Compare commits

..
Author SHA1 Message Date
sysadmin 7cc3460580 Merge branch 'prgs/master' into feat/issue-503-reviewer-active-worktree 2026-07-08 04:43:59 -04:00
sysadmin c54ac0d4de Merge pull request 'feat: isolate MCP workspace binding across role namespaces (Closes #510)' (#512) from feat/issue-510-workspace-binding-isolation into master 2026-07-08 03:37:17 -05:00
sysadmin 4d24c34548 Merge pull request 'feat: require post-merge cleanup safety-gate proof in reviewer reports (Closes #402)' (#415) from feat/issue-402-post-merge-cleanup-proof into master 2026-07-08 03:13:43 -05:00
sysadmin 3f3d6cb35d fix: resolve conflicts for PR #508
Merge prgs/master into feat/issue-503-reviewer-active-worktree.
Preserve reviewer mutation workspace binding (#503) and fold in
master's review_pr task preflight checks via _verify_reviewer_mutation_workspace.
2026-07-08 04:10:57 -04:00
sysadmin 3d11e1f12b feat: isolate MCP workspace binding across role namespaces (Closes #510)
Namespace-scoped workspace resolution prevents GITEA_AUTHOR_WORKTREE from
poisoning reviewer, merger, and reconciler purity checks. Each role uses
its own worktree env var with actionable binding-source errors and safe
reconnect guidance. Preserves #274/#475 author and control-checkout guards.
2026-07-08 04:08:08 -04:00
sysadmin ce61e424f6 Merge pull request 'fix: preserve capability preflight across safe reads (Closes #469)' (#502) from fix/issue-469-preflight-read-survival into master 2026-07-08 03:01:48 -05:00
sysadmin f89949cea9 Merge pull request 'fix: exempt reconciler close_pr from branches-only worktree guard (Closes #468)' (#472) from feat/issue-468-reconciler-close-guard into master 2026-07-08 02:48:57 -05:00
sysadmin b3edd53185 Merge pull request 'fix: guard PR/issue comment listing against non-list API payloads (Closes #485)' (#487) from feat/issue-485-lease-comments-non-list-guard into master 2026-07-08 02:46:03 -05:00
sysadminandClaude Opus 4.8 dad1dc8d51 fix: bind reviewer mutations to active branches worktree (Closes #503)
Reviewer mutation tools now resolve workspace from worktree_path,
GITEA_ACTIVE_WORKTREE, or the session reviewer lease before preflight
and branches-only validation. Preflight rejects metadata-only worktree_path
bindings that would mislead runtime_context while mutations still inspect
the MCP process root.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-08 03:42:07 -04:00
sysadmin 9ada4762c5 Merge pull request 'feat: expose explicit adoption proof in gitea_lock_issue response (Closes #477)' (#481) from feat/issue-477-lock-adoption-proof into master 2026-07-08 02:31:47 -05:00
sysadminandClaude Opus 4.8 3bce0a55fb test: reset preflight globals in read-survival setUp (#469)
Prevents cross-test leakage that masked the missing-capability fail-closed case.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-08 03:23:12 -04:00
sysadminandClaude Opus 4.8 afb4ba563c fix: preserve capability preflight across interleaved read-only whoami (#469)
Read-only gitea_whoami calls no longer clear a valid capability proof when
whoami was already verified clean. Capability is consumed on mutation (one
resolve per mutation), and verify_preflight_purity rejects task mismatches.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-08 03:23:12 -04:00
sysadmin 48cf3a334b Merge pull request 'feat: require validation cwd and HEAD proof in reviewer reports (Closes #398)' (#411) from feat/issue-398-validation-cwd-proof into master 2026-07-08 02:09:11 -05:00
sysadmin 974463163d Merge pull request 'feat: add first-class stacked PR support to author workflow (Closes #484)' (#489) from feat/issue-484-stacked-pr-support into master 2026-07-08 01:41:19 -05:00
sysadminandClaude Opus 4.8 0cbd1fc801 feat: add first-class stacked PR support to author workflow (#484)
Normal author work is unchanged: gitea_lock_issue still requires the worktree
to be base-equivalent to master/main/dev, and gitea_create_pr still targets a
normal base branch. A *stacked* PR (based on another unmerged PR's branch) is a
new explicit, proof-backed path — it never bypasses the issue lock.

- stacked_pr_support.py: pure policy. Approve a non-master base only when it is
  declared AND owned by a live OPEN PR whose number matches; reject arbitrary,
  mismatched, or stale (merged/closed) branches; require the PR body to document
  the stack (base branch, stacked-on PR, merge ordering).
- issue_lock_worktree.py: read_worktree_git_state / _find_matching_base_ref gain
  an opt-in extra_bases arg so an approved stacked base can anchor
  base-equivalence in addition to master/main/dev (empty by default = unchanged).
- gitea_mcp_server.py:
  - gitea_lock_issue gains stacked_base_branch + stacked_base_pr. When declared,
    it verifies the open PR owns the branch, anchors base-equivalence to it, and
    records approved_stacked_base = {branch, pr_number, verified_open} on the lock.
  - gitea_create_pr validates a non-master base against the lock's approved
    stacked base, re-checks the dependency PR is still open, and enforces the
    stacked-PR body fields. Master-based path untouched.
- Docs: llm-workflow-runbooks.md and work-issue.md document when stacked PRs are
  allowed and the required PR-body wording.

Tests: test_stacked_pr_support.py (18 policy cases), stacked base-equivalence in
test_issue_lock_worktree.py, approved_stacked_base persistence round-trip in
test_issue_lock_store.py. Existing lock/create_pr regressions still pass.

The #482 -> #479/#478 case motivates this: create_pr previously could not open a
stacked PR because the lock required a master-equivalent worktree.

Closes #484.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-08 02:22:54 -04:00
sysadmin d6b1e4be3c fix: resolve conflicts for PR #415 2026-07-08 02:20:04 -04:00
sysadmin 28f0f84de8 fix: resolve conflicts for PR #411 2026-07-08 02:20:04 -04:00
sysadminandClaude Opus 4.8 08202f7eaa fix: guard PR/issue comment listing against non-list API payloads (Closes #485)
_list_pr_lease_comments and gitea_list_issue_comments now fail safe to an
empty list when the comments API returns a non-list payload (e.g. an error
object), preventing KeyError: slice on the lease and listing paths.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-08 02:17:37 -04:00
sysadmin 7af73e1539 Merge pull request 'feat: add precise validation status vocabulary for reviewer reports (Closes #406)' (#412) from feat/issue-406-validation-status-vocabulary into master 2026-07-08 00:48:04 -05:00
sysadmin ae0478bf54 Merge pull request 'feat: add conflict-fix leases and stale-head protection (Closes #399)' (#416) from feat/issue-399-conflict-fix-leases into master 2026-07-08 00:34:20 -05:00
sysadmin cc5686b896 Merge pull request 'feat: require approval at current PR head for merge (#471)' (#480) from feat/issue-471-merge-head-approval into master 2026-07-08 00:29:17 -05:00
sysadmin d4dc9aa854 fix: align test mocks with #399 lease gates and head SHA requirements
PR #416 introduced pr_work_lease reviewer blocks and mandatory full
40-hex expected_head_sha on mark/merge paths. Update audit, review,
permission-report, terminal hard-stop, inventory, and schema tests to
patch lease comment fetches, seed ready review decisions, and use
consistent HEAD_SHA constants so the full suite passes again.
2026-07-08 01:26:36 -04:00
jcwalker3andClaude Opus 4.8 3496fc3952 feat: surface explicit adoption proof in gitea_lock_issue response (Closes #477)
When gitea_lock_issue adopts an existing own branch, the live tool
response now carries explicit, citable adoption-proof fields so #473-style
recovery sessions can quote the lock output directly instead of inferring
adoption from separate offline checks.

- issue_lock_adoption.py: add DECISION_LABELS + decision_label()/
  safe_next_action() helpers; extend build_adoption_proof() with
  adoption_decision, adopted, adopted_branch, adopted_branch_head,
  matcher_summary (boundary-safe reason), competing_branch_check, and
  safe_next_action for all outcomes; add build_non_adoption_lock_proof()
  for adoption-free NO_MATCH metadata.
- gitea_mcp_server.py: attach adoption-free adoption_check block to normal
  (non-adopt) lock responses so they cannot be misread as claiming adoption.
- docs/llm-workflow-runbooks.md: document the adoption/adoption_check
  response blocks and instruct recovery reports to cite them directly.
- tests: explicit-field proof for ADOPT/BLOCK_COMPETING/NO_MATCH,
  substring-collision boundary safety (issue-42 vs issue-420), non-adoption
  proof, and MCP-level live-response assertions.

BLOCK_COMPETING lock attempts still fail closed (raise); no raw API
fallback, branch deletion, force-push, or manual lock seeding introduced.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 23:40:12 -05:00
sysadminandClaude Opus 4.8 89582a0f2a feat: require approval at current PR head for merge (#471)
Add merge_approval_gate assessment so gitea_merge_pr fails closed when
visible APPROVED reviews do not pin the live head SHA. Expose
approval_at_current_head and stale_approval_block_reason in review
feedback. Document re-review requirement in canonical merge workflow.

Closes #471.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-08 00:37:54 -04:00
sysadmin a1de4ab8f9 Merge pull request 'feat: harden issue lock store against concurrent session clobbering (Closes #438)' (#464) from feat/issue-438-lock-hardening into master 2026-07-07 17:37:46 -05:00
sysadmin 3c41d2af5c Merge pull request 'feat: add own-branch lock adoption recovery to gitea_lock_issue (Closes #442)' (#461) from feat/issue-442-lock-adoption into master 2026-07-07 17:16:33 -05:00
sysadminandClaude Opus 4.8 d306c0a5ec feat: harden keyed issue lock store with flock serialization (#438)
Rebase onto master (#443 keyed store) and port #438 hardening: per-issue
fcntl.flock around bind_session_lock, assess_lock_freshness with dead-PID
detection, verify_lock_for_mutation before create_pr, and live lock
inventory for claim visibility.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 18:07:44 -04:00
sysadminandClaude Opus 4.8 85532059e2 fix: use exact issue-number boundary in own-branch adoption (#442)
Replace substring issue-marker matching with a numeric word-boundary
regex so issue-42 adoption is not false-blocked by unrelated issue-420
branches. Add regression tests for the #42 vs #420 collision.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 18:05:19 -04:00
sysadmin 54a75153a9 Merge pull request 'feat: keyed issue-lock store replaces global lock file (Closes #443)' (#465) from feat/issue-443-lock-store into master 2026-07-07 16:57:54 -05:00
sysadmin 5dfc00319f Merge pull request 'feat: align mutation guard with runtime_context workspace resolution (Closes #460)' (#462) from feat/issue-460-workspace-guard-alignment into master 2026-07-07 16:53:07 -05:00
sysadmin f51dc95153 Merge pull request 'feat: harden #274 branches-only guard for gitea_create_issue (Closes #274)' (#449) from feat/issue-274-branches-only-guard-hardening into master 2026-07-07 16:48:00 -05:00
sysadmin 55b3658d93 fix: resolve conflicts for PR #416
Merge current prgs/master into feat/issue-399-conflict-fix-leases.
Preserve reviewer PR lease gate (#407) before conflict-fix stale-head
gate (#399); document both as §26B and §26C. Align MCP and validator
tests with mandatory expected_head_sha and stale-head proof fields.
2026-07-07 17:38:45 -04:00
sysadminandClaude Opus 4.8 c7b462a034 fix: exempt reconciler close_pr from branches-only worktree guard (#468)
Reconciler profiles perform Gitea metadata mutations (close_pr) and must
not be blocked when the MCP workspace resolves to the stable control
checkout without GITEA_AUTHOR_WORKTREE. Author file mutations remain
guarded. Add regression tests for reconciler close vs author create_issue.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 17:36:33 -04:00
sysadminandClaude Opus 4.8 e956040b6b test: align duplicate-gate and artifact tests with keyed lock store (#443)
Update lock-issue and duplicate-recheck tests to mock branch listing,
use GITEA_ISSUE_LOCK_DIR session binding, and satisfy create_pr
provenance/worktree guards after the keyed persistent lock store landed.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 17:30:50 -04:00
sysadmin 43a17a7e8e resolve conflicts for PR #465 2026-07-07 17:30:35 -04:00
sysadminandClaude Opus 4.8 d042a9ca24 feat: replace global issue lock with keyed persistent store (Closes #443)
Store per remote/org/repo/issue locks under GITEA_ISSUE_LOCK_DIR with
atomic writes and per-session binding. Integrate own-branch adoption for
lock recovery, update worktree-start and cleanup reconcile, and add tests
documenting the ban on manual global lock seeding.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 17:30:35 -04:00
sysadmin f061fee18d Merge pull request 'feat: add per-PR reviewer leases for parallel review (Closes #407)' (#424) from feat/issue-407-reviewer-pr-leases into master 2026-07-07 16:27:19 -05:00
sysadminandClaude Opus 4.8 44a7c62dc3 fix: resolve conflicts for PR #415
Rebase onto current prgs/master and fix cleanup-mutations regex so
"Cleanup mutations: none" does not false-positive the proof checklist.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 17:19:37 -04:00
sysadminandClaude Opus 4.8 d5dc9f2708 feat: require post-merge cleanup safety-gate proof in reviewer reports (Closes #402)
Adds post_merge_cleanup_proof validation so cleanup claims must carry the
branch deletion and worktree removal checklists, or report CLEANUP_SKIPPED
with an exact blocker.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 17:19:37 -04:00
sysadminandClaude Opus 4.8 3e4b721d60 test: align duplicate gate lock fixtures with #447 provenance (#407)
Rebase onto master (#413/#463) requires sanctioned lock_provenance in
create_pr duplicate-recheck tests.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 17:18:00 -04:00
sysadminandClaude Opus 4.8 1830360850 test: fix reviewer lease gate regressions in MCP integration tests (#407)
Align merge/review/audit mocks with the per-PR reviewer lease gate introduced
in #407: install owned session leases after init_review_decision_lock, stub
_fetch_pr_comments and _authenticated_username to preserve mock ordering, and
update head-SHA mismatch expectations for lease-first fail-closed paths.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 17:16:06 -04:00
sysadminandClaude Opus 4.8 06c2069234 feat: add per-PR reviewer leases for parallel review (Closes #407)
Structured PR-thread leases with acquire/heartbeat MCP tools and
mutation-time enforcement so reviewers cannot race the same PR queue.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 17:16:06 -04:00
sysadmin cc4741a4ce Merge pull request 'feat: block manual issue-lock seeding and require lock disclosure (Closes #447)' (#463) from feat/issue-447-lock-provenance into master 2026-07-07 15:46:09 -05:00
sysadmin d95ea323d7 Merge pull request 'feat: move duplicate-work detection before author mutations (Closes #400)' (#413) from feat/issue-400-duplicate-work-preflight into master 2026-07-07 15:40:23 -05:00
sysadminandClaude Opus 4.8 e4adccd82a fix: inject duplicate-work context fetcher for testable lock/create_pr paths (#400)
Expose issue_duplicate_context_fetcher on the MCP server so lock_issue,
commit_files, and create_pr duplicate rechecks avoid live Gitea calls in
unit tests. Update affected test suites to patch the fetcher without
weakening duplicate-work gate assertions.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 16:33:31 -04:00
sysadmin 89a7d4dbfc Merge pull request 'fix(webui): suppress misleading empty queue copy on fail-closed fetch (#458)' (#459) from feat/issue-458-queue-fail-closed-ux into master 2026-07-07 15:19:13 -05:00
sysadminandClaude Opus 4.8 795f544047 feat: block manual issue-lock seeding and require lock disclosure (#447)
Add lock_provenance metadata on gitea_lock_issue writes and fail closed at
gitea_create_pr when provenance is missing. Final-report validation now
requires explicit External-state mutations disclosure for issue-lock
read/write/delete and blocks mixed author PR creation with reviewer approval.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 16:18:42 -04:00
sysadminandClaude Opus 4.8 b33844d74a feat: align mutation guard with runtime_context workspace resolution (Closes #460)
Share canonical workspace/repo-root resolution between gitea_get_runtime_context
and verify_preflight_purity so valid branches/ worktrees are not rejected when
the MCP process root differs from the stable control checkout. Fix relative
git-common-dir resolution and add regression tests.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 16:11:23 -04:00
sysadmin 22a1c4c2df fix(webui): suppress empty queue copy on fetch failure (#458)
When Gitea credentials are missing or the queue fetch fails closed,
show "Not loaded" instead of "No open items." and keep pagination
marked unavailable. Successful empty inventories still show the empty
state with complete pagination proof.

Closes #458
2026-07-07 15:53:10 -04:00
sysadmin ee8e9a0247 Merge pull request 'feat: add live PR/issue queue dashboard to web UI (Closes #429)' (#445) from feat/issue-429-queue-dashboard into master 2026-07-07 14:13:27 -05:00
sysadminandClaude Opus 4.8 f5370a94d3 test: harden queue dashboard classification and route coverage (#429)
Expand queue dashboard tests for stale/duplicate badges, title-linked
issues, and nav coverage; drop unused import in queue_loader.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 15:05:42 -04:00
sysadminandClaude Opus 4.8 c91e3642de feat: add live PR/issue queue dashboard to web UI (Closes #429)
Adds read-only /queue and /api/queue routes that load open PRs and issues
from the default registry project via gitea_auth, surface pagination proof,
and classify items with claimed/blocked/in-review/duplicate badges.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 15:05:41 -04:00
sysadmin 3599a9e12a Merge pull request 'feat: add canonical workflow prompt library to web UI (Closes #428)' (#441) from feat/issue-428-prompt-library into master 2026-07-07 13:58:57 -05:00
sysadmin f845864889 feat: add web UI prompt library from canonical workflows (#428)
Surface short copy/paste operator prompts derived from canonical workflow
files with workflow path and SHA-256 citations. Adds /prompts, /prompts/{id},
and /api/prompts with one-click copy.

Closes #428
2026-07-07 14:53:16 -04:00
sysadminandClaude Opus 4.8 6670c72e65 feat: add canonical workflow prompt library to web UI (Closes #428)
Expose short copy/paste operator prompts on /prompts derived from canonical
workflow files with workflow path and SHA-256 hash. Adds JSON export at
/api/prompts, copy buttons, and tests.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 14:53:16 -04:00
sysadmin 1529b9ff6b Merge pull request 'feat: thread worktree_path through gitea_create_issue (Closes #450)' (#451) from feat/issue-450-create-issue-worktree-guard into master 2026-07-07 13:51:00 -05:00
sysadmin 4a95c65f8b Merge pull request 'feat: add web UI project registry and onboarding (Closes #427)' (#439) from feat/issue-427-project-registry into master 2026-07-07 13:48:06 -05:00
sysadmin 16dcf65825 feat: add web UI project registry and onboarding (#427)
Load projects from versioned webui/data/projects.registry.json with profile
mappings, workflow/schema paths, and read-only onboarding checklist UI.
Seeds Gitea-Tools; exposes /projects, /projects/{id}, and /api/projects.

Closes #427
2026-07-07 14:44:27 -04:00
sysadmin 81fcdb09fd feat: thread worktree_path through gitea_create_issue (#450)
Hardens #274 branches-only guard for gitea_create_issue:

- verify_preflight_purity: validate resolved worktree exists, is a directory,
  and belongs to the target repository before author mutations.
- gitea_create_issue: add worktree_path threaded into preflight guard.
- tests/test_create_issue_workspace_guard.py: stable-control-checkout rejection
  and invalid-path fail-closed cases (PROJECT_ROOT simulated as control checkout).

Preserves WIP commit 8530109 implementation; test fix completes stable-checkout path.

Closes #450
2026-07-07 14:41:06 -04:00
sysadmin a9265efa82 feat: harden #274 branches-only guard for gitea_create_issue
Follow-up on the branches-only worktree guard (#274):

- verify_preflight_purity: validate resolved worktree exists, is a directory,
  and belongs to the target repository before author mutations.
- gitea_create_issue: add worktree_path for explicit branches/ workspace proof.
- tests/test_create_issue_workspace_guard.py: six regression cases including
  stable-control-checkout rejection with explicit PROJECT_ROOT simulation.

Closes #274
2026-07-07 14:38:10 -04:00
sysadmin 30ded9b712 Merge pull request 'feat: add internal web UI server skeleton (Closes #426)' (#437) from feat/issue-426-internal-web-ui-skeleton into master 2026-07-07 13:38:07 -05:00
sysadminandClaude Opus 4.8 a8fcf0e01c feat: add internal web UI server skeleton (Closes #426)
Starlette read-only MVP with shared layout, /health JSON liveness, and
route stubs for projects, prompts, runtime, audit, worktrees, and leases.
Includes scripts/run-webui, docs/webui-local-dev.md, and tests.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 13:23:09 -04:00
sysadminandClaude Opus 4.8 1320a55a7e feat: require validation cwd and HEAD proof in reviewer reports (Closes #398)
Rebased onto current prgs/master; merged with #396 validation failure
history by keeping both validator rules and workflow handoff fields.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 13:09:58 -04:00
sysadminandClaude Opus 4.8 0bad26230b fix: resolve conflicts for PR #413
Rebase onto current prgs/master and patch create_pr duplicate-gate test
with auth header mock to match post-rebase recheck path.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 13:08:53 -04:00
sysadminandClaude Opus 4.8 cf057d7829 feat: move duplicate-work detection before author mutations (Closes #400)
Extract issue_work_duplicate_gate for open PR, remote branch, and active
claim checks; wire it into lock_issue, commit_files recheck, and create_pr
recheck. Add gitea_assess_work_issue_duplicate read-only preflight, work-issue
report outcome validation, and workflow section 10A.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 13:08:21 -04:00
sysadmin 87beb44394 feat: add conflict-fix leases and stale-head protection (Closes #399)
Introduce structured PR work leases so author conflict-fix pushes cannot
race reviewer validation, approval, or merge on a moving head SHA.

- pr_work_lease.py: parse/acquire leases, push and reviewer mutation gates
- gitea_acquire_conflict_fix_lease, gitea_assess_conflict_fix_push MCP tools
- Enforce reviewed head SHA on mark_final_review_decision, submit_pr_review, merge_pr
- Final-report rules and workflow sections 20A / 26B
- tests/test_pr_work_lease.py (14 cases)
2026-07-07 13:06:07 -04:00
sysadminandClaude Opus 4.8 db032ef93c feat: add precise validation status vocabulary for reviewer reports (Closes #406)
Introduce validation_status_vocabulary with proof-path binding for baseline-
equivalent, merge-simulation-resolved, and transient-pass labels. Wire the
assessor into final_report_validator and document the taxonomy in the
review-merge workflow.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 12:53:55 -04:00
sysadmin 1a1e679246 Merge pull request 'feat: gate gitea_delete_branch on gitea.branch.delete capability (Closes #408)' (#410) from feat/issue-408-delete-branch-capability-gate into master 2026-07-07 10:48:51 -05:00
sysadmin 9fde4f3e76 Merge pull request 'feat: require transient validation failure history in reviewer reports (Closes #396)' (#409) from feat/issue-396-transient-validation-failure-history into master 2026-07-07 10:42:56 -05:00
sysadmin 6b5d2ad080 Merge pull request 'feat: require proof-backed claims in reviewer handoff reports (Closes #395)' (#397) from feat/issue-395-proof-backed-review-handoff into master 2026-07-07 10:34:35 -05:00
sysadmin f5654963eb Merge pull request 'feat: add prgs-reconciler profile model and role detection (Closes #304)' (#388) from feat/issue-304-reconciler-profile into master 2026-07-07 10:27:12 -05:00
sysadminandClaude Opus 4.8 af7131abf1 feat: gate gitea_delete_branch on gitea.branch.delete capability (Closes #408)
Add fail-closed profile gate at tool entry before preflight or API calls,
return structured permission reports on block, and record required_permission
in delete_branch audit metadata. Regression tests prove resolver-denied
sessions cannot bypass the gate through the raw tool.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 11:26:37 -04:00
sysadmin 027a3cb92c Merge pull request 'feat: register work-issue task routing for role-aware MCP (Closes #139)' (#385) from feat/issue-139-role-aware-work-issue-routing into master 2026-07-07 10:24:04 -05:00
sysadminandClaude Opus 4.8 71ccbca10f feat: require transient validation failure history in reviewer reports (Closes #396)
Add assess_validation_failure_history_report so reviewer final reports must
document every validation failure observed during a session, not only the
final passing result. Wire the verifier into final_report_validator,
gitea_validate_review_final_report, and the review-merge workflow template.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 11:22:00 -04:00
sysadmin 6220304f8c fix: resolve conflicts for PR #397
Merge prgs/master into feat/issue-395-proof-backed-review-handoff; keep
both proof-backed handoff (#395) and already-landed classification (#295)
downgrade gates in review_proofs.py.
2026-07-07 11:21:22 -04:00
sysadmin 6a37f7c8d5 fix: resolve conflicts for PR #388
Merge prgs/master into feat/issue-304-reconciler-profile; keep both
reconciler_profile and reconciliation_workflow imports in gitea_mcp_server.py.
2026-07-07 11:19:57 -04:00
sysadmin fcb9944378 Merge pull request 'feat: add reviewer final-report schema MCP validator (Closes #391)' (#394) from feat/issue-391-review-final-report-schema into master 2026-07-07 10:15:41 -05:00
sysadmin f75f2327f5 fix: resolve conflicts for PR #385
Merge prgs/master into feat/issue-139-role-aware-work-issue-routing and
keep both work-issue task routing entries and reconcile-landed-pr entries
in task_capability_map.py.
2026-07-07 11:13:45 -04:00
sysadmin 9f4dd4c39b Merge pull request 'feat: enforce canonical reconciliation handoff schema (Closes #307)' (#387) from feat/issue-307-canonical-reconciliation-handoff into master 2026-07-07 10:09:50 -05:00
sysadmin 6e212c0de7 Merge pull request 'feat: enforce author work leases in issue lock preflight (Closes #267)' (#386) from feat/issue-267-work-leases into master 2026-07-07 09:52:37 -05:00
sysadmin d446a31442 Merge pull request 'feat: add reconciliation workflow MCP tools for already-landed PRs (Closes #301)' (#384) from feat/issue-301-reconciliation-workflow into master 2026-07-07 09:42:31 -05:00
sysadmin b11160f236 Merge pull request 'feat: classify already-landed PRs as reconciliation-only (Closes #295)' (#382) from feat/issue-295-already-landed-classification into master 2026-07-07 09:37:23 -05:00
sysadmin d814a9ca81 fix: resolve conflicts for PR #388
Merge prgs/master; unify reconciler profile docs (#304) with master's
already-landed close tool guidance (#310).
2026-07-07 10:34:57 -04:00
sysadmin 10e64f6683 fix: resolve conflicts for PR #385
Merge prgs/master; keep work-issue author routing (#139) alongside
reconciler task entries from master (#309/#310).
2026-07-07 10:34:23 -04:00
sysadmin 5513bdf2aa fix: resolve conflicts for PR #384
Merge prgs/master; keep read-only reconciliation tools (#301) alongside
master's gitea_reconcile_already_landed_pr close path (#310). Unify task
routing: reconcile_landed_pr stays author/read-only; reconcile-landed-pr
and reconcile_already_landed_pr stay reconciler.
2026-07-07 10:33:42 -04:00
sysadmin 273eba7fb3 Merge pull request 'feat: enforce branches-only author mutation worktree guard (Closes #274)' (#374) from feat/issue-274-branches-only-worktrees into master 2026-07-07 09:33:18 -05:00
sysadmin cd61272837 fix: resolve conflicts for PR #374
Merge prgs/master into feat/issue-274-branches-only-worktrees; keep both
already_landed_reconcile and author_mutation_worktree imports. Allow
branches/ worktrees when PROJECT_ROOT is the session worktree; align
commit-payload tests with branches-only guard (#274).
2026-07-07 10:31:21 -04:00
sysadmin 7fbb1bf34a Merge pull request 'feat: require pagination proof in reconciliation PR inventory (Closes #308)' (#376) from feat/issue-308-reconcile-inventory-pagination into master 2026-07-07 09:26:02 -05:00
sysadmin c1d85b621a feat: require proof-backed claims in reviewer handoff reports (Closes #395) 2026-07-07 10:24:54 -04:00
sysadmin 4243b60ca3 Merge pull request 'feat: add prgs-reconciler profile and gated already-landed PR close (Closes #310)' (#381) from feat/issue-310-prgs-reconciler-profile into master 2026-07-07 09:22:55 -05:00
sysadmin f5953549aa fix: resolve conflicts for PR #376
Merge prgs/master; keep reconcile-inventory pagination verifier (#308)
alongside linked-issue live-proof gates from master (#300).
2026-07-07 10:22:35 -04:00
sysadmin ab6bb593bc fix: resolve conflicts for PR #382
Merge prgs/master; keep already-landed oldest-eligible-PR test from
#295 alongside master already-landed gate and target-branch freshness tests.
2026-07-07 10:20:52 -04:00
sysadmin 4cf75bcbc9 fix: resolve conflicts for PR #384
Merge prgs/master; keep reconciliation_workflow import alongside
review_merge_state_machine and preserve both reconciliation and
pr-queue-cleanup workflow split tests.
2026-07-07 10:20:14 -04:00
sysadmin 9ff861a1f3 fix: resolve conflicts for PR #388
Merge prgs/master into feat/issue-304-reconciler-profile; keep both
reconciler_profile and review_merge_state_machine imports.

Closes conflict with master landing review_merge_state_machine (#389 stack).
2026-07-07 10:19:40 -04:00
sysadmin 8d5fa06351 Merge commit 'b5ee6567aade4fdce4c57d826184bfd01ebb522e' into HEAD
# Conflicts:
#	tests/test_llm_workflow_split.py
2026-07-07 10:19:18 -04:00
sysadmin 7db8298cdd Merge pull request 'feat: require live linked issue proof in reconciliation handoffs (Closes #300)' (#372) from feat/issue-300-linked-issue-live-proof into master 2026-07-07 09:19:01 -05:00
sysadmin 167257b8a3 Merge commit 'b5ee6567aade4fdce4c57d826184bfd01ebb522e' into HEAD
# Conflicts:
#	tests/test_resolve_task_capability.py
2026-07-07 10:18:20 -04:00
sysadmin 23380d27fe Merge commit 'b5ee6567aade4fdce4c57d826184bfd01ebb522e' into HEAD
# Conflicts:
#	review_proofs.py
#	tests/test_llm_workflow_split.py
2026-07-07 10:16:59 -04:00
sysadmin bb525a8f94 Merge pull request 'Enforce already-landed review report state gates' (#383) from feat/issue-294-already-landed-review-gate into master 2026-07-07 09:15:19 -05:00
sysadmin d5b185ce38 Merge commit 'b5ee6567aade4fdce4c57d826184bfd01ebb522e' into HEAD
# Conflicts:
#	review_proofs.py
#	tests/test_llm_workflow_split.py
2026-07-07 10:15:10 -04:00
sysadmin b5ee6567aa Merge pull request 'feat: verify canonical workflow citation and version in review reports (Closes #296)' (#380) from feat/issue-296-workflow-version-proof into master 2026-07-07 09:11:13 -05:00
sysadmin 103364aa8d Merge pull request 'feat: add PR-only queue cleanup mode with per-PR dispatch gates (Closes #390)' (#393) from feat/issue-390-pr-queue-cleanup-mode into master 2026-07-07 09:07:51 -05:00
sysadminandClaude Opus 4.8 a2bd23da14 test: expand PR-queue-cleanup coverage to full acceptance matrix (#390)
Superset of the initial test slice: adds capability-map and router
routing cases (reviewer-role enforcement, author wrong_role_stop), the
full terminal-chain matrix (comment/skip stop, gates-failed stop, merge
completed/blocker stop, unknown-decision fail-closed), and report
verifier cases for two terminal mutations, missing next-suggested-PR,
branch mutations, and missing workflow citation.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 10:06:50 -04:00
sysadmin a164700ab9 feat: add reviewer final-report schema MCP validator (Closes #391)
Expose gitea_validate_review_final_report composing the composable
final-report validator with review-specific schema gates for legacy
fields, merge/issue claims, blocked handoff replay, and narrative drift.
2026-07-07 10:04:06 -04:00
sysadminandClaude Opus 4.8 91c67930ec feat: add PR-only queue cleanup mode with per-PR dispatch gates (Closes #390)
Adds canonical pr-queue-cleanup workflow, report verifier, reviewer-only task
routing, and runbook guidance so operators can drain open PRs one canonical
review at a time with fail-closed boundaries on batching and unauthorized merges.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 10:02:49 -04:00
sysadmin 0c18d0f54f Merge pull request 'feat: enforce PR review/merge workflow state machine gates (Closes #290)' (#378) from feat/issue-290-review-merge-state-machine into master 2026-07-07 09:02:42 -05:00
sysadmin ae566d5cd1 Merge pull request 'feat: reject stale already-landed controller handoff fields (Closes #299)' (#371) from feat/issue-299-already-landed-handoff into master 2026-07-07 09:00:07 -05:00
sysadmin d66d465e31 fix: resolve conflicts for PR #385
Merge prgs/master and keep work-issue author routing (#139) alongside
reconciler close task routing (#309).
2026-07-07 09:58:45 -04:00
sysadmin cf94112e96 fix: resolve conflicts for PR #384
Merge prgs/master and keep read-only reconciliation workflow tasks (#301)
alongside reconciler close capabilities (#309).
2026-07-07 09:58:45 -04:00
sysadmin 5a5bc8846f fix: resolve conflicts for PR #381
Merge prgs/master and keep already-landed reconcile close tool routing (#310)
alongside dedicated reconciler close tasks (#309).
2026-07-07 09:58:45 -04:00
sysadmin 7ef4c53cd8 Merge remote-tracking branch 'prgs/master' into feat/issue-304-reconciler-profile 2026-07-07 09:57:58 -04:00
sysadmin 60040e789b Merge remote-tracking branch 'prgs/master' into feat/issue-290-review-merge-state-machine 2026-07-07 09:57:58 -04:00
sysadmin 769f387267 Merge remote-tracking branch 'prgs/master' into feat/issue-308-reconcile-inventory-pagination 2026-07-07 09:57:54 -04:00
sysadmin 202be15764 Merge remote-tracking branch 'prgs/master' into feat/issue-300-linked-issue-live-proof 2026-07-07 09:57:54 -04:00
sysadmin fbfcc08640 Merge remote-tracking branch 'prgs/master' into feat/issue-299-already-landed-handoff 2026-07-07 09:57:54 -04:00
sysadmin 0bb6cae95a Merge pull request 'feat: dedicated reconciler capability for closing landed PRs (Closes #309)' (#379) from feat/issue-309-reconciler-close-capability into master 2026-07-07 08:57:04 -05:00
sysadmin 180490c640 fix: resolve conflicts for PR #384
Merge prgs/master and keep both reconciliation workflow (#301) and
native MCP preference (#270) imports in gitea_mcp_server.
2026-07-07 09:57:01 -04:00
sysadmin b604b468b5 fix: resolve conflicts for PR #388
Merge prgs/master and keep reconciler profile runtime fields (#304)
alongside native MCP shell health (#270).
2026-07-07 09:57:01 -04:00
sysadmin af064dd06b fix: resolve conflicts for PR #378
Merge prgs/master and keep both review-merge state machine (#290) and
native MCP preference (#270) imports, skill entries, and MCP tools.
2026-07-07 09:57:01 -04:00
sysadmin 6100187274 fix: resolve conflicts for PR #376
Merge prgs/master and keep both reconcile-inventory (#308) and
inventory-worktree (#293) verifiers wired in build_final_report.
2026-07-07 09:55:33 -04:00
sysadmin bf002aee2c fix: resolve conflicts for PR #372
Merge prgs/master and keep both reconcile-linked-issue (#300) and
inventory-worktree (#293) verifiers wired in build_final_report.
2026-07-07 09:55:33 -04:00
sysadmin b53e3de7cc fix: resolve conflicts for PR #371
Merge prgs/master and keep both already-landed-handoff (#299) and
inventory-worktree (#293) verifiers wired in build_final_report.
2026-07-07 09:55:33 -04:00
sysadmin 47b4ee4791 Merge pull request 'feat: add native MCP preference gate and shell circuit breaker (Closes #270)' (#377) from feat/issue-270-native-mcp-preference into master 2026-07-07 08:54:03 -05:00
sysadmin 1c506fa779 Merge pull request 'feat: comment-then-stop policy for partial PR reconciliation (Closes #302)' (#375) from feat/issue-302-partial-reconciliation-policy into master 2026-07-07 08:51:12 -05:00
sysadmin e7a1e183a0 Merge pull request 'feat: prove inventory pagination and worktree state in reviewer reports (Closes #293)' (#369) from feat/issue-293-inventory-worktree-proof into master 2026-07-07 08:47:05 -05:00
sysadminandClaude Opus 4.8 72f3bc8db8 fix: resolve conflicts for PR #376 against current master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:45:21 -04:00
sysadminandClaude Opus 4.8 e0a1717db2 fix: resolve conflicts for PR #372 against current master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:45:03 -04:00
sysadminandClaude Opus 4.8 3c2dd055fa fix: resolve conflicts for PR #371 against current master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:44:44 -04:00
sysadminandClaude Opus 4.8 56f8f9eeb4 fix: resolve conflicts for PR #369 against current master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:44:12 -04:00
sysadmin b6cf32c845 Merge pull request 'feat: add dedicated reconciliation controller-handoff schema (Closes #303)' (#373) from feat/issue-303-reconciliation-handoff into master 2026-07-07 08:42:40 -05:00
sysadminandClaude Opus 4.8 21ff12cef7 feat: add prgs-reconciler profile model and role detection (Closes #304)
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:39:52 -04:00
sysadmin a17e8561d0 Merge pull request 'feat: require infra_stop repair handoff for blocked reviewers (Closes #289)' (#366) from feat/issue-289-infra-stop-handoff into master 2026-07-07 08:39:35 -05:00
sysadmin e2771dfe9a fix: resolve conflicts for PR #385 against latest master 2026-07-07 09:38:36 -04:00
sysadmin 7635a17599 fix: resolve conflicts for PR #384 against latest master 2026-07-07 09:38:34 -04:00
sysadmin f9bf830067 fix: resolve conflicts for PR #381 against latest master 2026-07-07 09:38:31 -04:00
sysadmin 85c5d24b0e fix: resolve conflicts for PR #379 against latest master 2026-07-07 09:38:28 -04:00
sysadmin dc51e9a91b fix: resolve conflicts for PR #374 against latest master 2026-07-07 09:38:25 -04:00
sysadmin f749312b45 fix: resolve conflicts for PR #376 against latest master 2026-07-07 09:37:24 -04:00
sysadmin fa922cf8e3 fix: resolve conflicts for PR #373 against latest master 2026-07-07 09:37:21 -04:00
sysadmin 208dc40041 fix: resolve conflicts for PR #372 against latest master 2026-07-07 09:37:18 -04:00
sysadmin 8b555043e3 fix: resolve conflicts for PR #371 against latest master 2026-07-07 09:37:16 -04:00
sysadmin faf4484657 fix: resolve conflicts for PR #369 against latest master 2026-07-07 09:37:13 -04:00
sysadmin 93230457d8 fix: resolve conflicts for PR #366 against latest master 2026-07-07 09:37:10 -04:00
sysadmin b4c79fdbbc Merge pull request 'feat: add issue claim heartbeat leases and stale-claim reconciliation (Closes #268)' (#370) from feat/issue-268-heartbeat-leases into master 2026-07-07 08:36:55 -05:00
sysadmin 8b9a0705b6 Merge pull request 'feat: enforce precise mutation categories in controller handoffs (Closes #319)' (#365) from feat/issue-319-mutation-categories into master 2026-07-07 08:33:56 -05:00
sysadminandClaude Opus 4.8 bc8e2b7928 fix: resolve conflicts for PR #382 against latest master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:32:37 -04:00
sysadminandClaude Opus 4.8 9bb9bad4c7 fix: resolve conflicts for PR #382 against latest master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:31:55 -04:00
sysadminandClaude Opus 4.8 ab126da479 fix: resolve conflicts for PR #376 against latest master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:31:52 -04:00
sysadminandClaude Opus 4.8 ba4cf2e93d fix: resolve conflicts for PR #375 against latest master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:31:50 -04:00
sysadminandClaude Opus 4.8 4507457560 fix: resolve conflicts for PR #372 against latest master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:31:47 -04:00
sysadminandClaude Opus 4.8 dc24fe3afe feat: enforce canonical reconciliation handoff schema (Closes #307)
Reconciliation final reports must emit only the canonical reconciliation
handoff schema:

- Expand _RECONCILE_REQUIRED_FIELDS from 5 to the full 27-field canonical
  schema (Task through No review/merge confirmation), so incomplete
  reconciliation handoffs downgrade with per-field reasons.
- Add 'issue lock proof', 'claim/comment status', and 'workspace mutations'
  to _RECONCILE_STALE_FIELDS so stale author/reviewer fields block.
- Make the 'PR number opened' rejection conditional on a new
  session_pr_opened proof kwarg: the field is allowed only when a PR was
  actually opened in the session.
- Close the git-fetch reporting gap: a fetch observed only in the action
  log (not report text) now requires a Git ref mutations entry as well.
- Update the reconciliation handoff fixture to the canonical schema and
  add TestCanonicalReconcileSchema covering blocked, successful-close,
  and comment-only reconciliation plus missing-close-capability,
  stale-field, and conditional PR-number-opened paths.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:31:45 -04:00
sysadminandClaude Opus 4.8 3c9841760c fix: resolve conflicts for PR #371 against latest master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:31:44 -04:00
sysadminandClaude Opus 4.8 51d0c14009 fix: resolve conflicts for PR #369 against latest master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:31:42 -04:00
sysadminandClaude Opus 4.8 8a6cb6ac2c fix: resolve conflicts for PR #366 against latest master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:31:21 -04:00
sysadmin ce39fb52c9 Merge pull request 'Add validation integrity verifier for PR-head vs diagnostic tests (#316)' (#358) from feat/issue-316-validation-integrity into master 2026-07-07 08:31:02 -05:00
sysadminandClaude Opus 4.8 c260ef15fb feat: register work-issue task routing for role-aware MCP (#139)
- Map work_issue/work-issue to author role and gitea.pr.create in resolver
- Route work-issue sessions through role_session_router before mutations
- Expose work_issue in runtime context task capabilities
- Add work-issue workflow source verifier and canonical routing docs
- Tests for resolver, router, and final-report verifier

Closes #139

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:30:38 -04:00
sysadmin e823000301 fix: resolve conflicts for PR #365 against latest master 2026-07-07 09:28:23 -04:00
sysadmin 7b71113b4d Merge pull request 'feat: hard-stop review mutations on already-landed PR heads (Closes #292)' (#368) from feat/issue-292-already-landed-gate into master 2026-07-07 08:28:04 -05:00
sysadminandClaude Opus 4.8 c2226177b6 fix: resolve conflicts for PR #358 against latest master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:27:29 -04:00
sysadmin 749e480baf feat: add reconciliation workflow MCP tools for already-landed PRs (Closes #301)
Expose read-only assess/scan tools and capability planning so already-landed
open PRs can be reconciled without review or merge. Register the
gitea-reconcile-landed-pr skill and workflow-source verification.
2026-07-07 09:25:55 -04:00
sysadmin b41d9d6006 Merge pull request 'feat: reject eligible/reviewed wording for already-landed PRs (Closes #298)' (#367) from feat/issue-298-already-landed-wording into master 2026-07-07 08:25:44 -05:00
sysadmin 2577489c2a feat: enforce already-landed review report states 2026-07-07 09:24:30 -04:00
sysadminandClaude Opus 4.8 f89019f804 feat: classify already-landed PRs as reconciliation-only (Closes #295)
Add reviewer_already_landed_classification verifier to block oldest/next
eligible PR wording, review-eligible claims, and pinned reviewed head
usage when a PR is already landed. Wire the check into build_final_report
and tighten final_report_validator regex coverage.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:23:50 -04:00
sysadmin e26398b7a9 Merge pull request 'feat: block edits in PR validation worktrees (Closes #315)' (#363) from feat/issue-315-validation-worktree-no-edits into master 2026-07-07 08:22:57 -05:00
sysadmin 9cf1b41b77 feat: add prgs-reconciler profile and gated already-landed PR close (Closes #310)
Introduce a dedicated reconciler role with gitea_reconcile_already_landed_pr,
which closes open PRs only after live fetch and ancestor proof against a fresh
target branch. Document the gitea-reconciler namespace and extend task routing.
2026-07-07 09:20:50 -04:00
sysadmin 0fe8f57dda Merge pull request 'feat: classify merge simulation as worktree/index mutations (Closes #317)' (#356) from feat/issue-317-merge-simulation-mutations into master 2026-07-07 08:20:40 -05:00
sysadminandClaude Opus 4.8 6946534f3e feat: verify canonical workflow citation and version in review reports (Closes #296)
The canonical PR review/merge workflow already lives in
skills/llm-project-workflow/workflows/review-merge-pr.md and is exposed
through mcp_list_project_skills / mcp_get_skill_guide (#333-#335), and
its final-report schema already demands a workflow version. What was
missing is enforcement: add compute_workflow_hash (short sha256
version hash of workflow text) and assess_review_workflow_source to
review_proofs. Review reports must cite the canonical workflow file
and carry a populated Workflow version field; when the current
canonical hash is supplied, a mismatched citation fails as a stale
workflow copy, directing the agent to reload via mcp_get_skill_guide.
Fails closed. Tests cover hash determinism and content sensitivity,
matching-hash pass, missing citation, missing version field, stale
hash, version none, and offline validation without a canonical hash.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:19:33 -04:00
sysadmin 1487ff60b4 Merge pull request 'feat: enforce validation environment proof in reviewer reports (Closes #311)' (#352) from feat/issue-311-validation-environment-proof into master 2026-07-07 08:18:29 -05:00
sysadmin d0f52cbf19 fix: resolve conflicts for PR #376 against current master 2026-07-07 09:17:33 -04:00
sysadmin 17c51eaa86 fix: resolve conflicts for PR #372 against current master 2026-07-07 09:17:33 -04:00
sysadmin 2900add223 fix: resolve conflicts for PR #371 against current master 2026-07-07 09:17:33 -04:00
sysadmin a5fb7d2472 fix: resolve conflicts for PR #366 against current master 2026-07-07 09:17:25 -04:00
sysadmin c5ab43ed5a fix: resolve conflicts for PR #366 against current master 2026-07-07 09:17:25 -04:00
sysadmin a237b15559 fix: resolve conflicts for PR #365 against current master 2026-07-07 09:17:15 -04:00
sysadmin 07b23950ab fix: resolve conflicts for PR #363 against current master 2026-07-07 09:17:00 -04:00
sysadmin bbf5ee6d59 fix: resolve conflicts for PR #375 against current master 2026-07-07 09:15:26 -04:00
sysadmin 5131fc595a fix: resolve conflicts for PR #368 against current master 2026-07-07 09:15:26 -04:00
sysadmin e017d80256 Merge pull request 'fix: seed profile gate for TestIssueLockArtifactWarning (Closes #359)' (#364) from fix/issue-359-artifact-warning-test into master 2026-07-07 08:15:18 -05:00
sysadminandClaude Opus 4.8 958d470f98 feat: dedicated reconciler capability for closing landed PRs (Closes #309)
Already-landed open PRs block the review queue: the already-landed gate
correctly stops approval/merge, but no task path could close the
redundant PR. Add the reconciler close path:

- task_capability_map: new reconcile_close_landed_pr (gitea.pr.close)
  and reconcile_close_landed_issue (gitea.issue.close) tasks under a
  dedicated "reconciler" role. Exact close capabilities only — the
  role grants no review, approve, request-changes, or merge authority,
  and normal reviewer profiles keep no broad PR-close authority.
- role_session_router: reconciler tasks route wrong_role_stop in
  author/reviewer sessions with a dedicated recovery message; matching
  reconciler sessions route allowed_current_session.
- review_proofs.assess_reconciler_close_gate: PR close allowed only
  when every proof passes — PR live-verified open, candidate head SHA
  pinned and matching the live head, target branch freshly fetched
  with a full SHA, head an ancestor of the target, exact
  gitea.pr.close proven. Non-landed PRs return
  NOT_LANDED_CLOSE_BLOCKED; missing close capability returns
  RECOVERY_HANDOFF_REQUIRED; incomplete proof fails closed as
  GATE_NOT_PROVEN. Linked-issue close is skipped when the issue is
  already closed and allowed only for an open issue resolved by the
  landed commits with exact gitea.issue.close capability. The gate
  also returns the required final-report field list.

17 new tests cover the capability map, wrong-role routing, and every
gate path (landed close, non-landed block, stale target, changed head,
missing capabilities, linked-issue variants, review-mutation denial).

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:15:13 -04:00
sysadmin af806919b3 fix: resolve conflicts for PR #358 against current master 2026-07-07 09:14:47 -04:00
sysadmin 42cd0f9f4a fix: resolve conflicts for PR #356 against current master 2026-07-07 09:14:30 -04:00
sysadmin 8d32af14d6 feat: enforce PR review/merge workflow state machine gates (Closes #290)
Add fail-closed review/merge state progression helpers, pre-merge gate
requirements, blocked recovery handoff checks, and MCP assessment tool.

Closes #290
2026-07-07 09:14:29 -04:00
sysadmin 764e636c07 fix: resolve conflicts for PR #352 against current master 2026-07-07 09:14:04 -04:00
sysadmin c21e756a30 Merge pull request 'feat: prove reviewer worktree ownership before reset (Closes #312)' (#362) from feat/issue-312-worktree-ownership into master 2026-07-07 08:12:56 -05:00
sysadmin 6743cc11b9 feat: add native MCP preference gate and shell circuit breaker (Closes #270)
Enforce native MCP-first paths for Gitea mutations with fail-closed
assessment helpers, shell spawn circuit breaker tracking, and MCP tools
for operation-path assessment and shell health reporting.

Closes #270
2026-07-07 09:11:15 -04:00
sysadmin c20a93602d Merge pull request 'feat: block approval on full-suite failure without baseline proof (Closes #323)' (#361) from feat/issue-323-full-suite-approval-gate into master 2026-07-07 08:09:13 -05:00
sysadminandClaude Opus 4.8 6e774e191a feat: require pagination proof in reconciliation PR inventory (#308)
Add assess_reconcile_inventory_report verifier to block complete queue scan
and all-already-landed claims without final-page metadata, requested page size,
pages fetched, per-page counts, and no-next-page proof. Wire into
build_final_report and export from review_proofs.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 09:08:38 -04:00
sysadminandClaude Opus 4.8 852ac1f56b feat: comment-then-stop policy for partial PR reconciliation (Closes #302)
Already-landed PR reconciliation could prove a PR should be closed but
stop silently when gitea.pr.close was unavailable, even with comment
capability present. Adopt Option B (comment-then-stop) as the durable
policy and enforce it:

- resolve_partial_reconciliation_plan maps proven capabilities to one
  of four outcomes: FULL_RECONCILE_CLOSE_ALLOWED (close_pr proven),
  PARTIAL_RECONCILE_COMMENT_THEN_STOP (comment_pr proven, close_pr
  missing; one reconciliation comment with PR head SHA, target branch
  SHA, ancestor proof, linked issue status, and the missing capability,
  then stop), RECOVERY_HANDOFF_ONLY (no comment capability; no Gitea
  mutation), and GATE_NOT_PROVEN (no ancestry proof; no mutation
  regardless of capability). Missing capability dicts fail closed.
- assess_partial_reconciliation_report requires final reports to
  explain why the comment was or was not posted, name the missing
  capability, and state the close result or that no mutation ran.
- workflows/reconcile-landed-pr.md gains section 12A documenting the
  policy; the workflow contract test locks the new markers.

Tests cover close capability present, close missing with comment
allowed, comment capability missing, all capabilities missing,
unproven ancestry, fail-closed capability dict, and report checks for
each outcome.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:47:16 -04:00
sysadminandClaude Opus 4.8 012b8b387a feat: add dedicated reconciliation controller-handoff schema (Closes #303)
Add assess_reconciliation_handoff to review_proofs: already-landed PR
reconciliation runs must emit a dedicated handoff schema (selected PR,
live state, candidate head SHA, target branch + SHA, ancestor proof,
linked issue + live status, eligibility class
ALREADY_LANDED_RECONCILE_REQUIRED, capability proofs, per-category
mutation ledger, blocker, safe next action, no-review/merge
confirmation) instead of author or reviewer fields. Stale fields (PR
number opened, Pinned reviewed head, Scratch worktree used, Workspace
mutations, Issue lock proof, Claim/comment status) fail validation, a
wrong eligibility class fails, and an observed git fetch/pull must be
reported under Git ref mutations rather than claimed none. Fails
closed on a missing Controller Handoff section. Tests cover blocked
reconciliation, successful PR close, comment-only, no-op
already-closed, missing linked-issue proof, wrong eligibility class,
each stale field, and unreported observed fetch.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:45:51 -04:00
sysadmin 5e023dcc97 Merge pull request 'feat: verify linked-issue consistency for the selected PR (Closes #314)' (#360) from feat/issue-314-linked-issue-consistency into master 2026-07-07 04:45:10 -05:00
sysadminandClaude Opus 4.8 6779d903f2 feat: enforce branches-only author mutation worktree guard (Closes #274)
Add author_mutation_worktree assessment and integrate it into MCP
preflight so author mutations fail closed outside branches/ session
worktrees, with tests for control-checkout and branches-path cases.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:44:58 -04:00
sysadminandClaude Opus 4.8 3764ba0566 feat: require live linked issue proof in reconciliation handoffs (#300)
Add assess_reconcile_linked_issue_report verifier to block open/closed/resolved
linked issue status claims without gitea_view_issue proof in the current session,
require not-verified wording when the issue is missing, and gate recommended
issue close/reopen/comment actions on exact capability proof. Wire into
build_final_report and export from review_proofs.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:44:18 -04:00
sysadminandClaude Opus 4.8 2c65cf05ca feat: reject stale already-landed controller handoff fields (#299)
Add assess_already_landed_handoff_report verifier to block Pinned reviewed
head, Scratch worktree used, legacy Mutations/Workspace mutations None when
git fetch or other mutations occurred, and narrative/handoff drift after the
already-landed gate fires. Wire into build_final_report and review_proofs.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:42:08 -04:00
sysadmin f87101ccaa Merge pull request 'feat: require live blocker proof before skipping earlier PRs (Closes #318)' (#355) from feat/issue-318-prior-blocker-skip-proof into master 2026-07-07 04:41:56 -05:00
sysadminandClaude Opus 4.8 275a362a46 feat: add issue claim heartbeat leases and stale-claim reconciliation (Closes #268)
Structured claim/progress heartbeats post on issue claim and via
gitea_post_heartbeat. Read-only gitea_reconcile_issue_claims inventories
active, stale, phantom, and PR-backed claims; gitea_cleanup_stale_claims
supports dry-run cleanup of reclaimable labels.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:41:33 -04:00
sysadminandClaude Opus 4.8 1d3ee72274 feat: prove inventory pagination and worktree state in reviewer reports (#293)
Add assess_inventory_worktree_report verifier to reject partial PR inventory
claims, exactly-full first pages without finality proof, legacy scratch-worktree
flags that contradict branches/ review worktrees, and contradictory checkout
wording. Wire into build_final_report and export from review_proofs.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:39:43 -04:00
sysadminandClaude Opus 4.8 be1462dc03 feat: reject eligible/reviewed wording for already-landed PRs (Closes #298)
Add assess_already_landed_report_wording to review_proofs: when the
already-landed gate fires, final reports and controller handoffs must
use the reconciliation wording (Oldest open PR requiring action,
Eligibility class: ALREADY_LANDED_RECONCILE_REQUIRED, Candidate head
SHA, Reviewed head SHA: none, Review worktree used: false) and must not
use the legacy fields (oldest/next eligible PR, Pinned reviewed head,
Scratch worktree used) or claim a reviewed head SHA. In
non-already-landed reports, a populated Pinned reviewed head or
Reviewed head SHA fails unless validation and diff review actually
passed this session. Fails closed. Tests cover the correct
reconciliation wording, each forbidden legacy field, missing required
fields, populated reviewed SHA, normal reviewed reports, blocked-infra
reports, and validation-failed reports.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:38:32 -04:00
sysadmin fa0cb12464 Merge pull request 'feat: block reviewer local Gitea fallbacks during normal review (Closes #324)' (#350) from feat/issue-324-reviewer-no-fallback into master 2026-07-07 04:37:12 -05:00
sysadmin 37dc3a38b8 fix: resolve conflicts for PR #352 2026-07-07 05:36:52 -04:00
sysadminandClaude Opus 4.8 876f1ee3a3 feat: hard-stop review mutations on already-landed PR heads (Closes #292)
A reviewer workflow approved and attempted to merge PR #278 although its
head commit was already an ancestor of master, ending in a Gitea 405 and
manual reconciliation. Add the missing pre-mutation gate:

- assess_already_landed_review_gate classifies the pinned candidate
  head against the fetched target branch: NORMAL_REVIEW_CANDIDATE,
  ALREADY_LANDED_RECONCILE_REQUIRED, or GATE_NOT_PROVEN (fail closed).
  Already-landed PRs block approval and the merge API, allow
  request-changes only for a real blocker, and require a reconciliation
  handoff (PR number/title, candidate head SHA, target branch + SHA,
  ancestor proof, linked issue status, recommended action, capability
  proof for close/comment mutations). Unchecked ancestry, invalid or
  missing SHAs, and a head changed since pinning all fail closed.
  Mergeability stays a separate gate.

- assess_already_landed_report_state rejects APPROVED / MERGED /
  READY_TO_MERGE wording and missing ALREADY_LANDED_RECONCILE_REQUIRED
  state when the session's gate fired, complementing the text-marker
  rules from #327 which need the report to admit the landed state.

Tests cover already-landed, normal, non-mergeable, changed-head,
unchecked-ancestry, and invalid-SHA gate paths plus all report-state
verdicts.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:36:10 -04:00
sysadmin 0cc3ed8868 feat: enforce author work leases (Closes #267) 2026-07-07 05:35:44 -04:00
sysadmin 15c9add3f4 fix: resolve conflicts for PR #358 2026-07-07 05:35:13 -04:00
sysadmin c278e16053 fix: resolve conflicts for PR #356 2026-07-07 05:35:13 -04:00
sysadmin f1c7054871 fix: resolve conflicts for PR #360 2026-07-07 05:35:13 -04:00
sysadmin 4204503b41 fix: resolve conflicts for PR #355 2026-07-07 05:35:13 -04:00
sysadmin 77b29fe417 fix: resolve conflicts for PR #362 2026-07-07 05:33:47 -04:00
sysadmin a0d68ef66e fix: resolve conflicts for PR #361 2026-07-07 05:33:47 -04:00
sysadmin f2d82a93ed fix: resolve conflicts for PR #350 2026-07-07 05:33:47 -04:00
sysadmin e1dabd1b0f Merge pull request 'feat: reject legacy Workspace mutations field in reviewer handoffs (Closes #320)' (#357) from feat/issue-320-remove-workspace-mutations into master 2026-07-07 04:33:31 -05:00
sysadmin 7c7aa1ac2d fix: resolve conflicts for PR #363 2026-07-07 05:33:14 -04:00
sysadminandClaude Opus 4.8 6be8f16351 Add infra-stop repair handoff verifier for blocked reviewers (#289)
Stop PR queue advancement when infra_stop blocks capability and require
CONTROL-CHECKOUT REPAIR MODE handoffs with infra diagnostics.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:33:12 -04:00
sysadminandClaude Opus 4.8 b71d3aeab3 Add precise mutation category verifier for controller handoffs (#319)
Reject legacy Workspace mutations wording and require explicit file,
worktree/index, and git ref mutation fields in reviewer controller handoffs.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:31:24 -04:00
sysadmin 4d3e90e7c5 Merge pull request 'feat: enforce queue ordering proof in reviewer reports (Closes #321)' (#351) from feat/issue-321-queue-ordering-proof into master 2026-07-07 04:31:14 -05:00
sysadmin 634d8a17aa Merge pull request 'feat: composable final-report validator for reviewer and reconciliation handoffs (Closes #327)' (#348) from feat/issue-327-final-report-validator-rules into master 2026-07-07 04:29:04 -05:00
sysadminandClaude Opus 4.8 ef7c15f84d fix: seed profile gate for TestIssueLockArtifactWarning (Closes #359)
gitea_lock_issue is profile-gated via task_capability_map
(gitea.issue.comment); without GITEA_ALLOWED_OPERATIONS the call
fail-closes before the artifact-warning logic runs, so the test
asserted success on a permission block. Seed the environment in
setUp/tearDown exactly like tests/test_mcp_server.py::TestIssueLocking
and complete the mocked git state with base_equivalent,
inspected_git_root, and base_branch so the newer lock-worktree
base-equivalence gate passes. The test still verifies the artifact
warning fires for '?? _emit_payload.py'. Full suite now passes with
zero failures.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:28:41 -04:00
sysadminandClaude Opus 4.8 b842a7f923 Add validation worktree no-edit verifier for reviewer reports (#315)
Block file edits in PR validation worktrees, require separate diagnostic
scratch worktrees with explicit labeling, and reject contradictory
File edits by reviewer: none claims.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:28:19 -04:00
sysadmin bc227a9a6d feat: block reviewer local Gitea fallbacks during normal review (Closes #324) 2026-07-07 05:27:24 -04:00
sysadmin 2cf0ad5908 feat: add composable final-report validator for reviewer and reconciliation handoffs (Closes #327) 2026-07-07 05:27:24 -04:00
sysadmin 077cedc0c9 feat: enforce queue ordering proof in reviewer reports (Closes #321) 2026-07-07 05:27:18 -04:00
sysadmin c83c18e52a Merge pull request 'feat: enforce proof wording for pagination and live claims (Closes #330)' (#346) from feat/issue-330-proof-wording-enforcement into master 2026-07-07 04:26:44 -05:00
sysadminandClaude Opus 4.8 ce3bd784c0 Add worktree ownership verifier for reviewer reset safety (#312)
Require session-owned review worktrees or documented safe-reuse proof
before destructive reset/validation. Block contradictory workspace-none
claims when git reset --hard occurred.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:26:37 -04:00
sysadminandClaude Opus 4.8 5fa5760dd0 feat: block approval on full-suite failure without baseline proof (Closes #323)
Approving a PR whose full test suite fails is now gated behind complete
baseline proof. New assess_full_suite_failure_approval_gate:

- Full-suite pass: approval allowed, no baseline required.
- Full-suite failure: default action is REQUEST_CHANGES; approval is
  blocked unless the #325 baseline validation proof passes (branches/
  worktree, pinned baseline/PR SHAs, matching failure signatures,
  clean before/after) and the proof also carries the PR and baseline
  full-suite commands, evidence that new/changed tests passed, and a
  non-empty explanation of why the failures are unrelated to the PR.
- Unknown/unproven full-suite result blocks approval.
- Main-checkout baseline comparison is rejected.
- Vague "same as master" claims without proof are rejected via the
  existing pre-existing-failure claim detection.
- Non-approve decisions are never blocked; the gate only reports the
  request-changes default.

Tests cover matching baseline failures, mismatched signatures,
incomplete proof (missing commands, new-test evidence, explanation),
main-checkout baseline, unknown suite result, vague claims, and
full-suite pass.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:24:07 -04:00
sysadmin 1666e55e60 Merge pull request 'feat: reject workspace-none claims after worktree mutations (Closes #313)' (#354) from feat/issue-313-worktree-mutation-consistency into master 2026-07-07 04:23:54 -05:00
sysadminandClaude Opus 4.8 1e6bd34d6f feat: verify linked-issue consistency for the selected PR (Closes #314)
Add assess_linked_issue_consistency to review_proofs: reviewer final
reports must report a "Linked issue status" that matches the issue(s)
the selected PR was live-verified to link this session. Without live
proof the field may only say "not verified in this session"; with
proof, every linked issue must be named, and any other "Issue #N"
mention anywhere in the report fails as stale leakage from a previous
PR (the PR #280 / Issue #260-vs-#275 incident). Missing field fails
closed. Tests cover correct linked issue, stale previous issue in both
the status field and diagnostics, missing proof, unverified wording,
multiple linked issues, and PR-number mentions not being mistaken for
issue mentions.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:22:08 -04:00
sysadmin e9fdb356dd fix: resolve conflicts for issue 330 (master refresh) 2026-07-07 05:21:43 -04:00
sysadmin 897d256f98 Merge pull request 'feat: add queue-status report verifier for reviewer handoffs (Closes #339)' (#347) from feat/issue-339-queue-status-report-verifier into master 2026-07-07 04:20:09 -05:00
sysadminandClaude Opus 4.8 4ea618e8b4 Add validation integrity verifier for PR-head vs diagnostic tests (#316)
Separate official PR-head validation from post-edit diagnostic runs in
reviewer reports. Block treating diagnostic passes as official validation
and require contaminated sessions to report recovery-required status.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:19:22 -04:00
sysadmin 240e7adf46 fix: resolve conflicts for issue 339 2026-07-07 05:18:29 -04:00
sysadmin deb228d629 Merge pull request 'feat: require conflict proof for skipped non-mergeable PRs (Closes #322)' (#353) from feat/issue-322-mergeability-skip-proof into master 2026-07-07 04:17:09 -05:00
sysadminandClaude Opus 4.8 94b022b24d fix: resolve conflicts for issue 330
Merge prgs/master into feat/issue-330-proof-wording-enforcement and keep
both assess_proof_wording (#330) and git ref mutation reporting (#297).

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:16:57 -04:00
sysadminandClaude Opus 4.8 71b8d00897 feat: classify merge simulation as worktree/index mutations (Closes #317)
Add assess_merge_simulation_report to reject local merge simulations listed
as read-only diagnostics and require Worktree/index mutations documentation
with worktree path, pre/merge/abort/post-clean proof when simulation
commands appear in the session log.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:13:30 -04:00
sysadmin 6bf0210128 Merge pull request 'feat: enforce branches-only baseline validation in reviewer reports (Closes #325)' (#349) from feat/issue-325-baseline-worktree-validation into master 2026-07-07 04:13:22 -05:00
sysadminandClaude Opus 4.8 56661cd0a7 feat: require live blocker proof before skipping earlier PRs (Closes #318)
Add assess_prior_blocker_skip_proof to validate reviewer reports document
live REQUEST_CHANGES blocker proof for skipped earlier open PRs, reject
stale-memory skips when head changed after blocker, and require
BLOCKER_STATUS_UNVERIFIED when review feedback cannot be fetched.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:12:01 -04:00
sysadminandClaude Opus 4.8 c2bba27b86 feat: reject legacy Workspace mutations field in reviewer handoffs (Closes #320)
Reviewer controller handoffs must now report the precise mutation
categories (File edits by reviewer, Worktree/index mutations, Git ref
mutations, MCP/Gitea mutations, Review mutations, Merge mutations,
Cleanup mutations, External-state mutations, Read-only diagnostics)
instead of the ambiguous legacy "Workspace mutations" field.

- assess_controller_handoff(role="review") no longer requires
  "Workspace mutations" and fails closed when any legacy
  "Workspace mutations:" line is present, regardless of value.
- New HANDOFF_REVIEW_MUTATION_FIELDS makes the nine precise categories
  required fields for review-role handoffs, matching the canonical
  schemas/review-merge-final-report.md field set.
- Tests cover approval, request-changes, merge, worktree cleanup,
  fetch-only, and blocked handoffs, plus rejection of
  "Workspace mutations: none" and non-none values.
- Author/create-issue/inventory roles keep their existing contracts.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:11:05 -04:00
sysadminandClaude Opus 4.8 9fbe556466 feat: reject workspace-none claims after worktree mutations (Closes #313)
Add assess_workspace_mutation_consistency to review_proofs: final
reports fail validation when they claim "Workspace mutations: none"
while worktree mutations (reset --hard, checkout, clean, worktree
add/remove, stash, merge, rebase, ...) occurred — either observed in
the workflow command log or self-reported in the report's own
"Worktree mutations" field. Observed worktree and git ref mutations
must be reported under their precise category fields; reports pass
when they say "File edits by reviewer: none" alongside a non-none
"Worktree mutations" entry. Tests cover reset, checkout, worktree add,
clean, fetch-only, and no-op cases.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:09:44 -04:00
sysadmin 407483f6e0 Merge pull request 'feat: enforce session hard-stop after terminal review mutation (Closes #332)' (#345) from feat/issue-332-terminal-review-hard-stop into master 2026-07-07 04:09:19 -05:00
sysadminandClaude Opus 4.8 cf1cedf5e3 feat: require conflict proof for skipped non-mergeable PRs (Closes #322)
Add assess_non_mergeable_skip_proof to validate reviewer reports document
head SHA, mergeability result, conflict proof command, conflicting files,
and head-change status for each skipped non-mergeable PR, or classify
MERGEABILITY_UNVERIFIED when proof cannot be retrieved.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:08:53 -04:00
sysadminandClaude Opus 4.8 da9c0ba54c feat: enforce validation environment proof in reviewer reports (Closes #311)
Add assess_validation_environment_proof requiring exact validation command,
working directory, result evidence, and bare-pytest executable/version/venv
proof before reviewer reports can claim validation success.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:08:42 -04:00
sysadmin d7d2f92286 Merge pull request 'feat(reports): classify git fetch as a git ref mutation (closes #297)' (#344) from feat/issue-297-fetch-ref-mutation into master 2026-07-07 04:07:11 -05:00
sysadmin cbae9a3ecd Merge pull request 'docs: pin GlitchTip filing orchestrator contract (Closes #153)' (#343) from feat/issue-153-glitchtip-filing-orchestrator into master 2026-07-07 04:05:22 -05:00
sysadmin f92d1a29c2 Merge pull request 'Add pagination metadata to gitea_list_prs' (#342) from feat/issue-340-list-prs-pagination-metadata into master 2026-07-07 04:03:30 -05:00
sysadminandClaude Opus 4.8 a47083b9b8 feat: enforce branches-only baseline validation in reviewer reports (Closes #325)
Add assess_reviewer_baseline_validation_proof to block test-suite runs in
the main checkout and require complete baseline worktree proof when reports
claim pre-existing master failures.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 05:02:07 -04:00
sysadmin 0ae39c231d Merge pull request 'feat: add create-issue final-report verifier (Closes #337)' (#341) from feat/issue-337-create-issue-verifier into master 2026-07-07 03:59:00 -05:00
sysadmin 6c653ce32a feat: add queue-status report verifier for reviewer handoffs (#339)
Add assess_queue_status_report and assess_proof_wording to reject
contradictory queue-status-only reports: passed gates without a selected
PR, prior diagnostic conflict proof, pagination assumptions, and
blocker contradictions. Wire into build_final_report and document the
queue-status-only schema path.

Closes #339
2026-07-07 04:57:50 -04:00
sysadmin 5e145957a5 Merge pull request 'feat: extract review-merge-pr workflow from monolithic LLM skill' (#335) from feat/issue-334-review-merge-workflow-extract into master 2026-07-07 03:57:18 -05:00
sysadmin 46703eac3f feat: enforce proof wording for pagination and live claims (#330)
Add assess_proof_wording to reject unsupported phrases such as
inventory complete, live proof, same as master, and file edits none
unless matching evidence is present. Wire into build_final_report and
add doc-contract tests.

Closes #330
2026-07-07 04:46:48 -04:00
sysadmin 6e024db9eb Merge remote-tracking branch 'prgs/master' into feat/issue-337-create-issue-verifier 2026-07-07 04:44:03 -04:00
sysadmin fd6f7ff22c Merge pull request 'docs: add subagent tool-budget guardrails for MCP tasks (Closes #259)' (#284) from docs/issue-259-subagent-tool-budget into master 2026-07-07 03:43:47 -05:00
sysadmin a06af6f0e3 Merge pull request 'feat: merged PR cleanup reconciliation report (Closes #269)' (#282) from feat/issue-269-merged-pr-cleanup-reconcile into master 2026-07-07 03:39:48 -05:00
sysadmin 72c20ee1e1 Merge remote-tracking branch 'prgs/master' into feat/issue-334-review-merge-workflow-extract 2026-07-07 04:38:09 -04:00
sysadmin eb6f006a31 Merge pull request 'fix: recompute infra_stop live on capability resolve (Closes #285)' (#286) from feat/issue-285-live-infra-stop-recompute into master 2026-07-07 03:37:22 -05:00
sysadminandClaude Opus 4.8 3a490521e4 Merge prgs/master into feat/issue-269-merged-pr-cleanup-reconcile
Resolve task_capability_map.py conflict by keeping all three capability
entries: commit_files and gitea_commit_files (gitea.repo.commit, #262,
from master) and reconcile_merged_cleanups (gitea.read, #269, this
branch). Capability boundaries unchanged — reconciliation reporting stays
read-only; no close/merge capability broadened.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 04:37:11 -04:00
sysadmin 3c50db61eb merge: integrate remote branch history 2026-07-07 04:35:59 -04:00
sysadmin edf26410a0 merge: resolve master conflicts for #259 subagent tool-budget docs
Integrate split-workflow router SKILL.md (#333) with subagent tool-budget
guardrails (#259). Keep shell spawn hard-stop (#258) and subagent
delegation (#266) operator-guide rules alongside subagent_tool_budget.
2026-07-07 04:35:46 -04:00
sysadmin e1fbf0112a test: mock assess_infra_stop in route_task_session merge blocker test
route_task_session now calls assess_infra_stop() live instead of
check_mid_merge(). Update TestMCPHealth.test_route_task_session_blocks_during_merge
to mock the infra-stop assessment path and assert infra_stop_assessment
is returned when blocked.

Part of #285 / PR #286 review fix.
2026-07-07 04:35:34 -04:00
sysadminandClaude Opus 4.8 456e46d3fc feat: enforce session hard-stop after terminal review mutation (#332)
Extends the single-terminal-decision machinery (#211) to session-level
hard-stop semantics:

- terminal_review_hard_stop_reasons: after a live terminal verdict, only
  the merge sequence for the same approved PR may continue; REQUEST_CHANGES
  stops the run; operator-approved corrections re-open the review path
  only, never a cross-PR merge.
- gitea_merge_pr: new Gate 2b consults the hard-stop before any API call,
  so a run can never verdict one PR and merge another.
- gitea_mark_final_review_decision: hard-stop check with explicit stop
  guidance naming the consumed terminal mutation; plus duplicate
  REQUEST_CHANGES suppression — an unresolved request-changes at the
  current head SHA refuses a second request-changes (fail closed when
  feedback cannot be verified).
- tests/test_terminal_review_hard_stop.py: 13 tests covering the
  acceptance criteria; existing TestSubmitPrReview setups stub the new
  feedback fetch.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 04:32:15 -04:00
sysadmin 25c9d20870 merge: resolve PR #335 conflicts with master (fd396df)
Keep master router-style SKILL.md (full #333 split) and merge #334
review-merge test coverage into test_llm_workflow_split.py.

Conflicts resolved in:
- skills/llm-project-workflow/SKILL.md
- tests/test_llm_workflow_split.py
2026-07-07 04:31:42 -04:00
sysadmin 873bcd06e5 docs: pin GlitchTip filing orchestrator contract (Closes #153) 2026-07-07 04:27:02 -04:00
sysadminandClaude Opus 4.8 c41a698a08 feat(reports): classify git fetch as a git ref mutation (closes #297)
Reviewer reports treated 'git fetch prgs master' as read-only
diagnostics while claiming 'Git/worktree mutations: None'. Fetch edits
no files but updates remote-tracking refs, so hiding it under
diagnostics understates what the run mutated.

Add git_ref_mutating_commands() classifying fetch, pull, push, merge,
update-ref, remote update, branch -d/-D, and reset --hard as git ref
mutations, and assess_git_ref_mutation_report() which fails a report
when ref-updating commands ran but the report lacks a non-none 'Git ref
mutations' entry, omits 'fetched <remote>/<branch>' for a targeted
fetch, still claims 'Git/worktree mutations: None', or lists the
command under read-only diagnostics.

Tests cover fetch-only review, missing ledger entry, missing fetch
target, contradictory none-claim, read-only misclassification, worktree
creation (not a ref mutation), merge, cleanup branch deletion,
no-mutation blocked handoff, and dict command-log entries.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 04:26:58 -04:00
sysadmin 5c4153ad57 Merge branch 'master' into feat/issue-285-live-infra-stop-recompute 2026-07-07 03:25:51 -05:00
sysadmin 8929286276 feat: add pagination metadata to gitea_list_prs (#340)
Return wrapped {prs, pagination} from gitea_list_prs with has_more,
inventory_complete, and page traversal via api_fetch_page. Route
gitea_review_pr inventory through gitea_list_prs so trust gates can
prove pagination finality. Add assess_list_prs_pagination_proof verifier.

Closes #340
2026-07-07 04:25:02 -04:00
sysadmin 14f5c865b3 feat: add create-issue final-report verifier (#337)
Extend review_proofs with assess_create_issue_final_report to enforce
workflow-source proof, create-issue handoff schema fields, and rejection
of work-issue/review-merge cross-mode handoff fields.

Closes #337
2026-07-07 04:23:45 -04:00
sysadmin fd396df334 Merge pull request 'Add final-report mutation ledger verifier (#331)' (#338) from feat/issue-331-mutation-ledger-verifier into master 2026-07-07 03:10:30 -05:00
sysadminandClaude Opus 4.8 ae4dd0ff91 Add final-report mutation ledger verifier (#331)
Introduce assess_mutation_ledger_report so reviewer final reports cannot
claim no file edits when the action log records performed mutations.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 04:08:27 -04:00
sysadmin c73e68bd75 Merge pull request 'feat(reports): add no-email identity summary and disclosure validator (closes #305)' (#329) from feat/issue-305-report-email-redaction into master 2026-07-07 03:03:12 -05:00
sysadmin a73d5b351e Merge pull request 'Add REQUEST_CHANGES override proof helper before approval (#326)' (#328) from feat/issue-326-request-changes-override-proof into master 2026-07-07 02:59:33 -05:00
sysadmin 7fdf803a50 Merge pull request 'feat: add fail-closed subagent delegation gates (Closes #266)' (#291) from feat/issue-266-subagent-gate-inheritance into master 2026-07-07 02:56:00 -05:00
sysadmin f7621b155e Merge pull request 'docs: split LLM project workflow into task-specific modes' (#336) from feat/issue-333-work-issue-workflow-extract into master 2026-07-07 02:47:03 -05:00
sysadmin 6200e6bbec docs: split LLM project workflow into task-specific modes (Closes #333)
Consolidate all four canonical workflow files and router SKILL.md:
- review-merge-pr.md (from #334 canonical prompt)
- reconcile-landed-pr.md (reconciliation gates + handoff schema)
- create-issue.md (canonical issue-creation prompt)
- work-issue.md (canonical author/coder prompt)

SKILL.md is now a short router with mode isolation and universal rules.
Adds final-report schemas for all modes and 13 doc-contract tests.

Supersedes partial split work in PR #335; full #333 acceptance in one PR.
2026-07-07 03:44:21 -04:00
sysadmin 8d7203d701 feat: extract canonical work-issue workflow from monolithic skill (Closes #333)
Add workflows/work-issue.md with full author/coder gates (§0–§30),
work-issue final-report schema, task-mode router entry, start-issue template
link, and doc-contract tests.

Part of #333 workflow split (#334 review-merge extraction remains in PR #335).
2026-07-07 03:40:13 -04:00
sysadminandClaude Opus 4.8 c05b2f4c05 docs(skill): load canonical review-merge prompt into workflow file
Replace summary workflow with full §0–§38 canonical PR review/merge rules.
Align review-merge-final-report schema with §37 controller handoff fields.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 03:37:42 -04:00
sysadminandClaude Opus 4.8 3976f886e4 feat(skill): extract review-merge-pr workflow from monolithic SKILL.md
Closes #334

- Add workflows/review-merge-pr.md with reviewer-only gates and steps
- Add schemas/review-merge-final-report.md with mutation ledger fields
- Turn SKILL.md into task-mode router; replace inlined §F/§G with links
- Point review/merge templates at extracted workflow
- Add doc-contract tests for split structure

Part of #333.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 03:32:59 -04:00
sysadmin e4d78a5013 Merge pull request 'test: make sibling-worktree skip test hermetic (closes #283)' (#288) from fix/issue-283-hermetic-walk-root-test into master 2026-07-07 02:08:38 -05:00
sysadmin e96f5bda7a Merge pull request 'feat: add validation ledger and final-report verifier (Closes #271)' (#287) from feat/issue-271-validation-ledger-verifier into master 2026-07-07 02:03:33 -05:00
sysadmin f5f364739b Merge branch 'master' into feat/issue-271-validation-ledger-verifier 2026-07-07 02:01:25 -05:00
sysadminandClaude Opus 4.8 a3b79736e0 feat(reports): add no-email identity summary and disclosure validator (closes #305)
Workflow reports included the authenticated user's personal email even
when username/profile identity was sufficient (observed: 'Identity:
jcwalker3 / [email protected]' in a reconciliation handoff).

Add format_identity_summary() producing the canonical no-email identity
line ('<username> / <profile>' with optional role/remote, reducing an
email passed in the username slot to its local part), and
assess_email_disclosure() flagging personal email addresses in report
text unless the report itself justifies the disclosure (tool proof,
explicit request, or identity disambiguation).

Tests cover no-email summaries for author and reviewer identities,
role/remote suffixes, email-in-username reduction, clean reports,
single and multiple unjustified disclosures, and justified
disambiguation.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 02:57:45 -04:00
sysadminandClaude Opus 4.8 096bb878d5 Merge prgs/master into feat/issue-266-subagent-gate-inheritance
Resolve _GUIDE_RULES conflict by keeping both operator guide keys:
shell_spawn_hard_stop (#258, landed on master via PR #281) and
subagent_delegation (#266, this branch).

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 02:55:15 -04:00
sysadminandClaude Opus 4.8 efb0a2d076 Add REQUEST_CHANGES override proof helper before approval (#326)
Introduce assess_request_changes_approval_proof in review_proofs.py so
review workflows can fail closed when approving an unchanged head after a
prior REQUEST_CHANGES unless explicit override proof is provided.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 02:53:25 -04:00
sysadmin a5341684f6 Merge pull request 'feat: align claim/lock gates with branches-only worktrees (Closes #275)' (#280) from feat/issue-275-claim-lock-branches-worktree into master 2026-07-07 01:39:35 -05:00
sysadmin 2a28fee7f2 Merge pull request 'docs: add Shell Spawn Hard-Stop Rule for agent workflows (Closes #258)' (#281) from docs/issue-258-shell-spawn-hard-stop into master 2026-07-07 01:18:50 -05:00
sysadmin 95603b24a8 Merge pull request 'docs: forbid improvised commit fallbacks when MCP commit exists (Closes #260)' (#279) from docs/issue-260-forbid-commit-fallbacks into master 2026-07-07 01:01:16 -05:00
sysadmin ceb22e0d65 Merge branch 'master' into fix/issue-283-hermetic-walk-root-test 2026-07-07 01:01:11 -05:00
sysadmin aae189a801 Merge branch 'master' into feat/issue-271-validation-ledger-verifier 2026-07-07 01:01:01 -05:00
sysadmin 988a5cd975 Merge branch 'master' into feat/issue-285-live-infra-stop-recompute 2026-07-07 01:00:48 -05:00
sysadmin 52b729e86a Merge branch 'master' into docs/issue-259-subagent-tool-budget 2026-07-07 01:00:37 -05:00
sysadmin 3354da1973 Merge branch 'master' into docs/issue-258-shell-spawn-hard-stop 2026-07-07 01:00:06 -05:00
sysadmin bcf64928b1 Merge branch 'master' into feat/issue-275-claim-lock-branches-worktree 2026-07-07 00:59:50 -05:00
sysadmin d0e250e860 Merge branch 'master' into docs/issue-260-forbid-commit-fallbacks 2026-07-07 00:59:30 -05:00
sysadmin a4736a2346 Merge branch 'master' into feat/issue-266-subagent-gate-inheritance 2026-07-07 00:56:30 -05:00
sysadmin 042783076f Merge pull request 'docs: agent temp artifact cleanup guidance (Closes #261)' (#277) from docs/issue-261-agent-artifact-cleanup into master 2026-07-06 23:38:44 -05:00
sysadmin 8fc41ccf29 Merge pull request 'feat: map gitea_commit_files in task capability resolver (Closes #262)' (#276) from feat/issue-262-commit-files-capability into master 2026-07-06 23:35:31 -05:00
sysadminandClaude Opus 4.8 e09df4a348 feat: add fail-closed subagent delegation gates (#266)
Adds subagent_gate.py following the author_proofs/review_proofs pure-policy
pattern:

- assess_subagent_delegation: deterministic write tasks (issue claim,
  branch creation, code edits, commits, PR creation, review, merge,
  cleanup) are blocked for subagents unless explicitly allowed with a
  recorded justification and the full inherited gate context (issue lock,
  branch/worktree path, identity/profile, allowed tool class, command
  deny list, validation ledger requirement, final report schema);
  read-only delegation stays allowed; unknown task types fail closed.
- validate_subagent_report: subagent output is accepted only when it
  carries the same proof fields as a parent workflow final report.
- Operator guide: new subagent_delegation rule in _GUIDE_RULES so MCP
  clients discover the policy.
- tests/test_subagent_gate.py: 13 tests covering blocked write
  delegation, justification requirement, missing/blank inherited
  context, allowed read-only delegation, unknown tasks, and invalid
  subagent final reports.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 00:28:32 -04:00
sysadminandClaude Opus 4.8 b5a025102d feat: add validation ledger and final-report verifier (Closes #271)
Introduce validation_ledger with command ledger format, full-suite claim
checks, and structured final-report verification so sessions cannot
overclaim validation or misreport review/merge state.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 00:22:28 -04:00
sysadminandClaude Opus 4.8 5af1b796ef fix: recompute infra_stop live on every capability resolve (#285)
Add assess_infra_stop() to scan git state and conflict markers on each
reviewer capability check, return the exact conflict path in responses,
and clear stale capability-stop terminal state once the worktree is clean.

Closes #285

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 00:21:18 -04:00
sysadminandClaude Opus 4.8 2b04108f93 test: make sibling-worktree skip test hermetic (closes #283)
test_skip_scan_walk_root_skips_sibling_worktrees_only derived its
fixture paths from __file__, so the branches/ sibling-worktree skip
assertion only held when the suite ran from a branches/<slug> worktree
whose orchestration checkout has a real branches/ directory —
skip_python_scan_walk_root skips a branches/ walk root only when
<project_root>/branches exists on disk. From the main checkout or any
plain clone the test failed with 'AssertionError: False is not true'.

Build the <main>/branches/<worktree> layout in a TemporaryDirectory and
assert against those paths instead. All four behavioural assertions are
preserved and no production code changes.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 00:19:43 -04:00
sysadminandClaude Opus 4.8 82e6d3a3f4 docs: add subagent tool-budget guardrails for MCP tasks (Closes #259)
Document default tool-call and wall-time budgets for deterministic MCP
work, forbid subagent commit delegation when gitea_commit_files is
available, and pin the rules in runbooks, portable skill, operator guide,
and doc-contract tests.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 00:17:26 -04:00
sysadminandClaude Opus 4.8 4cc31a42f9 feat: add merged PR cleanup reconciliation report (#269)
Implement gitea_reconcile_merged_cleanups with dry-run reporting for merged
PR remote branches and local branches/ worktrees, plus confirmation-gated
execute mode. Safety gates cover protected branches, open PR references,
active issue locks, dirty worktrees, and delete_branch capability.

Closes #269

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-07 00:16:42 -04:00
sysadminandClaude Opus 4.8 cf292e4166 docs: add Shell Spawn Hard-Stop Rule for agent workflows (#258)
Documents fail-closed behavior when the shell executor fails to spawn
(exit_code: -1, empty stdout/stderr): probe once, mark shell unavailable,
hard-stop after two consecutive spawn failures, and emit a recovery report
(restart session, kill hung background terminals, prefer MCP-native paths)
instead of retry-spiralling.

- skills/llm-project-workflow/SKILL.md: portable rule section
- docs/llm-workflow-runbooks.md: Gitea runbook section
- gitea_mcp_server.py: shell_spawn_hard_stop operator guide rule
- tests/test_shell_spawn_hard_stop_docs.py: doc-contract tests (5)
- tests/test_operator_guide.py: require the new guide key

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:52:12 -04:00
sysadmin d73a070b5f Merge remote-tracking branch 'origin/feat/issue-275-claim-lock-branches-worktree' into feat/issue-275-claim-lock-branches-worktree 2026-07-06 14:51:02 -04:00
sysadminandClaude Opus 4.8 5e64c96851 feat: align claim/lock gates with branches-only worktrees (#275)
Allow issue lock from base-equivalent branches/ worktrees instead of
requiring the literal master/main branch name. Runtime preflight and
mark_issue/lock_issue now inspect the declared active task workspace so
dirty control-checkout state does not block clean task worktrees.

Closes #275

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:49:39 -04:00
sysadminandClaude Opus 4.8 67e34baece feat: align claim/lock gates with branches-only worktrees (#275)
Allow issue lock from base-equivalent branches/ worktrees instead of
requiring the literal master/main branch name. Runtime preflight and
mark_issue/lock_issue now inspect the declared active task workspace so
dirty control-checkout state does not block clean task worktrees.

Closes #275

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:47:56 -04:00
sysadminandClaude Opus 4.8 2f4672218c docs: forbid improvised commit fallbacks when MCP commit exists (#260)
Document MCP-native gitea_commit_files as the only approved commit path when
gitea.repo.commit is allowed; ban WebFetch, Playwright, and manual base64
workarounds in runbooks and safety model. Add doc tests.

Closes #260.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:44:41 -04:00
sysadmin b50d265247 Merge branch 'feat/issue-263-native-commit-payloads' into docs/issue-261-agent-artifact-cleanup 2026-07-06 14:44:33 -04:00
sysadmin 2a544b78d1 feat(mcp): support native commit payload preparation without shell (closes #263) 2026-07-06 14:42:59 -04:00
sysadminandClaude Opus 4.8 850b7c34b6 docs: agent temp artifact cleanup guidance and detection (#261)
Add runbook section, root-level .gitignore patterns, and detection helper
for _encode_/_emit_/_inline_ throwaway scripts. Surface non-blocking
warnings from assess_preflight_status and gitea_lock_issue.

Closes #261

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:42:33 -04:00
sysadminandClaude Opus 4.8 f464716e89 docs: document agent temp artifact cleanup and gitignore patterns (#261)
Add runbook guidance for deleting throwaway _encode/_emit/_inline commit
helpers after MCP commit attempts, ignore those patterns in .gitignore,
and cross-link the worktree-cleanup template.

Closes #261.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:42:22 -04:00
sysadmin 67634c399e Merge branch 'feat/issue-262-commit-files-capability' into feat/issue-263-native-commit-payloads 2026-07-06 14:41:07 -04:00
sysadmin 89c9b9a4ec feat(mcp): add commit_files gates and tests for capability resolver 2026-07-06 14:39:26 -04:00
sysadminandClaude Opus 4.8 23c6593b07 feat: map commit_files in task capability resolver (#262)
Add commit_files and gitea_commit_files resolver entries for
gitea.repo.commit, align gitea_commit_files tool gates with other
issue-write mutations, and add regression tests for author allow,
reviewer deny, preflight order, and resolver/tool alignment.

Refs #262

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:38:35 -04:00
sysadmin d6f4f936e3 Merge pull request 'feat(issue-filing): harden final report proofs for handoff and mutations (closes #191)' (#241) from feat/issue-191-issue-filing-reports into master 2026-07-06 13:37:41 -05:00
sysadminandClaude Opus 4.8 24d8891424 fix(issue-filing): enforce mutation-scope denials in generic loop (#191)
The forbidden-mutation loop in assess_issue_filing_mutation_scope never
appended reasons; align aggregation with issue-selection handoff checks.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:36:18 -04:00
sysadmin c1e47a5692 Merge pull request 'feat: expose role-aware runtime context capabilities (#139)' (#257) from feat/issue-139-role-aware-runtime-context into master 2026-07-06 13:34:37 -05:00
sysadminandClaude Opus 4.8 27922c67f8 feat(issue-filing): harden final report proofs for handoff and mutations (#191)
Add assess_issue_filing_final_report and supporting checks for Controller
Handoff (issue_filing role), exact issue number/title, full 40-char SHAs,
per-mutation capability proof, single-mutation scope statements, and
duplicate-search summaries. Workflow skill documents the issue-filing
handoff fields.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:32:48 -04:00
sysadminandClaude Opus 4.8 95d01b07fe feat: add session task capabilities to runtime context (#139)
gitea_get_runtime_context now reports per-task allowed_in_current_session
flags, matching configured profiles, and aggregate session_capabilities so
LLMs can determine author/comment/review/merge ability without guessing
launcher profile config.

Refs #139

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:32:07 -04:00
sysadmin 183e9f08b8 Merge pull request 'docs: separate Jenkins trigger onto jenkins-write-mcp boundary (#152)' (#264) from codex/issue-152-jenkins-trigger-boundary into master 2026-07-06 13:26:16 -05:00
sysadmin eb3560d949 Merge pull request 'docs: add Work Selection Rule for LLM operating rules (#272)' (#273) from docs/issue-272-work-selection-lease-rule into master 2026-07-06 13:23:17 -05:00
sysadminandClaude Opus 4.8 216fa5cf46 docs: add Global LLM Worktree Rule to operating rules (#272)
Require mutations only from branches/ worktrees after proving project
root, cwd, branch, stable main-checkout branch, and session worktree
path. Surface in portable skill, runbooks, wiki, operator guide, and
start-issue template.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:09:44 -04:00
sysadminandClaude Opus 4.8 7a28d09b5d fix(mcp): line-anchor conflict scan and worktree scan paths (#272)
Match git conflict markers on whole lines only so decorative equals
borders in mcp_discoverability.py no longer false-positive. Scan
session worktrees under branches/ without skipping the entire tree
when the worktree path contains "branches".

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 14:09:44 -04:00
sysadminandClaude Opus 4.8 2111c84e7d docs: add Work Selection Rule for LLM operating rules
Require lease verification (open PRs, branches, worktrees, leases,
merged completion) before any issue or PR work. Surface the rule in
the portable skill, Gitea runbooks, operator guide, and start-issue
template; add doc-contract tests.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 13:45:25 -04:00
sysadmin 0716157fa5 Merge pull request 'feat(mcp): implement operation-scoped role selection and automatic dispatch switching (#228)' (#239) from feat/issue-228-operation-scoped-role-selection into master 2026-07-06 12:39:07 -05:00
sysadmin 5798871cc2 Merge pull request 'feat(mcp): add integration and discoverability coverage for Jenkins/GlitchTip' (#255) from feat/issue-155-jenkins-glitchtip-discoverability into master 2026-07-06 12:37:41 -05:00
sysadmin 84ed137f66 Merge pull request 'fix: scope pre-flight workspace checks to session-owned deltas (closes #252)' (#254) from feat/issue-252-preflight-foreign-edits into master 2026-07-06 12:35:45 -05:00
sysadminandClaude Opus 4.8 b4e04f4dfb docs: separate Jenkins trigger onto jenkins-write-mcp boundary (#152)
Document that build triggers live on the gated jenkins-write-mcp server,
not the read-only jenkins-mcp surface. Add registration and safety-model
cross-links plus doc tests that fail if trigger tools reappear on the read
boundary.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 13:26:58 -04:00
sysadminandClaude Opus 4.8 4faf839dab test: align preflight tests with session-scoped porcelain baselines
Update TestPreflightVerification for #252 delta-based preflight checks
on the rebased #228 branch.

Refs #228

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 12:39:01 -04:00
sysadmin 7dde2f5405 fix(mcp): implement operation-scoped role selection and auto profile switching 2026-07-06 12:38:20 -04:00
sysadmin 75ed0632b4 feat(mcp): implement operation-scoped role selection and automatic dispatch switching (#228) 2026-07-06 12:37:30 -04:00
sysadmin 1fd929040c feat(mcp): add integration and discoverability coverage for Jenkins/GlitchTip 2026-07-06 12:37:04 -04:00
sysadminandClaude Opus 4.8 f2c8a8d5c1 fix: scope pre-flight workspace checks to session-owned deltas (#252)
Pre-flight order tracking now compares porcelain baselines captured at
process start, whoami, and capability resolution instead of treating any
shared-worktree dirt as a sticky violation. Fresh gitea_whoami re-evaluates
and clears violations; runtime context reports preflight_block_reasons when
mutations would fail.

Refs #252

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 12:34:45 -04:00
sysadmin 5965904c60 Merge pull request 'feat(reviewer): add queue-target lock before PR inventory trust gate (closes #200)' (#240) from feat/issue-200-queue-target-lock into master 2026-07-06 11:29:55 -05:00
sysadminandClaude Opus 4.8 70c868962a feat(reviewer): add queue-target lock before PR inventory trust gate (#200)
Operators can supply an active PR backlog for one repo while agents
inventory another and falsely report trusted_empty. reconcile_queue_target
locks the target repo from operator context/backlog before list_prs runs;
pr_inventory_trust_gate blocks target_repo_mismatch and unresolved locks.
assess_reviewer_queue_inventory wires the lock into the canonical path.

Tests cover Gitea-Tools backlog vs mcp-control-plane inventory mismatch,
ambiguous dual-repo context, conflicting directive vs backlog, and final-
report field requirements.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 12:29:16 -04:00
sysadmin 08ed5a82d2 Merge pull request 'feat(author): add continuation-mode selection wall for open-PR issues (closes #188)' (#246) from feat/issue-188-continuation-selection-wall into master 2026-07-06 11:26:10 -05:00
sysadmin a887da1f8f Merge pull request 'feat(mcp): bind issue lock to author scratch worktree (closes #249)' (#250) from feat/issue-249-issue-lock-scratch-worktree into master 2026-07-06 11:24:37 -05:00
sysadmin f94cb80fc9 Merge pull request 'fix(mcp): require visible APPROVED reviews before merge (closes #244)' (#251) from feat/issue-244-review-submit-visible into master 2026-07-06 11:22:14 -05:00
sysadmin 6e27911733 Merge pull request 'fix(mcp): clear sticky reviewer denial on allowed author task route (closes #238)' (#242) from feat/issue-238-sticky-denial-clear into master 2026-07-06 11:22:00 -05:00
sysadminandClaude Opus 4.8 9d8ab0a7b5 test: fix PR queue inventory mock for visible APPROVED verification (#244)
Update test_reviewer_profile_can_proceed_from_inventory_to_review_gates
to return APPROVED state on POST and include GET /reviews verification
responses required by #244 review submission gates.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 12:21:26 -04:00
sysadmin c502ae30d6 Merge pull request 'test: make test_health portable for scratch clones (closes #245)' (#248) from feat/issue-245-test-health-portable into master 2026-07-06 11:20:21 -05:00
sysadmin ff435ea13c Merge pull request 'fix(reviewer): close forbidden-git detection gaps via allowlist (closes #243)' (#247) from feat/issue-243-forbidden-git-gaps into master 2026-07-06 11:18:32 -05:00
sysadmin a1ba69eebb fix(mcp): require visible APPROVED reviews before merge (closes #244)
Gitea expects review event APPROVED, not APPROVE; the wrong event left
submissions as PENDING drafts. Verify terminal verdicts after submit,
submit pending drafts when needed, and block merge when approval_visible
is false.
2026-07-06 11:51:29 -04:00
sysadminandClaude Opus 4.8 df1104d3e7 feat(mcp): bind issue lock to author scratch worktree (#249)
gitea_lock_issue and gitea_create_pr accept worktree_path so lock
preconditions validate the caller's scratch clone instead of the shared
MCP server CWD. Lock records store the validated path; PR creation fails
closed when the declared path does not match.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 11:46:44 -04:00
sysadmin e441b81d3b Merge branch 'master' into feat/issue-188-continuation-selection-wall 2026-07-06 11:44:59 -04:00
sysadminandClaude Opus 4.8 d422bc0978 test: make test_health portable for scratch clones (#245)
Use sys.executable with optional GITEA_TOOLS_TEST_PYTHON override instead of
hard-coded venv/bin/python. Add subprocess timeouts/teardown and skip when no
interpreter is available. Note scratch-clone norm in review-pr template.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 11:42:34 -04:00
sysadminandClaude Opus 4.8 ec8f6abf5b fix(reviewer): close forbidden-git detection gaps via allowlist (#243)
Replace blocklist-based git mutation detection with fail-closed allowlist
semantics: any git command not matching the readonly set is forbidden.
Catches checkout HEAD --, checkout ., git switch variants, and uncommon
stash subcommands that bypassed the old regex.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 11:42:14 -04:00
sysadminandClaude Opus 4.8 dc41b685d0 feat(author): add continuation-mode selection wall for open-PR issues (#188)
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 11:40:35 -04:00
sysadminandClaude Opus 4.8 63a7ba8287 feat(author): add continuation-mode selection wall for open-PR issues (#188)
Issues already represented by open PRs must not be selected for fresh work
unless the operator explicitly requests continuation. Adds classification,
continuation report proofs (old/new head SHA, PR author, branch), contradictory
no-PR claim detection, and edited-PR inventory coverage checks.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 11:40:20 -04:00
sysadmin bab803ff3d Merge pull request 'test: require issue-lock proof in author handoffs (closes #208)' (#235) from feat/issue-208-issue-lock-proof into master 2026-07-06 10:39:12 -05:00
sysadminandClaude Opus 4.8 4bc02a8c7d fix(reviewer): restore #198 empty-queue helpers for PR #235 (#208)
Re-introduce assess_empty_queue_report and parse_trust_gate_status_from_report
dropped when #208 was reverted from master. Wire build_final_report grading,
inventory handoff trust-gate fields, and tests so capability_stop_terminal
imports resolve while issue-lock proof enforcement remains.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 11:37:07 -04:00
sysadminandClaude Opus 4.8 d4e89f7863 fix(mcp): clear sticky reviewer denial on allowed author task route (#238)
Add sync_from_capability_result() so reviewer terminal mode is
operation-scoped: a later allowed author capability resolution clears
stale denial state and unblocks read-only tools like gitea_list_prs.

Error messages now name the denied task and explain how to clear stale
state. Inline trust-gate parsing fixes broken imports in
assess_capability_stop_report on master.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 11:27:29 -04:00
jcwalker3 ec879df4c2 test: require issue-lock proof in author handoffs (#208) 2026-07-06 10:19:46 -05:00
jcwalker3 056a232ef8 revert: remove #208 from master (land via PR #235) 2026-07-06 10:19:26 -05:00
sysadmin 8fa94a07a8 Merge pull request 'feat(reviewer): require trust-gate proof in empty-queue reports (closes #198)' (#237) from feat/issue-198-empty-queue-trust-gate-report into master 2026-07-06 10:18:55 -05:00
sysadminandClaude Opus 4.8 ca3de3da53 feat(reviewer): require trust-gate proof in empty-queue reports (#198)
Add assess_empty_queue_report to block empty-queue claims without
pr_inventory_trust_gate.status == trusted_empty, reject weak merge-commit
corroboration, extend inventory handoff fields, and wire the gate into
build_final_report and capability-stop terminal validation.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 11:16:35 -04:00
sysadmin be6feabf70 Merge pull request 'feat(reviewer): wire PR inventory trust gate into live queue path (closes #196)' (#236) from feat/issue-196-pr-inventory-trust-gate-live-path into master 2026-07-06 10:14:22 -05:00
sysadminandClaude Opus 4.8 1071619532 feat(reviewer): fail closed on dirty worktrees and forbidden git cleanup (#233)
Add reviewer_worktree proofs that block stash/reset/checkout -- cleanup of
unrelated local files, require scratch worktree reporting when the main tree
is dirty outside PR scope, and integrate the gate into final report grading
and Controller Handoff review fields.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 11:13:38 -04:00
sysadminandClaude Opus 4.8 1033a22407 feat(reviewer): wire PR inventory trust gate into live queue path (#196)
Run pr_inventory_trust_gate on empty gitea_review_pr inventory results,
add assess_reviewer_queue_inventory for multi-repo completeness checks,
and require trusted_empty before empty-queue claims in reports/templates.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 02:08:55 -04:00
sysadmin 7966e70db6 Merge pull request 'feat(reviewer): fail closed on dirty worktrees and forbidden git cleanup (closes #233)' (#234) from feat/issue-233-reviewer-worktree-gate into master 2026-07-06 01:06:24 -05:00
sysadminandClaude Opus 4.8 4f466550ca feat(reviewer): fail closed on dirty worktrees and forbidden git cleanup (#233)
Add reviewer_worktree proofs that block stash/reset/checkout -- cleanup of
unrelated local files, require scratch worktree reporting when the main tree
is dirty outside PR scope, and integrate the gate into final report grading
and Controller Handoff review fields.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 02:04:34 -04:00
sysadminandClaude Opus 4.8 6ac6b9528c test: require issue-lock proof in author handoffs (#208)
Add Issue lock proof to author HANDOFF_ROLE_FIELDS and update controller
handoff tests. Strengthen create_pr tests to read and validate the issue
lock file (branch match, Closes #N, reveal opt-in unchanged).

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 02:02:26 -04:00
sysadmin 4dd32bb9f7 Merge pull request 'docs(wiki): refresh Gitea Wiki publication proof heads (closes #224)' (#231) from feat/issue-224-wiki-proof-refresh into master 2026-07-06 00:57:29 -05:00
sysadmin d5d3331498 docs(wiki): refresh Gitea Wiki publication proof heads (closes #224) 2026-07-06 01:49:59 -04:00
sysadmin 09a766be5d Merge pull request 'Bootstrap Gitea-Tools docs/wiki and wiki sync helper (#224)' (#230) from feat/issue-224-gitea-tools-wiki-bootstrap into master 2026-07-06 00:40:44 -05:00
sysadminandClaude Opus 4.8 342974f840 docs(wiki): record Gitea-Tools live wiki publication proof (#224)
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 01:36:49 -04:00
sysadminandClaude Opus 4.8 f58317d079 Add docs/wiki source tree and Gitea Wiki sync helper (#224)
Bootstrap repo-tracked wiki pages (10 required pages), the operator-
confirmed sync script, PR template publication gate, and safety tests so
Gitea-Tools can publish to the live Gitea Wiki per issue #224.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-06 01:36:10 -04:00
sysadmin 125f52a059 Merge pull request 'Fix test env isolation for credentials and create_pr gate tests' (#229) from feat/issue-227-fix-test-env-isolation into master 2026-07-06 00:29:21 -05:00
sysadmin 3320a5b3e6 test: isolate credentials and create_pr tests from MCP env leakage
Fix 8 suite failures triggered when GITEA_MCP_CONFIG/GITEA_MCP_PROFILE
are set in the operator environment.

- TestGetAuthHeader: clear MCP config env and DYNAMIC_CONFIGS in setUp
- TestCreatePR/TestIssueLocking: add CREATE_PR_ENV author fixture and
  patch reviewer-stop gate so tests reach intended code paths

Closes #227
2026-07-06 01:26:33 -04:00
sysadmin 8436276991 Merge pull request 'Fix MCP registration and naming for Jenkins/GlitchTip servers (Issue #151)' (#226) from feat/issue-151-mcp-registration-naming into master 2026-07-06 00:20:14 -05:00
sysadmin 6971a75818 Merge pull request 'Block role/tool namespace mismatch for author and reviewer mutations (Issue #209)' (#223) from feat/issue-209-namespace-mismatch into master 2026-07-06 00:14:48 -05:00
sysadmin 25c19b26d3 Merge pull request 'Enforce issue-write tool gates to match task capability resolver (mcp-control-plane #69)' (#222) from feat/issue-69-issue-write-tool-gates into master 2026-07-06 00:14:40 -05:00
sysadmin d747e0a2c9 Document external MCP registration for Jenkins and GlitchTip 2026-07-06 01:07:46 -04:00
sysadmin e727210aad Merge pull request 'Isolate Gitea MCP runtime from mutable development worktrees (Issue #221)' (#225) from feat/issue-221-runtime-isolation into master 2026-07-05 23:56:36 -05:00
sysadmin 2f5e0408fd test(review-proofs): add required Workspace mutations field to GOOD_HANDOFF mock 2026-07-05 21:14:37 -04:00
sysadmin 2cd61a90fe feat(runtime-isolation): loader, health checks, tests, and documentation for Issue #221 2026-07-05 21:13:40 -04:00
sysadmin cb974d659f feat(workflow-guard): check mid-merge in gitea_resolve_task_capability 2026-07-05 21:13:38 -04:00
sysadmin 1948eae184 feat(workflow-guard): verify MCP runtime clean/not mid-merge before starting reviewer work 2026-07-05 21:13:15 -04:00
sysadminandClaude Opus 4.8 cbce6cb2e7 Block author mutations from reviewer MCP namespace (Issue #209)
Add role_namespace_gate to fail closed when reviewer-bound sessions attempt
author mutations (create_pr always blocked; create_issue only when profile
explicitly grants gitea.issue.create). Extend mutation audit records with
mcp_namespace, task_role, and operation fields. Add handoff parity checks in
review_proofs for author role vs reviewer namespace mismatches.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-05 21:12:33 -04:00
sysadminandClaude Opus 4.8 00394d43b7 Enforce issue-write tool gates to match task capability resolver (Issue #69)
Extract TASK_CAPABILITY_MAP into task_capability_map.py as the single source
of truth shared by gitea_resolve_task_capability and issue-mutating tools.
Gate gitea_create_issue, gitea_close_issue, gitea_mark_issue, and
gitea_set_issue_labels with structured #142 permission reports. Add regression
tests proving resolver-denied tasks fail closed at the raw tool layer.

Implements mcp-control-plane #69.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-05 21:09:34 -04:00
sysadmin c89ae2eb5c Rename mcp_server.py to gitea_mcp_server.py 2026-07-05 21:07:42 -04:00
sysadmin 380ed5fde1 Merge pull request 'Harden author-run reporting and mutation capability proof per #183' (#187) from feat/issue-183-harden-author-run-reporting into master 2026-07-05 19:44:41 -05:00
sysadmin 5229dedb03 feat(author): harden reporting and claim capability per #183
- Add 'mark_issue' to TASK_MAP in resolve_task_capability for exact mutation proof (closes gap on unknown treated as allowed).
- Add assess_controller_handoff and integrate into build_final_report (downgrades missing handoff).
- Add tests for handoff and update existing.
- Update SKILL.md and review-pr.md to mandate exact 'Controller Handoff' section, exact sweep evidence, PR head SHA in reports.
- Author profile used throughout; no review/merge.

Refs #183
2026-07-05 20:41:34 -04:00
sysadmin 57ac56fcb3 Merge pull request 'feat(workflow): capability stop terminal mode after reviewer denial (Issue #197)' (#218) from feat/issue-197-capability-stop-terminal into master 2026-07-05 19:38:21 -05:00
sysadminandClaude Opus 4.8 7a7e9c0cbc feat: enforce capability stop terminal mode after reviewer denial (Issue #197)
Enter terminal mode when review_pr/merge_pr capability is denied. Block
reviewer queue tools (list_prs, eligibility checks) and add report-purity
validators for forbidden PR selection, fallback, and empty-queue claims.

Closes #197

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-05 20:36:23 -04:00
sysadmin 392266c553 Merge pull request 'feat: block workspace edits before preflight verification Closes #210' (#215) from feat/issue-210-block-workspace-edits-clean into master 2026-07-05 19:34:35 -05:00
sysadmin ab95f1b253 feat(preflight): block workspace edits before identity/capability verification (Closes #210) 2026-07-05 20:31:51 -04:00
sysadmin 8dd1b2ee34 Merge pull request 'feat(author-workflow): enforce exact issue lock before branch/commit/push/PR (Issue #204, recreation of PR #205)' (#213) from feat/issue-204-exact-issue-lock-v2 into master 2026-07-05 19:28:14 -05:00
sysadmin adb35f12b5 Merge pull request 'feat(reviewer-workflow): enforce single terminal review decision per PR review run' (#214) from feat/issue-211-single-terminal-review-decision into master 2026-07-05 19:19:34 -05:00
sysadmin 16f977fde0 fix(review): harden #211 decision lock and CLI gate per review blockers
- Bind review decision state to in-process session (pid + profile lock); drop /tmp file path.
- Fail closed on review_pr.py CLI; route live reviews through gated MCP tools only.
- Require operator_authorized on review correction; allow re-mark after correction.
- Validate remote/org/repo on mark and submit; wire review mutation proof into build_final_report.
- Add security regression tests for spoofed locks, correction flow, and CLI bypass.

Refs #211
2026-07-05 20:16:04 -04:00
sysadmin 7489107768 feat(review-workflow): enforce single terminal review decision per review run (#211) 2026-07-05 17:43:50 -04:00
sysadmin 5d0845df90 Initialize review decision lock in test suite to prevent flakiness and resolve test pollution 2026-07-05 17:43:36 -04:00
sysadmin 880fca81da Enforce single terminal review decision per PR review run (Issue #211)
Reviewer agents could post probe APPROVE/REQUEST_CHANGES reviews while
testing lock paths, polluting PR audit trails. Add a review decision lock
seeded by gitea_resolve_task_capability(review_pr) that requires
gitea_mark_final_review_decision and final_review_decision_ready=True
before gitea_submit_pr_review performs a live mutation.

Add gitea_dry_run_pr_review for read-only submission validation,
gitea_authorize_review_correction for operator-approved fixes, and
assess_review_mutation_final_report for final-report proof. One live
review mutation per run unless correction is explicitly authorized.
2026-07-05 17:43:36 -04:00
8ec69cdd35 feat(author-workflow): enforce exact issue lock before branch/commit/push/PR (Issue #204)
Recreation of the #204 work from closed PR #205 (invalid provenance), rebuilt
cleanly on master under the prgs author identity with no PR #203 content:

- Add gitea_lock_issue MCP tool: locks exactly one issue to its branch name,
  fails closed on branch/issue-number mismatch and on issues already tied to
  an open PR (by head branch or Closes/Fixes reference).
- gitea_create_pr now requires the issue lock: head must match the locked
  branch, title/body must contain Closes/Fixes #<locked issue> exactly, and
  ambiguous references (equivalent / related / same as) are rejected.
- scripts/worktree-start refuses to create an issue-linked worktree unless the
  lock file exists and matches the requested branch.
- assess_controller_handoff rejects handoffs whose selected issue / opened PR
  fields carry multiple numbers or fuzzy equivalence wording.
- Tests: TestIssueLocking (lock + create_pr gates), handoff exact-reference
  tests, worktree-start lock coverage.

Closes #204

Co-Authored-By: Claude Fable 5 <[email protected]>
2026-07-05 17:41:50 -04:00
210 changed files with 48606 additions and 4736 deletions
+9
View File
@@ -46,3 +46,12 @@ GITEA_TOKEN_SOURCE=GITEA_TOKEN
# profile's values. Leave unset for pure env-based configuration.
GITEA_MCP_CONFIG=/Users/jasonwalker/.config/gitea-tools/profiles.json
GITEA_MCP_PROFILE=prgs
# Namespace-scoped active task workspaces (#510). Each MCP namespace uses only
# its own role env var; foreign bindings (e.g. GITEA_AUTHOR_WORKTREE in a
# merger process) are ignored.
# GITEA_AUTHOR_WORKTREE=/path/to/repo/branches/issue-123-work
# GITEA_REVIEWER_WORKTREE=/path/to/repo/branches/review-pr456
# GITEA_MERGER_WORKTREE=/path/to/repo/branches/merge-pr456
# GITEA_RECONCILER_WORKTREE=/path/to/repo/branches/reconcile-pr456
# GITEA_ACTIVE_WORKTREE=/path/to/repo/branches/session-override
+20
View File
@@ -0,0 +1,20 @@
## Description
[Summary of changes and issue number closed.]
Closes #[Issue Number]
## Checklist
- [ ] I have verified my identity matches the required role.
- [ ] No secrets, tokens, keychain IDs, or raw service URLs are committed.
- [ ] All tests pass for touched code.
- [ ] `git diff --check` is clean.
## Documentation and Wiki
- [ ] Does this change require a wiki update (workflows, tools, profiles, runbooks)?
- [ ] If yes, has `docs/wiki/` been updated accordingly?
- [ ] If wiki pages changed, plan the Gitea Wiki sync after merge (`scripts/sync-gitea-wiki.sh`, see Runbooks).
- [ ] Readiness gate (#224): the Gitea Wiki is populated and current for this repo — verify the repo **Wiki tab**, not `docs/wiki/`. If stale or empty, record the required sync as a follow-up before approval.
- [ ] If this PR closes a wiki-related issue: closure requires live Gitea Wiki proof links (Wiki Home plus page listing or wiki git log). Markdown in `docs/wiki/`, sync-helper code, or policy docs alone are not sufficient to close a wiki issue.
+4
View File
@@ -10,3 +10,7 @@ gitea-mcp*.json
.vscode/
graphify-out/
branches/
# Throwaway agent commit-encoding helpers (#261) — never commit.
/_encode_*.py
/_emit_*.py
/_inline_*.py
+8
View File
@@ -172,6 +172,14 @@ Recognized environment fields (see [`.env.example`](.env.example) for placeholde
| `GITEA_MCP_CONFIG` | Optional path to a JSON file defining multiple named runtime profiles. Unset ⇒ pure env behaviour. |
| `GITEA_MCP_PROFILE` | Name of the profile (from `GITEA_MCP_CONFIG`) to activate for this runtime. |
#### External MCP Control Plane servers
Jenkins and GlitchTip are separate MCP trust boundaries, not tools inside this
Gitea MCP runtime. Register them as `jenkins-mcp` and `glitchtip-mcp` in the
client that will use them, then reconnect or reload the client and verify the
expected tools are visible before claiming readiness. See
[`docs/mcp-client-registration.md`](docs/mcp-client-registration.md).
Notes:
- This provides **one token + one profile per process**. It does not implement
+26
View File
@@ -0,0 +1,26 @@
"""Detect throwaway agent helper scripts left in the repo root (#261)."""
from __future__ import annotations
import fnmatch
AGENT_TEMP_BASENAME_PATTERNS = (
"_encode_*.py",
"_emit_*.py",
"_inline_*.py",
)
def find_agent_temp_artifacts_from_porcelain(porcelain: str) -> list[str]:
"""Return untracked repo-root helper paths matching agent temp patterns."""
found: list[str] = []
for line in (porcelain or "").splitlines():
if not line.startswith("??"):
continue
path = line[3:].strip()
if not path or "/" in path or "\\" in path:
continue
basename = path.split("/")[-1]
if any(fnmatch.fnmatch(basename, pat) for pat in AGENT_TEMP_BASENAME_PATTERNS):
found.append(path)
return sorted(found)
+142
View File
@@ -0,0 +1,142 @@
"""Already-landed open PR reconciliation gates (#310).
Reconciler workflows may close an open PR only when the PR head SHA is proven
an ancestor of a freshly fetched target branch. Arbitrary PR closure is denied.
"""
from __future__ import annotations
import subprocess
from typing import Any
from merged_cleanup_reconcile import extract_linked_issue, is_head_ancestor_of_ref
ELIGIBILITY_ALREADY_LANDED = "ALREADY_LANDED_RECONCILE_REQUIRED"
ELIGIBILITY_NOT_LANDED = "NOT_ALREADY_LANDED"
ELIGIBILITY_STALE_TARGET = "TARGET_BRANCH_UNVERIFIED"
ELIGIBILITY_PR_NOT_OPEN = "PR_NOT_OPEN"
def fetch_target_branch(project_root: str, remote: str, branch: str) -> dict[str, Any]:
"""Fetch *branch* from *remote* and return the resolved SHA."""
ref = f"{remote}/{branch}"
fetch = subprocess.run(
["git", "-C", project_root, "fetch", remote, branch],
capture_output=True,
text=True,
check=False,
)
if fetch.returncode != 0:
return {
"success": False,
"target_branch": branch,
"target_ref": ref,
"target_branch_sha": None,
"reasons": [
f"git fetch {remote} {branch} failed: "
f"{(fetch.stderr or fetch.stdout or '').strip()}"
],
}
rev = subprocess.run(
["git", "-C", project_root, "rev-parse", ref],
capture_output=True,
text=True,
check=False,
)
if rev.returncode != 0:
return {
"success": False,
"target_branch": branch,
"target_ref": ref,
"target_branch_sha": None,
"reasons": [
f"git rev-parse {ref} failed: "
f"{(rev.stderr or rev.stdout or '').strip()}"
],
}
return {
"success": True,
"target_branch": branch,
"target_ref": ref,
"target_branch_sha": (rev.stdout or "").strip(),
"reasons": [],
"git_fetch_command": f"git fetch {remote} {branch}",
}
def assess_already_landed_reconciliation(
*,
pr: dict[str, Any],
project_root: str,
remote: str,
target_branch: str,
target_fetch: dict[str, Any] | None = None,
) -> dict[str, Any]:
"""Return eligibility and proof for reconciling an open PR."""
pr_number = int(pr.get("number") or 0)
pr_state = (pr.get("state") or "").strip().lower()
head_sha = (pr.get("head") or {}).get("sha") if isinstance(pr.get("head"), dict) else None
if not head_sha and isinstance(pr.get("head"), str):
head_sha = None
head_ref = (pr.get("head") or {}).get("ref") if isinstance(pr.get("head"), dict) else pr.get("head")
base_ref = (pr.get("base") or {}).get("ref") if isinstance(pr.get("base"), dict) else pr.get("base")
title = pr.get("title") or ""
body = pr.get("body") or ""
fetch_result = target_fetch or fetch_target_branch(project_root, remote, target_branch)
linked_issue = extract_linked_issue(title, body)
result: dict[str, Any] = {
"pr_number": pr_number,
"pr_state": pr_state,
"candidate_head_sha": head_sha,
"head_ref": head_ref,
"base_ref": base_ref or target_branch,
"target_branch": target_branch,
"target_branch_sha": fetch_result.get("target_branch_sha"),
"linked_issue": linked_issue,
"git_ref_mutations": [],
"reasons": [],
}
if fetch_result.get("git_fetch_command"):
result["git_ref_mutations"].append(fetch_result["git_fetch_command"])
if pr_state != "open":
result["eligibility_class"] = ELIGIBILITY_PR_NOT_OPEN
result["ancestor_proof"] = None
result["close_allowed"] = False
result["reasons"].append(f"PR #{pr_number} state is {pr_state!r}, not open")
return result
if not fetch_result.get("success"):
result["eligibility_class"] = ELIGIBILITY_STALE_TARGET
result["ancestor_proof"] = None
result["close_allowed"] = False
result["reasons"].extend(fetch_result.get("reasons") or [])
return result
target_ref = fetch_result.get("target_ref") or f"{remote}/{target_branch}"
ancestor = is_head_ancestor_of_ref(project_root, head_sha, target_ref)
result["ancestor_proof"] = ancestor
if ancestor is None:
result["eligibility_class"] = ELIGIBILITY_STALE_TARGET
result["close_allowed"] = False
result["reasons"].append(
f"ancestor check failed for head {head_sha!r} against {target_ref}"
)
return result
if ancestor:
result["eligibility_class"] = ELIGIBILITY_ALREADY_LANDED
result["close_allowed"] = True
return result
result["eligibility_class"] = ELIGIBILITY_NOT_LANDED
result["close_allowed"] = False
result["reasons"].append(
f"PR head {head_sha} is not an ancestor of {target_ref}"
)
return result
+234
View File
@@ -0,0 +1,234 @@
"""Branches-only author mutation worktree guard (#274).
Author/coder mutations must run from a session-owned worktree under the
project's ``branches/`` directory, never from the stable control checkout.
"""
from __future__ import annotations
import os
import subprocess
BASE_BRANCHES = frozenset({"master", "main", "dev"})
ACTIVE_WORKTREE_ENV = "GITEA_ACTIVE_WORKTREE"
AUTHOR_WORKTREE_ENV = "GITEA_AUTHOR_WORKTREE"
# Author-only: reviewer/merger/reconciler namespaces use role-specific env vars
# via namespace_workspace_binding (#510).
def _normalize_path(path: str) -> str:
return (path or "").replace("\\", "/").rstrip("/")
def is_path_under_branches(path: str, project_root: str | None = None) -> bool:
"""True when *path* resolves inside ``<project_root>/branches/``."""
normalized = _normalize_path(path)
if not normalized:
return False
if "/branches/" in f"{normalized}/":
return True
if normalized.endswith("/branches"):
return True
if project_root:
root = _normalize_path(os.path.realpath(project_root))
real = _normalize_path(os.path.realpath(path))
if real.startswith(f"{root}/"):
rel = real[len(root) + 1 :]
return rel == "branches" or rel.startswith("branches/")
return False
def resolve_mutation_workspace(
worktree_path: str | None,
project_root: str,
*,
active_worktree_env: str | None = None,
author_worktree_env: str | None = None,
) -> str:
"""Resolve the workspace path inspected before author mutations."""
for candidate in (worktree_path, active_worktree_env, author_worktree_env):
text = (candidate or "").strip()
if text:
return os.path.realpath(os.path.abspath(text))
return os.path.realpath(project_root)
def _realpath_git_common_dir(workspace_path: str, common_dir: str) -> str:
"""Resolve ``git rev-parse --git-common-dir`` relative to *workspace_path*."""
raw = (common_dir or "").strip()
if not raw:
return raw
if os.path.isabs(raw):
return os.path.realpath(raw)
return os.path.realpath(os.path.join(workspace_path, raw))
def resolve_canonical_repo_root(workspace_path: str, fallback_project_root: str) -> str:
"""Return the stable repository root for *workspace_path* via git metadata (#460)."""
path = (workspace_path or "").strip()
fallback = os.path.realpath(fallback_project_root)
if not path:
return fallback
try:
res = subprocess.run(
["git", "-C", path, "rev-parse", "--git-common-dir"],
capture_output=True,
text=True,
check=True,
)
common = _realpath_git_common_dir(path, res.stdout)
except Exception:
return fallback
if common.endswith(f"{os.sep}.git"):
return os.path.dirname(common)
if os.path.basename(common) == ".git":
return os.path.dirname(common)
return fallback
def resolve_author_mutation_context(
worktree_path: str | None,
process_project_root: str,
*,
active_worktree_env: str | None = None,
author_worktree_env: str | None = None,
) -> dict:
"""Shared workspace resolution for runtime_context and mutation guards (#460)."""
workspace = resolve_mutation_workspace(
worktree_path,
process_project_root,
active_worktree_env=active_worktree_env,
author_worktree_env=author_worktree_env,
)
process_root = os.path.realpath(process_project_root)
# Canonical repository identity comes from the MCP process checkout (#460),
# not from the declared task workspace being validated.
canonical_root = resolve_canonical_repo_root(process_root, process_root)
return {
"workspace_path": workspace,
"process_project_root": process_root,
"canonical_repo_root": canonical_root,
"roots_aligned": canonical_root == process_root,
}
def assess_workspace_repo_membership(
*,
workspace_path: str,
canonical_repo_root: str,
) -> dict:
"""Fail closed when *workspace_path* is not a git worktree of *canonical_repo_root*."""
workspace = os.path.realpath(workspace_path)
root = os.path.realpath(canonical_repo_root)
reasons: list[str] = []
if not os.path.exists(workspace):
reasons.append(f"worktree path '{workspace}' does not exist")
return _membership_assessment(False, reasons, workspace, root, None)
if not os.path.isdir(workspace):
reasons.append(f"worktree path '{workspace}' is not a directory")
return _membership_assessment(False, reasons, workspace, root, None)
try:
res = subprocess.run(
["git", "-C", workspace, "rev-parse", "--git-common-dir"],
capture_output=True,
text=True,
check=True,
)
common_dir = _realpath_git_common_dir(workspace, res.stdout)
except Exception:
reasons.append(f"worktree '{workspace}' is not a valid git repository")
return _membership_assessment(False, reasons, workspace, root, None)
expected_dir = os.path.realpath(os.path.join(root, ".git"))
if common_dir != expected_dir:
reasons.append(
f"worktree '{workspace}' does not belong to the target repository '{root}'"
)
return _membership_assessment(not reasons, reasons, workspace, root, common_dir)
def _membership_assessment(
proven: bool,
reasons: list[str],
workspace: str,
root: str,
common_dir: str | None,
) -> dict:
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"workspace_path": workspace,
"canonical_repo_root": root,
"git_common_dir": common_dir,
}
def format_workspace_repo_membership_error(assessment: dict) -> str:
workspace = assessment.get("workspace_path") or "(unknown)"
root = assessment.get("canonical_repo_root") or "(unknown)"
reasons = "; ".join(assessment.get("reasons") or ["unknown repository membership violation"])
return (
f"Branches-only mutation guard (#274): {reasons} (fail closed). "
f"canonical repository root: {root}; workspace: {workspace}."
)
def assess_author_mutation_worktree(
*,
workspace_path: str,
project_root: str,
current_branch: str | None = None,
base_branches: frozenset[str] | None = None,
) -> dict:
"""Fail closed when author mutations are not rooted under ``branches/``."""
bases = base_branches or BASE_BRANCHES
reasons: list[str] = []
root = os.path.realpath(project_root)
workspace = os.path.realpath(workspace_path)
branch = (current_branch or "").strip()
under_branches = is_path_under_branches(workspace, root)
if not under_branches:
if workspace == root:
reasons.append(
"author mutation blocked: workspace is the stable control checkout; "
"create or switch to a session-owned worktree under branches/"
)
else:
reasons.append(
f"author mutation blocked: workspace '{workspace}' is not under "
f"'{root}/branches/'; create a branches/<task> worktree first"
)
if not under_branches and workspace == root and branch and branch not in bases:
reasons.append(
f"control checkout drift: branch '{branch}' is not a stable base "
f"branch ({'/'.join(sorted(bases))})"
)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"project_root": root,
"workspace_path": workspace,
"under_branches": is_path_under_branches(workspace, root),
"current_branch": branch or None,
}
def format_author_mutation_worktree_error(assessment: dict) -> str:
"""Single RuntimeError message for MCP preflight gates."""
workspace = assessment.get("workspace_path") or "(unknown)"
root = assessment.get("project_root") or "(unknown)"
reasons = "; ".join(assessment.get("reasons") or ["unknown branches-only violation"])
return (
f"Branches-only mutation guard (#274): {reasons}. "
f"project root: {root}; workspace: {workspace}. "
"Create a session-owned worktree under branches/ before mutating."
)
+234
View File
@@ -0,0 +1,234 @@
"""Hard-stop terminal mode after reviewer capability denial (#197)."""
from __future__ import annotations
import re
TERMINAL_REPORT_HEADING = (
"Cannot perform reviewer task under current profile. "
"No reviewer mutations performed."
)
REVIEWER_CAPABILITY_TASKS = frozenset({
"review_pr",
"merge_pr",
"blind_pr_queue_review",
"pr_queue_cleanup",
"pr-queue-cleanup",
"request_changes_pr",
"approve_pr",
})
BLOCKED_QUEUE_TOOLS = frozenset({
"list_prs",
"check_pr_eligibility",
"view_pr",
"submit_pr_review",
"dry_run_pr_review",
"merge_pr",
"review_pr",
})
_session_terminal: dict | None = None
def enter_from_capability_result(capability: dict) -> dict | None:
"""Enter terminal mode when a reviewer/merge task is denied."""
global _session_terminal
task = (capability or {}).get("requested_task", "")
required_role = (capability or {}).get("required_role_kind")
if not capability.get("stop_required"):
return None
if required_role != "reviewer" and task not in REVIEWER_CAPABILITY_TASKS:
return None
record = {
"active": True,
"requested_task": task,
"required_role_kind": required_role,
"active_profile": capability.get("active_profile"),
"active_identity": capability.get("active_identity"),
"stop_required": True,
"exact_safe_next_action": capability.get("exact_safe_next_action"),
"terminal_message": TERMINAL_REPORT_HEADING,
}
_session_terminal = record
return dict(record)
def _is_reviewer_denial(capability: dict) -> bool:
task = (capability or {}).get("requested_task", "")
required_role = (capability or {}).get("required_role_kind")
return (
required_role == "reviewer"
or task in REVIEWER_CAPABILITY_TASKS
)
def sync_from_capability_result(capability: dict) -> dict | None:
"""Enter or clear terminal mode from a capability resolution (#238).
Reviewer denials activate terminal mode for the denied operation only.
A later allowed task route clears stale denial state so author read-only
tools (e.g. ``list_prs``) are not permanently blocked.
"""
if (capability or {}).get("stop_required") and _is_reviewer_denial(capability):
return enter_from_capability_result(capability)
clear()
return None
def enter_from_route_result(route: dict) -> dict | None:
"""Enter terminal mode from a role router wrong_role_stop (#206 compat)."""
if (route or {}).get("route_result") != "wrong_role_stop":
return None
if route.get("required_role") != "reviewer":
return None
return enter_from_capability_result({
"requested_task": route.get("task_type"),
"required_role_kind": "reviewer",
"stop_required": True,
"active_profile": route.get("active_profile"),
"active_identity": None,
"exact_safe_next_action": route.get("message"),
})
def is_active() -> bool:
return bool(_session_terminal and _session_terminal.get("active"))
def active_record() -> dict | None:
if not is_active():
return None
return dict(_session_terminal)
def clear():
global _session_terminal
_session_terminal = None
def check_reviewer_queue_tool(tool_name: str) -> tuple[bool, list[str]]:
"""Return (allowed, reasons). False when terminal mode blocks queue work."""
if not is_active():
return True, []
name = (tool_name or "").strip().lower().removeprefix("gitea_")
if name in BLOCKED_QUEUE_TOOLS:
denied_task = (_session_terminal or {}).get("requested_task") or "unknown"
return False, [
TERMINAL_REPORT_HEADING,
f"Reviewer queue tool '{tool_name}' is blocked by the current "
f"capability denial for task '{denied_task}' (fail closed).",
"Resolve or route an allowed author task to clear stale denial "
"state, or relaunch a reviewer MCP namespace for reviewer work.",
]
return True, []
def validate_eligibility_wording(text: str) -> tuple[bool, list[str]]:
"""Reject session-based eligibility reasoning (#197)."""
lower = (text or "").lower()
violations = []
if "not authored by this session" in lower:
violations.append(
"eligibility must use authenticated account identity, not "
"'this session' wording"
)
if re.search(r"not (?:self-)?authored by (?:the )?session", lower):
violations.append("session-based eligibility reasoning is invalid")
return (len(violations) == 0), violations
def assess_capability_stop_report(
report_text: str,
*,
trust_gate_status: str | None = None,
capability_denied: bool = True,
) -> dict:
"""Validate final report purity after reviewer capability denial."""
text = report_text or ""
lower = text.lower()
violations = []
if capability_denied and TERMINAL_REPORT_HEADING.lower() not in lower:
violations.append("missing required terminal report heading")
forbidden_patterns = [
("pr selection", re.compile(
r"selected pr|pr #\d+ (?:to review|selected)|eligible pr|"
r"next pr to review", re.I)),
("sibling repo inventory", re.compile(
r"sibling repo|other repo|mcp-control-plane|gitea-tools and", re.I)),
("author fallback", re.compile(
r"rebase conflicted|author-side fallback|have me rebase|"
r"implement the fix|push a branch|open a pr for", re.I)),
("invalid session eligibility", re.compile(
r"not authored by this session", re.I)),
]
for label, pattern in forbidden_patterns:
if pattern.search(text):
violations.append(f"forbidden after hard stop: {label}")
empty_queue_patterns = re.compile(
r"\b0 open pr|\bno open pr|\bno eligible pr|\bempty (?:review )?queue|"
r"inventory empty",
re.I,
)
parsed_status = None
for line in text.splitlines():
if "pr_inventory_trust_gate.status:" in line.lower():
parsed_status = line.split(":", 1)[1].strip()
break
effective_status = trust_gate_status or parsed_status
if empty_queue_patterns.search(text):
if effective_status != "trusted_empty":
violations.append(
"empty-queue claim after capability stop without "
"pr_inventory_trust_gate.status == trusted_empty"
)
ok, elig_violations = validate_eligibility_wording(text)
violations.extend(elig_violations)
if violations:
return {
"pure": False,
"downgraded": True,
"violations": violations,
"reasons": violations,
}
return {
"pure": True,
"downgraded": False,
"violations": [],
"reasons": [],
}
def build_terminal_report(capability: dict) -> dict:
"""Minimal allowed report fields after hard stop."""
return {
"terminal_mode": True,
"heading": TERMINAL_REPORT_HEADING,
"authenticated_profile": capability.get("active_profile"),
"authenticated_identity": capability.get("active_identity"),
"denied_task": capability.get("requested_task"),
"required_role_kind": capability.get("required_role_kind"),
"stop_required": capability.get("stop_required"),
"required_action": capability.get("exact_safe_next_action"),
"mutations_performed": False,
"allowed_sections": [
"authenticated identity/profile",
"denied capability result",
"reason task cannot proceed",
"required reviewer profile/identity",
"mutation confirmation (none)",
],
"forbidden_sections": [
"PR selection",
"sibling-repo queue recommendations",
"author-side fallback suggestions",
"empty-queue claims without trusted_empty",
"session-based eligibility wording",
],
}
@@ -22,13 +22,13 @@ Strictly read-only, per ADR-0001:
tools — never one dual-credential server).
- **This server never holds Gitea write credentials.**
## 2. Boundary placement (namespace pending)
## 2. Boundary placement
These tools belong to the GlitchTip observability boundary of the MCP Control
Plane`glitchtip-mcp` (ADR-0001's recommendation), `observability-mcp`, or
folded into `ops-mcp`. **ADR-0001 open owner decision #2 picks the name; this
design does not assume it.** Tool names below use the `glitchtip_` prefix for
readability and rename mechanically with the decision.
Plane. The canonical MCP server name is `glitchtip-mcp`; client registration
and reload instructions live in
[`../mcp-client-registration.md`](../mcp-client-registration.md). Tool names
below use the `glitchtip_` prefix.
Fixed regardless of the name (per `tool-boundaries.md`,
`credential-isolation.md`):
@@ -160,11 +160,13 @@ Mocked-GlitchTip unit tests only, per `docs/developer-testing-guidelines.md`:
## 10. Implementation-readiness checklist
Ready to implement once:
Ready to operate once:
1. ADR-0001 owner decision #2 (namespace/placement) is made — mechanical
rename of the `glitchtip_` prefix if needed.
2. ADR-0001 owner decision #1 (repo home) is made.
1. The MCP client registers the external server under the exact `glitchtip-mcp`
name and is reconnected/reloaded.
2. Tool discoverability proves the expected `glitchtip_*` read tools are
visible; if the server is enabled but exposes no usable tools, report
`SKIPPED` and stop.
3. #76 profile schema exists (or a minimal `glitchtip-readonly` profile is
hand-rolled to the same rules).
4. A pinned GlitchTip version is chosen for API-subset testing (§3).
@@ -1,21 +1,34 @@
# GlitchTip-to-Gitea Issue Filing Workflow Design
# GlitchTip-to-Gitea Issue Filing Workflow Contract
- **Status:** Design (no implementation in this repo)
- **Issue:** #74 (parent umbrella: #75)
- **Status:** Contract for the library-only implementation in
`Scaled-Tech-Consulting/mcp-control-plane` (#57)
- **Issue:** #153 (supersedes #74/#78 as the Gitea-Tools tracker)
- **Related:** #78 (deduplication design, child of #74)
- **Date:** 2026-07-02
- **Date:** 2026-07-07
## 1. Boundary and Orchestration
* **GlitchTip-to-Gitea filing is NOT a GlitchTip MCP capability.** The `glitchtip-mcp` boundary remains strictly read-only per ADR-0001.
* The filing capability lives in an **orchestrator / runbook / release workflow**. It **composes** separate GlitchTip **read** tools (from the `glitchtip-mcp` server) and Gitea **issue** tools (from the `gitea-mcp` server).
* The filing capability lives in a **library-only orchestrator** in
`mcp-control-plane` and is not exposed through `glitchtip-mcp`.
* The orchestrator composes the real GlitchTip read path with Gitea
issue-write tools. It must not use mocked GlitchTip issue data in production
filing paths.
* The orchestrator **must not centralize credentials** into a single server. The GlitchTip MCP holds only GlitchTip tokens, and the Gitea MCP holds only Gitea tokens.
* If a future MCP surface exposes filing, it must be a separate write-boundary
server/profile with explicit Gitea issue-write permission and the same audit
gates. It must not be added to the read-only `glitchtip-mcp` surface.
## 2. Invocation and Safety
* **Explicit invocation only:** There is no automatic, unsupervised filing in phase 1. A human or an explicitly-triggered automation must initiate the workflow.
* **Dry-run / Preview required:** The orchestrator must present a preview of the drafted Gitea issue (title, body, labels) and obtain explicit confirmation before calling the Gitea mutation tool to file the issue.
* **Gitea Profile Checks & Audit Logging:** The actual Gitea issue creation relies on `gitea-mcp`, and therefore inherently subjects the mutation to Gitea profile checks and audit logging as standard.
* **Gitea Profile Checks & Audit Logging:** The actual Gitea issue creation
relies on `gitea-mcp`, and therefore must pass Gitea profile checks and
fail-closed mutation audit before create/link/comment actions.
* **Deduplication before create:** The orchestrator must run the
GlitchTip-to-Gitea dedup/linking logic before any Gitea create action. Create,
link, and skip decisions must be represented in tests.
## 3. Gitea Issue Format
@@ -51,6 +64,23 @@ To prevent PII or secret leakage into Gitea, the orchestrator and the underlying
The principle is: **"Link, don't dump"**. The generated issue acts as an alert/pointer, while the raw context remains protected inside GlitchTip.
## 5. Deduplication and Linking (Deferred)
## 5. Deduplication and Linking
Deduplication logic (e.g. searching existing Gitea issues, managing GlitchTip issue IDs, and race condition handling) is specifically handled by **Issue #78** and will augment this design.
Deduplication logic (e.g. searching existing Gitea issues, managing GlitchTip
issue IDs, and race condition handling) is integrated into the library-only
filing orchestrator in `mcp-control-plane`. The orchestrator must reuse that
dedup/linking path instead of duplicating or bypassing it.
## 6. Gitea-Tools Acceptance Contract for #153
Gitea-Tools does not host the filing implementation. This repository owns the
operator-facing contract:
* `glitchtip-mcp` remains read-only and exposes only GlitchTip inspection tools.
* Filing is implemented in `mcp-control-plane`, not in this Gitea MCP runtime.
* Filing uses the real GlitchTip read path, not mocked issue data.
* Dedup runs before any Gitea create action.
* Create/link/skip decisions and audit failure are covered by orchestrator tests.
* Mutation audit fails closed before any Gitea create, link, or comment mutation.
* Any future MCP exposure requires a separate write-boundary server/profile and
must not add Gitea write credentials to `glitchtip-mcp`.
@@ -5,7 +5,9 @@
- **Related:** #77 (repo/branch/PR → job mapping, designed separately)
- **Date:** 2026-07-02
Note on naming: This design used historical `jenkins-readonly` skill name in Gitea-Tools. Actual package/server is `jenkins-mcp` (see mcp-control-plane registration in #55). The server boundary now contains gated trigger (see #56), but read tools remain as designed.
Note on naming: This design used historical `jenkins-readonly` skill name in Gitea-Tools. Actual package/server is `jenkins-mcp` (see mcp-control-plane registration in #55). The read server boundary remains read-only; gated triggers live on the separate `jenkins-write-mcp` / `jenkins_mcp.write_server` boundary (see #56 / #152).
Client registration and reload instructions live in
[`../mcp-client-registration.md`](../mcp-client-registration.md).
## 1. Purpose and scope
@@ -16,7 +18,9 @@ detail (build URL, number, timing, result) to report or investigate.
Phase 1 is **primarily read-only**, per ADR-0001
([`adr-0001-mcp-control-plane-boundaries.md`](adr-0001-mcp-control-plane-boundaries.md)):
- Build triggers are gated behind dedicated profile + exact confirmation (landed in #4, boundary correction in #56).
- Build triggers are outside this read-only surface and require the separate
`jenkins-write-mcp` boundary, a dedicated profile, exact confirmation, and
fail-closed mutation audit (landed in #4, boundary correction in #56 / #152).
- **Excluded: deploy triggers.**
- **Excluded: parameterized job launches.**
- Excluded: job creation/deletion/config changes, queue manipulation, node
@@ -107,7 +111,7 @@ by #76):
`forbidden_operations: ["jenkins.build.trigger", "jenkins.deploy", "jenkins.job.configure"]`
as belt-and-braces even though no mutating tool exists.
- Missing URL/user/token/profile ⇒ **fail closed** with a clear message.
- Since every tool is read-only, no confirmation gates are needed — but
- Since every tool on `jenkins-mcp` is read-only, no confirmation gates are needed — but
identity (`jenkins_whoami`) must still work so workflows can prove which
Jenkins account they act as.
@@ -141,12 +145,16 @@ repo's conventions (`docs/developer-testing-guidelines.md`):
## 10. Implementation-readiness checklist
Ready to implement in `jenkins-mcp` once:
Ready to operate through `jenkins-mcp` once:
1. ADR-0001 owner decision #1 (where `jenkins-mcp` lives) is made.
2. #76 profile schema exists (or a minimal `jenkins-readonly` profile is
1. The MCP client registers the external server under the exact `jenkins-mcp`
name and is reconnected/reloaded.
2. Tool discoverability proves the expected `jenkins_*` read tools are visible;
if the server is enabled but exposes no usable tools, report `SKIPPED` and
stop.
3. #76 profile schema exists (or a minimal `jenkins-readonly` profile is
hand-rolled to the same rules).
3. #77 mapping design is accepted (or tools ship path-addressed only, mapping
4. #77 mapping design is accepted (or tools ship path-addressed only, mapping
deferred).
Explicitly **not** unlocked by this document: build triggers, deploys,
+1
View File
@@ -23,6 +23,7 @@ launched with exactly one static execution profile:
|-----------------------------|----------------|-------------|
| `gitea-author` | an author profile | implement issues, push branches, open PRs, comment |
| `gitea-reviewer` | a reviewer profile | review, approve/request changes, merge |
| `gitea-reconciler` | a reconciler profile | close already-landed open PRs after ancestry proof (#304 profile; #310 close tool) |
Properties:
+36
View File
@@ -226,6 +226,31 @@ explicit operator-directed closure of a contaminated PR). If `close_pr` ever
resolves as unknown, agents must fail closed rather than fall back to the
edit path.
## Reconciler profile for already-landed open PRs (#304 / #310)
Normal author and reviewer profiles must not gain broad `gitea.pr.close`
authority. Already-landed open PRs (head SHA is an ancestor of the target
branch) need a dedicated reconciler profile such as `prgs-reconciler` with a
narrow operation set:
- `gitea.read`
- `gitea.pr.comment`
- `gitea.issue.comment`
- `gitea.issue.close`
- `gitea.pr.close`
Forbidden on reconciler profiles: `gitea.pr.approve`, `gitea.pr.merge`,
`gitea.pr.review`, `gitea.pr.create`, `gitea.branch.push`, and
`gitea.repo.commit`.
Launch a static `gitea-reconciler` MCP namespace with
`GITEA_MCP_PROFILE=prgs-reconciler`. Profile shape is validated by
`reconciler_profile.assess_reconciler_profile` (#304). Use the
`gitea_reconcile_already_landed_pr` tool (#310). The resolver task is
`reconcile_already_landed_pr`. PR close is allowed only after live PR fetch,
fresh target-branch fetch, recorded target SHA, and ancestor proof. PRs whose
heads are not already landed cannot be closed through this path.
## Identity and fail-closed rules
Before **any** mutating action, a workflow must know both:
@@ -298,6 +323,17 @@ When dynamic profile switching is enabled and a profile is activated via `gitea_
2. Call `gitea_whoami` with the target remote to prove and verify the fresh Gitea authenticated identity.
This guarantees the active profile operations align with the actual Gitea authenticated user credential.
## Gitea MCP Runtime Isolation and Worktree Safety
To ensure high availability and prevent broken feature worktrees from disabling essential security/identity controls, the Gitea MCP server implements runtime isolation:
- **Startup Conflict Check:** The MCP server (`mcp_server.py`) acts as a conflict-free loader. On startup, it scans all Python files in the directory for unresolved git merge conflicts (`<<<<<<<`, `=======`, `>>>>>>>`). If any are found, it prints an `infra_stop` message and exits immediately.
- **Workflow Guard:** Before starting reviewer work, task routing checks if the MCP runtime source is mid-merge (by checking for `.git/MERGE_HEAD` or conflict markers). If dirty, it returns `infra_stop` (never `wrong_role_stop` or empty queue) to prevent unsafe mutations.
- **Recovery Instructions:** To recover from an `infra_stop` state:
1. Resolve all merge conflicts in the local repository or abort the merge (`git merge --abort` / `git rebase --abort`).
2. Restart the Gitea MCP server process.
3. Retry the task.
## Relationship to roadmap issues
This document defines the **model only**. Related work is tracked separately
+325
View File
@@ -30,6 +30,11 @@ audit logging). See [Related documents](#related-documents).
> the standard rules; operator prompts still control task-specific scope.
> See issue #129 for the skill registry design.
Jenkins and GlitchTip workflows use separate MCP servers, not this Gitea MCP
runtime. Register them as `jenkins-mcp` and `glitchtip-mcp`, reconnect or
reload the client, and verify visible tools before claiming either integration
is usable. See [`mcp-client-registration.md`](mcp-client-registration.md).
For cross-project use, copy the portable workflow skill at
[`../skills/llm-project-workflow/SKILL.md`](../skills/llm-project-workflow/SKILL.md).
It extracts the issue-first, isolated-worktree, no-self-review, profile-safety,
@@ -164,6 +169,25 @@ To avoid the bottleneck of relaunching/restarting the MCP server to switch betwe
* `mcp__gitea-reviewer__*` (for reviewing PRs, approving, requesting changes, merging)
* **Trust Model:** Separate tokens remain separate in the keychain/environment. Each instance operates under its own `GITEA_MCP_PROFILE` and enforces its own `allowed_operations`. A runtime `whoami` identity check is still performed independently, and self-review/self-merge checks remain strictly mandatory. The dual-server pattern is a operational convenience and never a security bypass.
* **Reviewer-Identity PR Creation Deadlock:** Reviewer/merge identities must not create PRs or push branches. Doing so makes the reviewer identity the PR author in Gitea, blocking subsequent independent review and causing a review deadlock. Normally, PRs must be created by the author/work identity (`gitea-author`), leaving the reviewer identity (`gitea-reviewer`) clean and available for independent review and merge.
* **Reconciler namespace (#310):** Register a third static instance for
already-landed PR cleanup when review queues block on open PRs whose heads
already landed on `master`:
```json
"gitea-reconciler": {
"command": "/path/to/Gitea-Tools/venv/bin/python3",
"args": ["/path/to/Gitea-Tools/mcp_server.py"],
"env": {
"GITEA_MCP_CONFIG": "/path/to/.config/gitea-tools/profiles.json",
"GITEA_MCP_PROFILE": "prgs-reconciler"
}
}
```
The reconciler profile grants `gitea.pr.close` only for
`gitea_reconcile_already_landed_pr` after ancestry proof — not for normal
review or author workflows.
* **Fallback:** If the dual-profile MCP launcher pattern is not supported or configured in the client, the LLM must relaunch or restart the client/MCP with the correct profile environment variable before claiming or working on any tasks.
## Setup runbook — interactive menu
@@ -224,6 +248,195 @@ Legacy environment-only setups keep working unchanged until migrated.
Each runbook names the **profile role** it runs under, the steps, and a safe
prompt. Confirm the active profile first (`gitea_get_profile` / `gitea_whoami`).
## Work Selection Rule for LLMs
Before starting any issue or PR work, acquire or verify a work lease. Do not
begin coding, reviewing, fixing, branching, committing, pushing, commenting,
or creating a PR until you prove the target is not already being worked.
Required checks:
1. List open PRs.
2. Search for PRs linked to the target issue.
3. Search local and remote branches for the issue number.
4. Search registered worktrees for the issue branch.
5. Check dirty worktrees.
6. Check active leases or recent handoffs.
7. Check whether the issue was already completed by a merged PR.
If another active LLM/session owns the lease, stop. Allowed responses:
continue as the lease owner; review the existing PR if reviewer capability
allows; produce a handoff; request takeover after lease expiry; stop with
"work already claimed."
Never create a parallel branch or PR for the same issue unless the old branch
is proven abandoned and the takeover is recorded.
Gitea-Tools lease gates: `gitea_lock_issue` (fail-closed before author
mutations), `status:in-progress`, and claim comments. `gitea_lock_issue`
records an `author_issue_work` lease in a keyed lock file under
`GITEA_ISSUE_LOCK_DIR` (default `~/.cache/gitea-tools/issue-locks`), one file
per `remote` + `org` + `repo` + `issue_number`. The current MCP session binds
its active lock through a per-process pointer so concurrent repos/issues never
share one overwrite-prone slot (#443).
Each lock payload includes issue number, optional PR number, branch, worktree
path, claimant identity/profile, created timestamp, expiry timestamp, and last
heartbeat timestamp. An active same-issue/same-operation lease blocks duplicate
work. An expired lease still blocks takeover until a recovery review records why
the prior work is abandoned, completed, or unsafe to continue.
**Stacked PRs (#484).** By default the lock worktree must be base-equivalent to
`master`/`main`/`dev` — ordinary work is unchanged. A *stacked* PR (deliberately
based on another unmerged PR's branch) is an explicit, opt-in path: pass
`stacked_base_branch` **and** `stacked_base_pr` to `gitea_lock_issue`. The lock
fails closed unless that branch is owned by a live **open** PR whose number
matches `stacked_base_pr`, so arbitrary or stale branches cannot be used as
bases. When approved, the lock payload records
`approved_stacked_base = {branch, pr_number, verified_open}` and the worktree may
be base-equivalent to that branch instead of master. `gitea_create_pr` then
allows `base = <that branch>` only when it matches the recorded approval, the
dependency PR is **still open**, and the PR body documents the stack:
- `Stacked on PR #<X> / issue #<Y>`
- `Base branch: <feature-branch>`
- `Head branch: <this-issue-branch>`
- `Do not merge before PR #<X>` (merge ordering)
- retarget/rebase to `master` after the dependency lands, if required
Stacked support never bypasses the issue lock — the base is recorded *on* the
lock and re-verified at PR time. A merged/closed dependency base fails closed;
retarget onto `master` or re-lock against a live base.
**Do not manually seed `/tmp/gitea_issue_lock.json` or any lock file as a normal
recovery path.** That global slot is deprecated and can clobber unrelated live
leases (#438). After an MCP restart, call `gitea_lock_issue` again — own-branch
adoption rebinds the session when the issue's exact branch already exists (#442).
`gitea_create_pr` resolves the durable keyed lock by session pointer or by
matching `head` branch without unsafe manual seeding.
**Issue-lock recovery (#447):** Do not manually seed, restore, or delete
`/tmp/gitea_issue_lock.json` as a normal recovery path. That file is global
shared state and manual writes can clobber another session's live lease. Use
`sanctioned recovery` instead:
1. `gitea_lock_issue` on a clean `branches/` worktree (normal path).
2. Own-branch adoption via #442 when the issue's exact branch is already pushed.
3. Operator override only when explicitly authorized — record
`External-state mutations` and `operator override proof` in the final report.
**Adoption proof in the live lock response (#477):** when `gitea_lock_issue`
adopts an existing own branch, the response carries an `adoption` block with
citable fields — `adoption_decision` (`ADOPT`), `adopted` (`true`),
`adopted_branch`, `adopted_branch_head`, `matcher_summary` (boundary-safe reason
the branch qualified), `competing_branch_check`, and `safe_next_action`. A normal
lock instead returns an `adoption_check` block with `adoption_decision`
(`NO_MATCH`) and `adopted: false`, so a non-adoption response can never be misread
as claiming adoption. Recovery reports should quote the live lock response
`adoption`/`adoption_check` block directly instead of inferring adoption from
separate offline checks.
`gitea_create_pr` rejects lock files that lack sanctioned `lock_provenance`
metadata. Final-report validation blocks handoffs that hide lock read/write/delete
under `External-state mutations: none` or mix author PR creation with reviewer
approval in one run. See also #438 (global lock redesign).
Remote branches matching the issue number are also treated as active work unless
the recovery review proves the branch is abandoned or superseded. Never delete
or clean up a branch when it has an active lease, dirty worktree, open PR, or is
the only copy of unmerged work.
Full portable wording:
[`skills/llm-project-workflow/SKILL.md`](../skills/llm-project-workflow/SKILL.md).
## Global LLM Worktree Rule
The main project checkout is a stable control checkout. It must stay on the
configured stable branch: `master`, `main`, or `dev`.
All LLM task work must happen inside the project's `branches/` directory.
Before any mutation, prove:
1. current project root
2. current working directory
3. current branch
4. stable branch for the main checkout
5. session-owned worktree path under `branches/`
If `cwd` is not inside `branches/`, stop. Do not edit, create, delete, format,
test-write, commit, merge, rebase, checkout task branches, resolve conflicts,
or run cleanup.
There are no exceptions for small fixes, docs, tests, cleanup, PR review fixes,
conflict resolution, or emergencies.
The main checkout may only be used for read-only inspection, fetching,
stable-branch update after merged PRs, creating `branches/` worktrees, or
explicit control-checkout repair.
Portable wording: [`skills/llm-project-workflow/SKILL.md`](../skills/llm-project-workflow/SKILL.md).
## Shell Spawn Hard-Stop Rule
Symptom: a shell tool call returns `exit_code: -1` with empty stdout/stderr.
That is an executor spawn failure, not a command failure — the command never
ran, and retrying the identical call cannot succeed.
Required behavior (fail closed, issue #258):
1. **Probe once.** On the first spawn failure, run one trivial probe
(`echo ok` or `pwd`). If the probe also returns `exit_code: -1`, mark
shell unavailable for the session.
2. **Hard-stop at two.** After two consecutive spawn failures, stop all
further shell tool use for the session; never retry the same failing
spawn. A hundred retries produce a hundred identical failures (session
`019f382e`: 100+ tool calls stalled on a trivial encode-and-commit task).
3. **Emit a recovery report.** The report must direct the operator to:
- restart the session,
- kill hung background terminals (a hung test runner holding the
executor is a known contributor),
- prefer MCP-native paths for remaining mutations (for example
`gitea_commit_files` under `gitea.repo.commit`) instead of shell.
4. **No improvised fallbacks.** Shell unavailability never authorizes
WebFetch/browser/manual-encoding workarounds (see #260). No shell means
stop-and-report.
Doc-contract tests: `tests/test_shell_spawn_hard_stop_docs.py`.
## Subagent Tool-Budget Guardrails
General-purpose subagents tasked with **deterministic MCP work** (for example a
single `gitea_commit_files` call) have expanded to 100122 tool calls,
WebFetch/Playwright fallbacks, and throwaway helper-script generation instead of
calling the native MCP tool once (observed during #152 closure, issue #259).
**Default budgets** (fail closed when exceeded):
| Task class | Max tool calls | Max wall time |
|------------|----------------|---------------|
| Single-step MCP mutation (`commit_files`, `create_pr`, `lock_issue`) | 15 | 5 minutes |
| Review / merge queue inspection | 40 | 15 minutes |
| Exploration / codebase search (non-mutating) | 60 | 20 minutes |
**Required behavior:**
1. **Main session first.** When the active author profile allows
`gitea.repo.commit` and `gitea_commit_files` is visible, the main session
must call it directly — do not delegate commit authority to a subagent
(see #260).
2. **Native MCP before fallback.** After a shell spawn failure (#258), attempt
the native MCP tool once before any alternate path. Shell unavailability
never authorizes WebFetch, Playwright, or manual base64 encoding.
3. **No retry spirals.** Never resume a failed subagent into a larger retry
loop or spawn a second subagent for the same deterministic step. Stop and
emit a recovery report instead.
4. **Forbidden detours** when `gitea_commit_files` is available: WebFetch,
Playwright/browser automation, manual LLM-generated base64, and ad-hoc
`_encode_*` / `_emit_*` helper scripts left in the repo.
Doc-contract tests: `tests/test_subagent_tool_budget_docs.py`.
## Branch worktree isolation
All LLM implementation and review work happens in an isolated branch worktree
@@ -255,6 +468,30 @@ may edit another issue's branch folder unless explicitly assigned to that issue.
No LLM may clean another issue's branch folder unless the PR is merged or closed
and cleanup is explicitly part of the task.
## Agent temp artifact cleanup (#261)
Failed or aborted MCP commit attempts sometimes leave throwaway helper scripts in
the **repository root**. These are not part of any issue scope and pollute
`git status`, which can break `gitea_lock_issue` and preflight checks.
**Patterns (repo root only, untracked):**
- `_encode_*.py` — base64 payload encoders
- `_emit_*.py` — commit payload emitters
- `_inline_*.py` — inline encoding helpers
**Required cleanup (after MCP commit completes or aborts):**
1. Delete any matching files at the repo root (`rm ./_encode_*.py` etc.).
2. Confirm `git status` is clean on the orchestration checkout before
`gitea_lock_issue`.
3. Prefer native `gitea_commit_files` / gated commit paths — do not leave shell
encoding fallbacks behind.
Root-level matches are listed in `.gitignore` so they never get committed.
`gitea_get_runtime_context` and `gitea_lock_issue` surface **warnings** (not
hard blocks) when these artifacts are still present.
Implementation work and review work must use separate branch folders. For
example, an implementation branch might live under
`branches/fix-issue-123-example`, while a review branch for the resulting PR
@@ -309,6 +546,36 @@ git branch -d fix/issue-123-example
All three helpers accept `--dry-run` to print the exact commands/paths without
touching anything.
### MCP-native commit path (#260)
When the active author profile allows **`gitea.repo.commit`** and
**`gitea_commit_files`** is visible in the client, that is the **only** approved
path for committing files to the tracked repository. Do not improvise alternate
encoding or transport when MCP commit is available.
**Required before commit:**
1. Call `gitea_resolve_task_capability` for `commit_files` or
`gitea_commit_files` and confirm `allowed_in_current_session` is true.
2. Use `gitea_commit_files` with file payloads prepared in the author worktree.
3. Stage only issue-scoped paths; never commit throwaway `_encode_*` /
`_emit_*` / `_inline_*` helpers.
**Explicitly forbidden workarounds** when MCP commit is reachable:
- `WebFetch` / HTTP calls to external decode sites (for example httpbin base64
endpoints)
- Playwright or other browser automation to bypass MCP
- Manual LLM-generated base64 pasted into ad-hoc scripts
- Delegating commit authority to a subagent while the main session has
`gitea.repo.commit` on an author profile
**If shell encoding is unavailable** (spawn failure, hung terminal) **and** MCP
commit cannot run: **stop** with a recovery report. Mention restarting the
session, clearing hung background terminals, switching to MCP-native commit, and
the agent temp artifact cleanup checklist. Do **not** retry shell encoding in a
loop and do **not** substitute WebFetch/Playwright/manual base64.
### Create an issue / child issues
- **Profile:** issue-manager or author (any profile allowed to create issues).
@@ -556,6 +823,45 @@ scripts/release-tag v0.4.0 --notes-file /tmp/release-notes.md
scripts/release-tag v0.4.0 --notes-file /tmp/release-notes.md --push
```
## Namespace workspace binding (#510)
Each MCP namespace resolves its **own** active task workspace. Foreign role
worktree environment variables must not poison another namespace's purity
checks.
| Namespace | Workspace env vars (in priority under `GITEA_ACTIVE_WORKTREE`) | Allowed roots |
|-----------|------------------------------------------------------------------|---------------|
| author | `GITEA_AUTHOR_WORKTREE` | `branches/<task>` worktree only (#274) |
| reviewer | `GITEA_REVIEWER_WORKTREE` | clean `branches/<review>` worktree |
| merger | `GITEA_MERGER_WORKTREE` | clean `branches/<merge>` worktree **or** clean control checkout |
| reconciler | `GITEA_RECONCILER_WORKTREE` | clean `branches/<reconcile>` worktree **or** clean control checkout |
`GITEA_AUTHOR_WORKTREE` is **author-only**. Reviewer, merger, and reconciler
MCP processes ignore it even when it points at a dirty author WIP tree.
### Safe reconnect / rebind procedure
When a mutation blocks on workspace binding:
1. Read the error — it names the **resolved workspace path**, **role
namespace**, and **binding source** (tool arg, env var, or process root).
2. Reconnect or relaunch the correct namespace MCP server from the intended
workspace (or set the role-specific env var before launch).
3. Pass `worktree_path` on reviewer/merger mutation tools when the active
branches/ worktree differs from the MCP process root.
4. **Do not** clean, reset, or discard foreign role worktrees to unblock your
own namespace — that destroys another agent's WIP.
### CTH guidance for workspace binding blockers
When posting a Canonical Thread Handoff after a binding blocker:
- State which namespace was active (author / reviewer / merger / reconciler).
- Quote the resolved workspace path and binding source from the error.
- Name the safe reconnect action (relaunch MCP from `branches/...`, set
`GITEA_*_WORKTREE`, or pass `worktree_path`).
- Explicitly note that foreign worktrees must not be cleaned to unblock.
## Safety notes
- Never place raw tokens or passwords in any LLM MCP config; reference secrets
@@ -574,3 +880,22 @@ scripts/release-tag v0.4.0 --notes-file /tmp/release-notes.md --push
- [`credential-isolation.md`](credential-isolation.md) — credential handling.
- [`release-workflows.md`](release-workflows.md) — release/merge workflow.
- [`../README.md`](../README.md) — canonical config, thin launchers, the menu.
## PR-only queue cleanup mode (#390)
Use `pr-queue-cleanup` mode when the queue holds many open PRs and the goal
is to drain reviews deterministically. One run = one PR = one canonical
review (`skills/llm-project-workflow/workflows/pr-queue-cleanup.md`).
* Cleanup runs are reviewer-role only (`pr_queue_cleanup` resolver task);
author sessions get `wrong_role_stop`.
* Forbidden in cleanup mode: issue claiming, branch creation, implementation
edits, new issue filing, and any second-PR review after a terminal
mutation.
* Terminal chain: stop after `REQUEST_CHANGES`; after `APPROVED` continue
only to same-PR merge with explicit per-PR operator authorization and
passing gates; stop after merge or merge blocker.
* Every run reports the next suggested PR without continuing to it; the next
PR requires a fresh run with fresh identity/capability/inventory proof.
* Use full `review-merge-pr` mode instead when the operator wants a single
targeted review, and `work-issue` mode for any authoring.
+108
View File
@@ -0,0 +1,108 @@
# MCP Client Registration for External Control Plane Servers
Issue #151 fixes the registration and naming contract for the Jenkins and
GlitchTip MCP servers that live outside this Gitea MCP runtime.
## Canonical Server Names
Use these exact MCP server names in clients:
| Server name | Boundary | Default capability |
|---|---|---|
| `jenkins-mcp` | Jenkins CI inspection (read) | Read-only build/job inspection |
| `jenkins-write-mcp` | Jenkins build trigger (write) | Gated `jenkins_trigger_build` only |
| `glitchtip-mcp` | GlitchTip observability inspection | Read-only issue/event inspection |
The write boundary (`jenkins-write-mcp`) is **not** registered by default (#152).
It exposes a single mutating tool and requires operator approval of a dedicated
trigger profile before any client config references it.
Historical names such as `jenkins-readonly` and `glitchtip-readonly` are
descriptive profile labels only. They are not the canonical MCP server names
unless an operator intentionally creates aliases and documents them.
## Registration Pattern
Register each external server as its own MCP entry. Do not add Jenkins or
GlitchTip credentials to the Gitea MCP server. Also, do not add Gitea write
credentials to the GlitchTip server.
```json
{
"mcpServers": {
"jenkins-mcp": {
"command": "/path/to/mcp-control-plane/venv/bin/python3",
"args": ["-m", "jenkins_mcp"],
"env": {
"JENKINS_MCP_CONFIG": "/path/to/mcp-control-plane/profiles.json",
"JENKINS_MCP_PROFILE": "jenkins-readonly"
}
},
"glitchtip-mcp": {
"command": "/path/to/mcp-control-plane/venv/bin/python3",
"args": ["-m", "glitchtip_mcp"],
"env": {
"GLITCHTIP_MCP_CONFIG": "/path/to/mcp-control-plane/profiles.json",
"GLITCHTIP_MCP_PROFILE": "glitchtip-readonly"
}
}
}
}
```
Client-specific wrappers may differ, but the server names, trust boundaries,
and profile separation must remain the same. After adding or changing either
entry, reconnect or reload the MCP client before claiming the tools are usable.
## Discoverability Validation
Before using either server in a task, prove the expected tools are visible in
the client. It is not enough for the config entry to exist.
Expected Jenkins read tools (`jenkins-mcp` only):
- `jenkins_whoami`
- `jenkins_list_jobs`
- `jenkins_latest_build`
- `jenkins_build_status`
- `jenkins_get_build`
`jenkins_trigger_build` must **not** appear on `jenkins-mcp`. When an operator
explicitly enables the write boundary, the only expected tool on
`jenkins-write-mcp` is `jenkins_trigger_build`.
Expected GlitchTip tools:
- `glitchtip_whoami`
- `glitchtip_list_projects`
- `glitchtip_list_unresolved`
- `glitchtip_get_issue`
- `glitchtip_recent_events`
- `glitchtip_search`
If a client reports the server as enabled but exposes no usable tools, report
`SKIPPED: server enabled but no usable tools visible`, then stop. Do not fall
back to shell commands, raw service APIs, or unrelated MCP servers.
## Boundary Rules
- `jenkins-mcp` read profiles must not expose build trigger tools.
- Build triggers live on the separate `jenkins-write-mcp` server
(`jenkins_mcp.write_server`), not on `jenkins-mcp`.
- Jenkins build triggers require a dedicated trigger profile with
`jenkins.build.trigger` allowed, exact confirmation
(`TRIGGER BUILD <job-path>`), and fail-closed mutation audit.
- Do not register `jenkins-write-mcp` until an operator approves a trigger
profile; no shipped profile carries trigger capability by default.
- `glitchtip-mcp` remains read-only. It must not file or mutate Gitea issues.
- GlitchTip-to-Gitea filing is a separate library-only orchestrator in
`mcp-control-plane` that composes the real GlitchTip read path with Gitea
issue-write tools.
- The filing orchestrator must run GlitchTip/Gitea dedup before any Gitea
create action, and its create/link/skip decisions must be covered by tests.
- The filing orchestrator must fail closed on mutation-audit failure before
any Gitea create, link, or comment mutation.
- If filing is ever MCP-exposed, it must use a separate write-boundary
server/profile. It must not be exposed by `glitchtip-mcp`.
- Gitea credentials never enter Jenkins or GlitchTip runtimes.
- Jenkins and GlitchTip credentials never enter the Gitea MCP runtime.
+25 -2
View File
@@ -21,5 +21,28 @@ Note on naming: Historical design docs used `jenkins-readonly` / `glitchtip-read
## 5. Mutation Gating
Any mutating action (e.g., Gitea issue creation from GlitchTip, or Jenkins builds) must be explicitly allowed by the execution profile.
- **Jenkins build triggers** exist in jenkins-mcp (landed #4) but require dedicated profile/identity and exact confirmation; not on standard reader profiles. See #56 for boundary correction.
- **GlitchTip to Gitea issue filing** is documented as a gated, orchestrated workflow (not in glitchtip-mcp), currently partial (mocked, dedup not wired, audit missing). See #57.
- **Jenkins build triggers** are gated on a separate write boundary
(`jenkins-write-mcp` / `jenkins_mcp.write_server`), not on the read-only
`jenkins-mcp` surface. Triggers require a dedicated profile with
`jenkins.build.trigger`, exact confirmation, and fail-closed mutation audit.
No default profile carries trigger capability (#152 / mcp-control-plane #56).
- **GlitchTip to Gitea issue filing** is a library-only orchestrator in
mcp-control-plane (not on `glitchtip-mcp`). See #153 / mcp-control-plane #57.
## 6. Agent Commit Path (no improvised fallbacks)
When an author execution profile allows **`gitea.repo.commit`** and the
**`gitea_commit_files`** tool is visible, agents must use that MCP path for
repository commits. Fail closed instead of improvising alternate transports.
Forbidden when MCP commit is available:
- WebFetch or other HTTP calls to external base64/decode services
- Playwright or browser automation used to work around MCP commit
- Manual LLM-generated base64 embedded in throwaway scripts as the primary
commit transport
If shell helpers are unavailable and MCP commit cannot run, stop with a recovery
report (restart session, clear hung terminals, use MCP-native commit). See
[`llm-workflow-runbooks.md`](llm-workflow-runbooks.md) § MCP-native commit path
(#260) and agent temp artifact cleanup (#261).
+89
View File
@@ -0,0 +1,89 @@
# Internal web UI — local development (#426)
Read-only MVP skeleton for the MCP Control Plane operator console. Gitea,
MCP capability gates, and `skills/llm-project-workflow/` remain the source of
truth; this UI only provides route stubs and layout.
## Prerequisites
- Python 3.11+ with project dependencies installed (`pip install -r requirements.txt`)
- No secrets in repo, config, or client bundle
## Start the server
From the repository root (or an issue worktree):
```bash
./scripts/run-webui
```
Or directly:
```bash
python3 -m webui
```
Optional environment variables:
| Variable | Default | Purpose |
|----------|---------|---------|
| `WEBUI_HOST` | `127.0.0.1` | Bind address (keep local for MVP) |
| `WEBUI_PORT` | `8765` | Listen port |
## Routes (MVP)
| Path | Description |
|------|-------------|
| `/` | Home / operator overview |
| `/health` | JSON liveness (`status`, `service`, `mode`, `timestamp`) |
| `/queue` | Live PR and issue queue dashboard (#429) |
| `/api/queue` | JSON queue export with pagination metadata |
| `/projects` | Project registry list (#427) |
| `/projects/{id}` | Project detail + onboarding checklist |
| `/api/projects` | JSON registry export |
| `/prompts` | Prompt library with per-prompt copy buttons (#428) |
| `/api/prompts` | JSON prompt export with workflow hashes |
| `/runtime` | Stub — MCP runtime health (#430) |
| `/audit` | Stub — report audit paste (#431) |
| `/worktrees` | Stub — hygiene dashboard (#432) |
| `/leases` | Stub — lease visibility (#433) |
All routes are GET-only. POST/PUT/PATCH/DELETE return `405` with
`read-only-mvp`.
## Project registry (#427)
Versioned registry file: `webui/data/projects.registry.json` (schema version `1`).
Override path with `WEBUI_PROJECT_REGISTRY` when operators keep a machine-local
copy outside git. The registry stores repo identity, remotes, profile names,
workflow/schema path references, and onboarding checklist steps — never tokens
or credentials.
Seed entry: **Gitea-Tools** on `https://gitea.prgs.cc` with `prgs-author`,
`prgs-reviewer`, and `prgs-reconciler` profiles.
## Prompt library (#428)
Prompts are generated at load time from canonical workflow files under
`skills/llm-project-workflow/workflows/`. SHA-256 hashes are computed from
`WEBUI_REPO_ROOT` (defaults to the repository root). Prompt bodies are short
copy/paste starters; canonical workflow files remain the only full policy
source.
## Live queue dashboard (#429)
`/queue` loads open PRs and issues for the default registry project (seed:
**Gitea-Tools** on `https://gitea.prgs.cc`) using existing `gitea_auth` read
credentials. The UI surfaces pagination proof (returned count, pages fetched,
`has_more`, `inventory_complete`) and classification badges (`claimed`,
`blocked`, `in-review`, `duplicate`) when evidence exists.
If credentials are missing or the fetch fails, the page shows an explicit error
instead of an empty queue (fail closed).
## Tests
```bash
pytest tests/test_webui_skeleton.py tests/test_webui_project_registry.py tests/test_webui_prompt_library.py tests/test_webui_queue_dashboard.py -q
```
+15
View File
@@ -0,0 +1,15 @@
# Project History
## Navigation
- [Home](Home.md) | [Operator Guide](Operator-Guide.md) | [Repositories Map](Repositories.md) | [Identity and Profiles](Identity-and-Profiles.md) | [Workflow Model](Workflow.md) | [Safety and Gates](Safety-and-Gates.md) | [MCP Tools Reference](MCP-Tools.md) | [Operator Runbooks](Runbooks.md) | [Open Decisions](Open-Decisions.md) | [Project History](History.md)
## 2026-07-06
- **Wiki bootstrap (#224)** — Added repo-tracked `docs/wiki/` (10 pages), `scripts/sync-gitea-wiki.sh`, PR template gate, and sync safety tests for Gitea-Tools publication.
## Prior milestones
- Gated review/merge path (#16), task capability resolver (#69), issue-write tool gates.
- Canonical JSON execution profiles (#19) and thin MCP launchers.
- Role session router (#206) and review decision lock (#211).
+35
View File
@@ -0,0 +1,35 @@
# Gitea-Tools Project Wiki
Welcome to the version-controlled wiki for the Gitea-Tools MCP server and CLI tooling.
## Navigation
- [Home](Home.md)
- [Operator Guide](Operator-Guide.md)
- [Repositories Map](Repositories.md)
- [Identity and Profiles](Identity-and-Profiles.md)
- [Workflow Model](Workflow.md)
- [Safety and Gates](Safety-and-Gates.md)
- [MCP Tools Reference](MCP-Tools.md)
- [Operator Runbooks](Runbooks.md)
- [Open Decisions](Open-Decisions.md)
- [Project History](History.md)
## Gitea Wiki mirror
These repo-tracked pages are the **source of truth**. The Gitea native Wiki
for this repository is a read-only convenience mirror generated from this
directory with `scripts/sync-gitea-wiki.sh` (dry-run by default; see
[Runbooks](Runbooks.md)). Never edit the Gitea Wiki directly — change the
pages here through a PR, then sync.
## Project overview
Gitea-Tools provides the Gitea MCP server, execution-profile configuration,
identity handling, and CLI helpers used across MCP Control Plane workflows.
It enforces author/reviewer separation, gated review/merge, task-capability
resolution, and fail-closed safety rails in code.
Canonical long-form docs also live under `docs/` in the repository (for example
`docs/gitea-execution-profiles.md` and `docs/llm-workflow-runbooks.md`). The
wiki summarizes operator-facing rules; the repo docs carry implementation detail.
+39
View File
@@ -0,0 +1,39 @@
# Identity and Profiles
## Navigation
- [Home](Home.md) | [Operator Guide](Operator-Guide.md) | [Repositories Map](Repositories.md) | [Identity and Profiles](Identity-and-Profiles.md) | [Workflow Model](Workflow.md) | [Safety and Gates](Safety-and-Gates.md) | [MCP Tools Reference](MCP-Tools.md) | [Operator Runbooks](Runbooks.md) | [Open Decisions](Open-Decisions.md) | [Project History](History.md)
The LLM is not the role — the **MCP execution profile** is the role. Profiles
bind an authenticated Gitea identity to an allowed operation set.
## Reference profiles (prgs)
### Author / implementer
- **Profile:** `prgs-author`
- **Typical identity:** `jcwalker3`
- **Allowed:** branch create/push, PR create, issue comment/create/close, repo commit, read.
- **Forbidden:** PR approve, merge, request_changes.
### Reviewer / merger
- **Profile:** `prgs-reviewer`
- **Typical identity:** `sysadmin`
- **Allowed:** PR review/approve/merge/request_changes, issue comment, read.
- **Forbidden:** branch push, PR create, repo commit.
## Configuration
Profiles are defined in the canonical JSON config (`GITEA_MCP_CONFIG`, typically
`~/.config/gitea-tools/profiles.json`). Launchers are thin: they set
`GITEA_MCP_PROFILE` and point at the config file. Credentials resolve from
keychain or env references — never inline in client configs.
See `docs/gitea-execution-profiles.md` in the repository for the full model.
## Profile switching
Use separate MCP server namespaces (`gitea-tools` author vs `gitea-reviewer`)
or distinct launcher entries. Runtime in-place profile switching is disabled by
default (fail closed).
+34
View File
@@ -0,0 +1,34 @@
# MCP Tools Reference
## Navigation
- [Home](Home.md) | [Operator Guide](Operator-Guide.md) | [Repositories Map](Repositories.md) | [Identity and Profiles](Identity-and-Profiles.md) | [Workflow Model](Workflow.md) | [Safety and Gates](Safety-and-Gates.md) | [MCP Tools Reference](MCP-Tools.md) | [Operator Runbooks](Runbooks.md) | [Open Decisions](Open-Decisions.md) | [Project History](History.md)
## Identity and capability
- `gitea_whoami` — authenticated user and active profile metadata.
- `gitea_get_profile` / `gitea_get_runtime_context` — allowed and forbidden operations.
- `gitea_resolve_task_capability` — required pre-flight for gated mutations.
- `gitea_route_task_session` — role/session router before task execution.
## Author tools
- `gitea_create_issue`, `gitea_create_issue_comment`, `gitea_close_issue`
- `gitea_mark_issue`, `gitea_set_issue_labels`, `gitea_lock_issue`
- `gitea_create_pr`, `gitea_edit_pr`, `gitea_commit_files`
- `gitea_delete_branch`
## Reviewer tools
- `gitea_check_pr_eligibility` — read-only eligibility check.
- `gitea_dry_run_pr_review` — validation-phase review mechanics.
- `gitea_mark_final_review_decision` — mark validation complete.
- `gitea_submit_pr_review` / `gitea_review_pr` — gated live review.
- `gitea_merge_pr` — gated merge (only merge path).
## Read tools
- `gitea_list_prs`, `gitea_view_pr`, `gitea_list_issues`, `gitea_view_issue`
- `gitea_get_file`, `gitea_list_labels`, `gitea_mirror_refs`
See the repository `README.md` for the full tool table and client setup.
+11
View File
@@ -0,0 +1,11 @@
# Open Decisions
## Navigation
- [Home](Home.md) | [Operator Guide](Operator-Guide.md) | [Repositories Map](Repositories.md) | [Identity and Profiles](Identity-and-Profiles.md) | [Workflow Model](Workflow.md) | [Safety and Gates](Safety-and-Gates.md) | [MCP Tools Reference](MCP-Tools.md) | [Operator Runbooks](Runbooks.md) | [Open Decisions](Open-Decisions.md) | [Project History](History.md)
Pending architectural and workflow decisions:
- **Operation-scoped role selection (#228)** — Move from launcher-profile-dependent routing to per-operation profile resolution.
- **Gitea-Tools wiki for dadeschools remote** — This bootstrap covers `prgs`; dadeschools instance wiki parity is undecided.
- **Server-side self-merge block** — Complement tool gates with Gitea branch protection where available.
+32
View File
@@ -0,0 +1,32 @@
# Operator Guide
## Navigation
- [Home](Home.md) | [Operator Guide](Operator-Guide.md) | [Repositories Map](Repositories.md) | [Identity and Profiles](Identity-and-Profiles.md) | [Workflow Model](Workflow.md) | [Safety and Gates](Safety-and-Gates.md) | [MCP Tools Reference](MCP-Tools.md) | [Operator Runbooks](Runbooks.md) | [Open Decisions](Open-Decisions.md) | [Project History](History.md)
Handbook for LLM operators and human developers using the Gitea-Tools MCP server.
## Key rules
1. **Verify identity first** — Run `gitea_whoami` and confirm the active profile before any mutation.
2. **Resolve task capability** — Call `gitea_resolve_task_capability` for the intended task before gated tools.
3. **One unit of work per session** — Implement one claimed issue *or* review/merge one PR; do not mix author and reviewer mutations in one session.
4. **No self-review / no self-merge** — The authenticated Gitea user must not approve or merge a PR they authored.
5. **Follow the gates** — Prompts express intent; MCP tools enforce safety. Never bypass gates via prompt instructions.
6. **Global LLM Worktree Rule** — Main checkout stays on `master`/`main`/`dev`; all mutations happen under `branches/`. Prove project root, `cwd`, branch, stable main-checkout branch, and session worktree path before editing. No exceptions.
## Supported Gitea instances
| Remote | Host | Default org/repo |
|--------|------|------------------|
| `dadeschools` | `gitea.dadeschools.net` | `Contractor / Timesheet` |
| `prgs` | `gitea.prgs.cc` | `Scaled-Tech-Consulting / Timesheet` |
Always pass `remote` explicitly on tool calls. The server default is `dadeschools`; forgetting `remote` on a `prgs` task hits the wrong host.
## New session checklist
1. `gitea_whoami` — confirm authenticated user and profile.
2. `gitea_get_runtime_context` — allowed/forbidden operations for this session.
3. `gitea_resolve_task_capability` — prove the session may perform the planned task.
4. For reviewer work: dry-run validation (`gitea_dry_run_pr_review`) before live review mutations.
+38
View File
@@ -0,0 +1,38 @@
# Repositories Map
## Navigation
- [Home](Home.md) | [Operator Guide](Operator-Guide.md) | [Repositories Map](Repositories.md) | [Identity and Profiles](Identity-and-Profiles.md) | [Workflow Model](Workflow.md) | [Safety and Gates](Safety-and-Gates.md) | [MCP Tools Reference](MCP-Tools.md) | [Operator Runbooks](Runbooks.md) | [Open Decisions](Open-Decisions.md) | [Project History](History.md)
Active MCP Control Plane repositories (authoritative inventory for wiki publication gate #224 / #87):
| Repository | Purpose | Wiki required |
|------------|---------|---------------|
| `Scaled-Tech-Consulting/Gitea-Tools` | Gitea MCP server, execution profiles, workflow tooling | Yes — live Gitea Wiki on Wiki tab |
| `Scaled-Tech-Consulting/mcp-control-plane` | Orchestrators, audit trails, multi-service controller workflows | Yes — live Gitea Wiki on Wiki tab |
Repo-tracked `docs/wiki/` is the source of truth; the Gitea Wiki is a mirror.
Wiki-related issues cannot be closed until the live Wiki is verified — see
[Safety and Gates](Safety-and-Gates.md) and [Runbooks](Runbooks.md).
### Gitea-Tools
- **Repository:** `Scaled-Tech-Consulting/Gitea-Tools`
- **Purpose:** Gitea MCP server, profile configuration, CLI scripts, safety gates.
- **Base branch:** `master`
### mcp-control-plane
- **Repository:** `Scaled-Tech-Consulting/mcp-control-plane`
- **Purpose:** Controller workflows, Jenkins/GlitchTip integrations, audit orchestration.
- **Base branch:** `master`
## Wiki publication status (#224 readiness gate)
| Repository | `docs/wiki/` source | Gitea Wiki published | Proof |
|---|---|---|---|
| `Scaled-Tech-Consulting/Gitea-Tools` | yes (10 pages) | published (verified 2026-07-06) | [Wiki Home](https://gitea.prgs.cc/Scaled-Tech-Consulting/Gitea-Tools/wiki/Home); 10 pages; wiki git log head `d1f0693` |
| `Scaled-Tech-Consulting/mcp-control-plane` | yes (10 pages) | published (verified 2026-07-06) | [Wiki Home](https://gitea.prgs.cc/Scaled-Tech-Consulting/mcp-control-plane/wiki/Home); 10 pages (History, Home, Identity-and-Profiles, MCP-Tools, Open-Decisions, Operator-Guide, Repositories, Runbooks, Safety-and-Gates, Workflow); wiki git log head `ef3dec2` |
Update this table whenever a wiki is published, re-synced, or found stale.
+40
View File
@@ -0,0 +1,40 @@
# Operator Runbooks
## Navigation
- [Home](Home.md) | [Operator Guide](Operator-Guide.md) | [Repositories Map](Repositories.md) | [Identity and Profiles](Identity-and-Profiles.md) | [Workflow Model](Workflow.md) | [Safety and Gates](Safety-and-Gates.md) | [MCP Tools Reference](MCP-Tools.md) | [Operator Runbooks](Runbooks.md) | [Open Decisions](Open-Decisions.md) | [Project History](History.md)
## PR review and merge
1. `gitea_resolve_task_capability(task="review_pr")`
2. `gitea_check_pr_eligibility` for review and merge actions.
3. Validate locally: tests, `py_compile`, `git diff --check`.
4. `gitea_mark_final_review_decision` → approve via `gitea_review_pr`.
5. `gitea_merge_pr` with pinned head SHA and `confirmation="MERGE PR <n>"`.
## Gitea Wiki sync
The Gitea Wiki mirrors `docs/wiki/` (source of truth). After merging wiki changes:
1. Preview (no network):
```bash
scripts/sync-gitea-wiki.sh
```
2. Operator-confirmed push:
```bash
GITEA_WIKI_SYNC_CONFIRM="SYNC WIKI Gitea-Tools" scripts/sync-gitea-wiki.sh --push
```
3. If clone fails on first run, bootstrap Home via Gitea API or UI, then re-run.
The script mirrors only `docs/wiki/*.md`, never deletes wiki pages, and never
prints credentials.
## Wiki Publication Readiness Gate (#224)
Actual Gitea Wiki publication is a **required repo readiness gate**.
1. **Closure prevention** — No wiki issue may close on markdown/helper work alone.
Closing requires live-Wiki proof: Wiki Home link plus page listing or wiki git log.
2. **Reviewer checklist** — PR template requires verifying the repo **Wiki tab**.
3. **Publication authority** — `--push` requires exact `GITEA_WIKI_SYNC_CONFIRM` phrase.
4. **Per-repo status** — Update [Repositories Map](Repositories.md) when published.
+19
View File
@@ -0,0 +1,19 @@
# Safety and Gates
## Navigation
- [Home](Home.md) | [Operator Guide](Operator-Guide.md) | [Repositories Map](Repositories.md) | [Identity and Profiles](Identity-and-Profiles.md) | [Workflow Model](Workflow.md) | [Safety and Gates](Safety-and-Gates.md) | [MCP Tools Reference](MCP-Tools.md) | [Operator Runbooks](Runbooks.md) | [Open Decisions](Open-Decisions.md) | [Project History](History.md)
*Prompts express intent; MCP tools enforce safety.*
## Primary gates
1. **Fail closed** — Unknown tasks, missing capability resolution, or profile mismatches block mutations.
2. **Task capability map**`gitea_resolve_task_capability` must precede gated issue/PR mutations.
3. **Issue lock**`gitea_lock_issue` required before author implementation on a tracked issue.
4. **Head SHA pinning** — Reviews and merges refuse when the PR head moved.
5. **Explicit merge confirmation**`gitea_merge_pr` requires `confirmation="MERGE PR <n>"`.
6. **No self-review / self-merge** — Authenticated user must differ from PR author for approve/merge.
7. **Review decision lock** — Live review mutations require validation-phase dry-run and `gitea_mark_final_review_decision`.
8. **Redaction** — Tokens, passwords, and keychain material never appear in tool output.
9. **Wiki publication (#224)**`docs/wiki/` and the sync helper are prerequisites only. Closing a wiki issue requires live Gitea Wiki proof on the repo Wiki tab. See [Runbooks](Runbooks.md#wiki-publication-readiness-gate-224).
+38
View File
@@ -0,0 +1,38 @@
# Workflow Model
## Navigation
- [Home](Home.md) | [Operator Guide](Operator-Guide.md) | [Repositories Map](Repositories.md) | [Identity and Profiles](Identity-and-Profiles.md) | [Workflow Model](Workflow.md) | [Safety and Gates](Safety-and-Gates.md) | [MCP Tools Reference](MCP-Tools.md) | [Operator Runbooks](Runbooks.md) | [Open Decisions](Open-Decisions.md) | [Project History](History.md)
## Step 1: Review queue first (reviewer profile)
1. `gitea_resolve_task_capability(task="review_pr")`
2. List open PRs; pick the oldest eligible PR (not self-authored, mergeable).
3. Pin head SHA; validate diff scope and run tests locally.
4. `gitea_mark_final_review_decision``gitea_review_pr` / `gitea_submit_pr_review`
5. `gitea_merge_pr` only with `confirmation="MERGE PR <n>"` and pinned head SHA.
## Step 2: Implement issues (author profile)
0. Work Selection Rule — verify a work lease before any mutations (open PRs,
issue-linked PRs, branches, worktrees, dirty worktrees, active leases/
handoffs, merged-PR completion). Stop if another session owns the lease.
`gitea_lock_issue` records an operation-scoped `author_issue_work` lease
with issue, branch, worktree, claimant, created, expiry, and heartbeat
fields; active or expired same-operation leases require recovery review
before takeover.
0b. Global LLM Worktree Rule — main checkout on `master`/`main`/`dev` only;
mutate only from a `branches/` worktree after proving root, cwd, branch,
stable main-checkout branch, and session worktree path (no exceptions).
1. `gitea_resolve_task_capability` for the author task.
2. `gitea_lock_issue` before implementation mutations.
3. Claim with `gitea_mark_issue` / `status:in-progress` label.
4. Branch, implement, test, push, `gitea_create_pr`.
5. Release claim when done; never self-review the PR.
## Step 3: Operator follow-ups
Wiki publication, credential provisioning, and cross-repo mirror operations are
operator-confirmed actions — see [Runbooks](Runbooks.md).
Full Gitea-specific runbooks: `docs/llm-workflow-runbooks.md` in the repository.
File diff suppressed because it is too large Load Diff
+10 -2
View File
@@ -163,12 +163,13 @@ def audit_enabled():
def build_event(*, action, result, remote=None, server=None, repository=None,
issue_number=None, pr_number=None, profile_name=None,
audit_label=None, authenticated_username=None, target_branch=None,
head_sha=None, reason=None, request_metadata=None, now=None):
head_sha=None, reason=None, request_metadata=None, now=None,
mcp_namespace=None, task_role=None, operation=None):
"""Build a redacted, JSON-able audit record for a mutating action."""
ts = now or datetime.datetime.now(datetime.timezone.utc)
if isinstance(ts, datetime.datetime):
ts = ts.isoformat()
return {
event = {
"timestamp": ts,
"action": action,
"action_type": "mutating",
@@ -186,6 +187,13 @@ def build_event(*, action, result, remote=None, server=None, repository=None,
"reason": _redact_str(reason) if reason else reason,
"request_metadata": redact(request_metadata) if request_metadata is not None else None,
}
if mcp_namespace is not None:
event["mcp_namespace"] = mcp_namespace
if task_role is not None:
event["task_role"] = task_role
if operation is not None:
event["operation"] = operation
return event
def write_event(event, path=None):
+37
View File
@@ -420,6 +420,43 @@ def api_get_all(url, auth_header, *, limit=None, page_size=50, max_pages=100,
return results
def api_fetch_page(url, auth_header, *, page=1, limit=50, **kwargs):
"""Fetch one page from a Gitea list endpoint with explicit pagination metadata.
Returns ``(items, pagination)`` where *pagination* includes ``has_more``,
``next_page``, and ``is_final_page`` derived from the returned page length.
"""
page = max(1, int(page))
limit = max(1, min(50, int(limit)))
page_url = _add_query(url, page=page, limit=limit)
data = api_request("GET", page_url, auth_header, **kwargs)
if data is None:
pagination = {
"page": page,
"per_page": limit,
"returned_count": 0,
"has_more": False,
"next_page": None,
"is_final_page": True,
}
return [], pagination
if not isinstance(data, list):
raise RuntimeError(
f"expected a list page from Gitea, got {type(data).__name__}"
)
returned = len(data)
has_more = returned >= limit
pagination = {
"page": page,
"per_page": limit,
"returned_count": returned,
"has_more": has_more,
"next_page": page + 1 if has_more else None,
"is_final_page": not has_more,
}
return data, pagination
def gitea_url(host, path):
"""Build a full URL for *host* and *path*, using http for loopback and https for others."""
if not path.startswith("/"):
+9 -1
View File
@@ -205,7 +205,7 @@ def selected_profile_name():
def is_runtime_switching_enabled(path=None):
"""Check if runtime profile switching is explicitly enabled in config."""
"""Check if runtime profile switching is enabled in config."""
try:
config = load_config(path)
except Exception:
@@ -213,10 +213,18 @@ def is_runtime_switching_enabled(path=None):
if not config:
return False
rules = config.get("rules") or {}
if rules.get("allow_runtime_switching") is False:
return False
if config.get("allow_runtime_switching") is False:
return False
if rules.get("allow_runtime_switching") is True:
return True
if config.get("allow_runtime_switching") is True:
return True
# Default to True if multiple profiles exist in the config
profiles = config.get("profiles") or {}
if len(profiles) > 1:
return True
return False
+7902
View File
File diff suppressed because it is too large Load Diff
+295
View File
@@ -0,0 +1,295 @@
"""Issue claim heartbeat leases and stale-claim reconciliation (#268).
Structured issue-thread comments prove live ownership beyond the
``status:in-progress`` label alone. Queue inventory can classify claims as
active, stale, reclaimable, PR-backed, or phantom.
"""
from __future__ import annotations
import re
from datetime import datetime, timedelta, timezone
from typing import Any
MARKER = "<!-- gitea-issue-claim-heartbeat:v1 -->"
IN_PROGRESS_LABEL = "status:in-progress"
_KIND_CLAIM = "claim"
_KIND_PROGRESS = "progress"
_KIND_CLEANUP = "cleanup"
_FIELD_RE = re.compile(
r"^\s*-\s*([a-z_]+)\s*:\s*(.+?)\s*$",
re.IGNORECASE | re.MULTILINE,
)
_ISSUE_REF_RE = re.compile(r"issue-(\d+)", re.IGNORECASE)
def _parse_timestamp(value: str | None) -> datetime | None:
if not value:
return None
text = value.strip()
if text.endswith("Z"):
text = text[:-1] + "+00:00"
try:
parsed = datetime.fromisoformat(text)
except ValueError:
return None
if parsed.tzinfo is None:
return parsed.replace(tzinfo=timezone.utc)
return parsed
def format_heartbeat_body(
*,
kind: str,
issue_number: int,
branch: str,
phase: str,
profile: str | None = None,
pr: str = "none",
next_action: str = "none",
blocker: str = "none",
) -> str:
"""Return a structured, machine-parseable issue comment body."""
profile_value = (profile or "unknown").strip() or "unknown"
lines = [
MARKER,
"**Issue claim heartbeat**",
f"- kind: {kind}",
f"- issue: #{issue_number}",
f"- branch: {branch}",
f"- phase: {phase}",
f"- profile: {profile_value}",
f"- pr: {pr}",
f"- blocker: {blocker}",
f"- next_action: {next_action}",
]
return "\n".join(lines)
def parse_heartbeat_comment(body: str) -> dict[str, Any] | None:
"""Parse one structured heartbeat comment, or None when not a heartbeat."""
text = body or ""
if MARKER not in text:
return None
fields: dict[str, str] = {}
for match in _FIELD_RE.finditer(text):
fields[match.group(1).strip().lower()] = match.group(2).strip()
if not fields:
return None
issue_raw = fields.get("issue", "")
issue_digits = re.sub(r"[^\d]", "", issue_raw)
issue_number = int(issue_digits) if issue_digits.isdigit() else None
return {
"kind": fields.get("kind"),
"issue_number": issue_number,
"branch": fields.get("branch"),
"phase": fields.get("phase"),
"profile": fields.get("profile"),
"pr": fields.get("pr"),
"blocker": fields.get("blocker"),
"next_action": fields.get("next_action"),
"raw_fields": fields,
}
def extract_issue_heartbeats(
comments: list[dict],
*,
issue_number: int | None = None,
) -> list[dict]:
"""Return parsed heartbeats newest-last, optionally filtered to *issue_number*."""
heartbeats: list[dict] = []
for comment in comments or []:
parsed = parse_heartbeat_comment(comment.get("body") or "")
if not parsed:
continue
if issue_number is not None and parsed.get("issue_number") != issue_number:
continue
heartbeats.append(
{
**parsed,
"comment_id": comment.get("id"),
"author": (comment.get("user") or {}).get("login")
or comment.get("author"),
"created_at": comment.get("created_at"),
"updated_at": comment.get("updated_at"),
}
)
return heartbeats
def issue_has_in_progress_label(issue: dict) -> bool:
labels = issue.get("labels") or []
for label in labels:
name = label if isinstance(label, str) else label.get("name")
if name == IN_PROGRESS_LABEL:
return True
return False
def _linked_open_pr(issue_number: int, open_prs: list[dict]) -> dict | None:
pattern = f"issue-{issue_number}"
closes = f"closes #{issue_number}"
fixes = f"fixes #{issue_number}"
for pr in open_prs or []:
head = (pr.get("head") or {}).get("ref") or ""
text = f"{pr.get('title', '')} {pr.get('body', '')}".lower()
if pattern in head.lower():
return pr
if closes in text or fixes in text:
return pr
return None
def _matching_branch_names(issue_number: int, branch_names: list[str]) -> list[str]:
pattern = f"issue-{issue_number}"
return [name for name in branch_names if pattern in (name or "").lower()]
def classify_issue_claim(
*,
issue: dict,
comments: list[dict],
open_prs: list[dict] | None = None,
branch_names: list[str] | None = None,
now: datetime | None = None,
heartbeat_lease_minutes: int = 30,
reclaim_after_minutes: int = 60,
) -> dict[str, Any]:
"""Classify one issue's claim state from labels, heartbeats, PRs, and branches."""
issue_number = int(issue.get("number") or 0)
open_prs = open_prs or []
branch_names = branch_names or []
current = now or datetime.now(timezone.utc)
has_label = issue_has_in_progress_label(issue)
heartbeats = extract_issue_heartbeats(comments, issue_number=issue_number)
latest = heartbeats[-1] if heartbeats else None
latest_at = _parse_timestamp(
(latest or {}).get("updated_at") or (latest or {}).get("created_at")
)
age_minutes = None
if latest_at:
age_minutes = int((current - latest_at).total_seconds() // 60)
linked_pr = _linked_open_pr(issue_number, open_prs)
matching_branches = _matching_branch_names(issue_number, branch_names)
if not has_label:
status = "not_claimed"
reasons = ["issue lacks status:in-progress label"]
elif linked_pr:
status = "awaiting_review"
reasons = [f"open PR #{linked_pr.get('number')} covers this issue"]
elif not heartbeats:
status = "phantom"
reasons = [
"status:in-progress label present without structured claim heartbeat"
]
elif latest_at is None:
status = "phantom"
reasons = ["heartbeat comments present but timestamps could not be parsed"]
elif age_minutes is not None and age_minutes <= heartbeat_lease_minutes:
status = "active"
reasons = [f"heartbeat age {age_minutes}m within {heartbeat_lease_minutes}m lease"]
elif matching_branches and age_minutes is not None and age_minutes <= reclaim_after_minutes:
status = "active"
reasons = [
f"matching branch(es) {', '.join(matching_branches)} with heartbeat "
f"age {age_minutes}m"
]
elif age_minutes is not None and age_minutes > reclaim_after_minutes and not matching_branches:
status = "reclaimable"
reasons = [
f"no heartbeat within {reclaim_after_minutes}m and no matching branch"
]
elif age_minutes is not None and age_minutes > heartbeat_lease_minutes:
status = "stale"
reasons = [
f"heartbeat age {age_minutes}m exceeds {heartbeat_lease_minutes}m lease"
]
else:
status = "active"
reasons = ["claim has structured heartbeat proof"]
return {
"issue_number": issue_number,
"title": issue.get("title"),
"status": status,
"has_in_progress_label": has_label,
"heartbeat_count": len(heartbeats),
"latest_heartbeat": latest,
"heartbeat_age_minutes": age_minutes,
"linked_open_pr": linked_pr.get("number") if linked_pr else None,
"matching_branches": matching_branches,
"reasons": reasons,
"reclaimable": status == "reclaimable",
"stale": status in {"stale", "phantom", "reclaimable"},
}
def build_claim_inventory(
*,
issues: list[dict],
comments_by_issue: dict[int, list[dict]],
open_prs: list[dict],
branch_names: list[str],
now: datetime | None = None,
heartbeat_lease_minutes: int = 30,
reclaim_after_minutes: int = 60,
) -> dict[str, Any]:
"""Build a queue inventory report for in-progress issue claims."""
entries: list[dict] = []
for issue in issues or []:
if not issue_has_in_progress_label(issue):
continue
number = int(issue["number"])
entry = classify_issue_claim(
issue=issue,
comments=comments_by_issue.get(number, []),
open_prs=open_prs,
branch_names=branch_names,
now=now,
heartbeat_lease_minutes=heartbeat_lease_minutes,
reclaim_after_minutes=reclaim_after_minutes,
)
entries.append(entry)
counts: dict[str, int] = {}
for entry in entries:
counts[entry["status"]] = counts.get(entry["status"], 0) + 1
return {
"entries": entries,
"counts": counts,
"heartbeat_lease_minutes": heartbeat_lease_minutes,
"reclaim_after_minutes": reclaim_after_minutes,
"in_progress_total": len(entries),
}
def build_cleanup_plan(inventory: dict[str, Any]) -> list[dict]:
"""Return stale/reclaimable/phantom claims that may be cleaned up."""
plan: list[dict] = []
for entry in inventory.get("entries") or []:
if entry.get("status") in {"reclaimable", "phantom"}:
plan.append(
{
"issue_number": entry["issue_number"],
"status": entry["status"],
"action": "remove_status_in_progress_and_comment",
"reasons": list(entry.get("reasons") or []),
}
)
elif entry.get("status") == "stale" and not entry.get("linked_open_pr"):
plan.append(
{
"issue_number": entry["issue_number"],
"status": entry["status"],
"action": "report_only",
"reasons": list(entry.get("reasons") or []),
}
)
return plan
+272
View File
@@ -0,0 +1,272 @@
"""Own-branch lock adoption / recovery for ``gitea_lock_issue`` (#442 / #443).
When an issue's own already-pushed branch exists, lock reacquisition must be
allowed (adoption) instead of being treated as #400 duplicate competing work.
This module isolates the pure decision so it can be unit-tested apart from the
MCP server's live Gitea calls.
Adoption is granted only for the issue's *exact* requested branch. Any other
branch that merely contains the same ``issue-<n>`` marker is competing work and
stays fail-closed. Open-PR, competing-live-lock, capability, and worktree
safety checks are enforced by the caller before this decision is consulted;
this module additionally records whether they passed for proof purposes.
"""
from __future__ import annotations
import re
ADOPT = "adopt_existing_branch"
BLOCK_COMPETING = "block_competing_branch"
NO_MATCH = "no_matching_branch"
# Citable decision labels aligned with the ``assess_own_branch_adoption``
# outcomes, surfaced verbatim in the live ``gitea_lock_issue`` response so
# recovery reports (#473-style) can quote the lock tool output directly
# instead of inferring adoption from separate offline checks (#477).
DECISION_LABELS = {
ADOPT: "ADOPT",
BLOCK_COMPETING: "BLOCK_COMPETING",
NO_MATCH: "NO_MATCH",
}
_SAFE_NEXT_ACTIONS = {
ADOPT: (
"Own existing branch adopted for lock recovery; proceed to "
"gitea_create_pr for this issue and cite this adoption proof."
),
BLOCK_COMPETING: (
"Competing same-issue branch(es) exist; resolve branch ownership "
"before locking. No adoption performed (fail closed)."
),
NO_MATCH: (
"No existing branch carries this issue marker; normal lock path "
"applied. No adoption performed."
),
}
def decision_label(outcome: str) -> str:
"""Map an ``assess_own_branch_adoption`` outcome to its citable label."""
return DECISION_LABELS.get(outcome, "UNKNOWN")
def safe_next_action(outcome: str) -> str:
"""Return the safe next action string for an adoption *outcome*."""
return _SAFE_NEXT_ACTIONS.get(
outcome, "Unknown adoption outcome; treat as fail closed."
)
def _branch_name(entry) -> str:
if isinstance(entry, dict):
return str(entry.get("name") or "")
return str(entry or "")
def _branch_sha(entry) -> str | None:
if isinstance(entry, dict):
sha = entry.get("commit_sha")
if sha:
return str(sha)
return None
def _branch_carries_issue_marker(branch_name: str, issue_number: int) -> bool:
"""Return True when *branch_name* references issue *issue_number* exactly.
Uses a numeric word-boundary so ``issue-42`` does not match inside
``issue-420`` (AC6 / #440).
"""
name = (branch_name or "").strip()
if not name:
return False
pattern = rf"(?:^|/)issue-{int(issue_number)}(?![0-9])"
return re.search(pattern, name) is not None
def assess_own_branch_adoption(
*,
issue_number: int,
requested_branch: str,
existing_branches,
) -> dict:
"""Decide whether an existing matching branch is adoptable.
Args:
issue_number: The tracking issue number being locked.
requested_branch: The exact branch the caller wants to lock.
existing_branches: Iterable of remote branch entries — either names or
dicts with ``name`` and optional ``commit_sha``.
Returns:
dict with:
* ``outcome`` — one of ADOPT / BLOCK_COMPETING / NO_MATCH
* ``adopt`` (bool), ``block`` (bool)
* ``reason`` (str)
* ``matched_branch`` (str | None), ``matched_head_sha`` (str | None)
* ``competing_branches`` (list[str])
ADOPT: the issue's exact branch exists and no other same-issue branch does.
BLOCK_COMPETING: at least one same-issue branch is not the requested branch.
NO_MATCH: no branch carries the issue marker — normal lock path applies.
"""
requested = (requested_branch or "").strip()
matches: list[tuple[str, str | None]] = []
for entry in existing_branches or []:
name = _branch_name(entry).strip()
if _branch_carries_issue_marker(name, issue_number):
matches.append((name, _branch_sha(entry)))
competing = sorted({name for name, _ in matches if name != requested})
exact = [(name, sha) for name, sha in matches if name == requested]
# Fail closed whenever any non-requested same-issue branch exists, even if
# the requested branch is also present: ownership is then ambiguous.
if competing:
return {
"outcome": BLOCK_COMPETING,
"adopt": False,
"block": True,
"reason": (
f"issue #{issue_number} already has matching branch(es) "
f"{competing} that are not the requested branch "
f"'{requested}' (fail closed)"
),
"matched_branch": None,
"matched_head_sha": None,
"competing_branches": competing,
}
if exact:
name, sha = exact[0]
return {
"outcome": ADOPT,
"adopt": True,
"block": False,
"reason": (
f"existing branch '{name}' is the exact requested branch for "
f"issue #{issue_number}; adopting it for lock recovery"
),
"matched_branch": name,
"matched_head_sha": sha,
"competing_branches": [],
}
return {
"outcome": NO_MATCH,
"adopt": False,
"block": False,
"reason": f"no existing branch matches issue #{issue_number}",
"matched_branch": None,
"matched_head_sha": None,
"competing_branches": [],
}
def _matcher_summary(issue_number: int, assessment: dict) -> str:
"""Explain, citably, why the assessed branch did or did not qualify.
Names the numeric word-boundary rule so reports can show that
``issue-42`` was not matched inside ``issue-420`` (#440 / #477 AC3).
"""
outcome = assessment.get("outcome")
matched = assessment.get("matched_branch")
competing = assessment.get("competing_branches") or []
if outcome == ADOPT and matched:
return (
f"branch '{matched}' exactly matches the issue-{int(issue_number)} "
f"marker (numeric word-boundary; 'issue-{int(issue_number)}' is not "
f"matched inside 'issue-{int(issue_number)}0')"
)
if outcome == BLOCK_COMPETING:
return (
f"competing same-issue branch(es) {competing} carry the "
f"issue-{int(issue_number)} marker but are not the requested "
f"branch; ownership is ambiguous (fail closed)"
)
return (
f"no existing branch carries the issue-{int(issue_number)} marker "
f"under the numeric word-boundary rule"
)
def _competing_branch_check(assessment: dict) -> dict:
"""Structured competing-branch verdict for the proof block."""
competing = list(assessment.get("competing_branches") or [])
return {
"result": "blocked" if competing else "clear",
"competing_branches": competing,
}
def build_adoption_proof(
*,
issue_number: int,
branch_name: str,
assessment: dict,
open_pr_checked: bool,
competing_lock_checked: bool,
lock_file_path: str,
lock_file_status: str,
) -> dict:
"""Assemble the proof block returned by ``gitea_lock_issue`` on adoption.
Requirement #4: adoption results must carry issue number, branch name,
branch head commit, adoption reason, no-existing-PR proof, no-competing-
live-lock proof, and lock file path/status.
#477: additionally surface explicit, citable adoption-proof fields tied to
the ``assess_own_branch_adoption`` outcome (``adoption_decision``,
``adopted``, ``adopted_branch``, ``adopted_branch_head``,
``matcher_summary``, ``competing_branch_check``, ``safe_next_action``) so a
recovery session can quote the live lock response directly. The explicit
fields are populated for any outcome; ``adopted_branch`` /
``adopted_branch_head`` are set only when the outcome is ADOPT so a
non-adoption proof can never be misread as claiming adoption.
"""
outcome = assessment.get("outcome")
adopted = outcome == ADOPT
return {
"issue_number": issue_number,
"branch_name": branch_name,
"branch_head_commit": assessment.get("matched_head_sha"),
"adoption_reason": assessment.get("reason"),
"no_existing_pr_proof": bool(open_pr_checked),
"no_competing_live_lock_proof": bool(competing_lock_checked),
"lock_file_path": lock_file_path,
"lock_file_status": lock_file_status,
# Explicit citable fields (#477).
"adoption_decision": decision_label(outcome),
"adopted": adopted,
"adopted_branch": branch_name if adopted else None,
"adopted_branch_head": assessment.get("matched_head_sha") if adopted else None,
"matcher_summary": _matcher_summary(issue_number, assessment),
"competing_branch_check": _competing_branch_check(assessment),
"safe_next_action": safe_next_action(outcome),
}
def build_non_adoption_lock_proof(*, issue_number: int, branch_name: str) -> dict:
"""Safe, adoption-free proof metadata for a normal (NO_MATCH) lock.
Requirement #477 AC2: non-adoption lock responses must stay clear and must
not imply adoption. This returns explicit ``adopted: False`` metadata with
the ``NO_MATCH`` decision so a normal lock response can carry citable proof
without ever asserting a branch was adopted.
"""
return {
"issue_number": issue_number,
"branch_name": branch_name,
"adoption_decision": DECISION_LABELS[NO_MATCH],
"adopted": False,
"adopted_branch": None,
"adopted_branch_head": None,
"matcher_summary": (
f"no existing branch carries the issue-{int(issue_number)} marker; "
f"normal lock path (no adoption)"
),
"competing_branch_check": {"result": "clear", "competing_branches": []},
"safe_next_action": safe_next_action(NO_MATCH),
}
+269
View File
@@ -0,0 +1,269 @@
"""Issue-lock provenance and external-state disclosure (#447).
Sanctioned locks are written only by ``gitea_lock_issue`` (or adoption recovery
#442). Manual seeding of ``/tmp/gitea_issue_lock.json`` is unsafe and must be
blocked at PR creation unless explicit operator override proof is recorded.
"""
from __future__ import annotations
import os
import re
from datetime import datetime, timezone
ISSUE_LOCK_FILE = os.environ.get("GITEA_ISSUE_LOCK_FILE", "/tmp/gitea_issue_lock.json")
SOURCE_LOCK_ISSUE = "gitea_lock_issue"
SOURCE_LOCK_ADOPTION = "gitea_lock_issue_adoption"
SOURCE_OPERATOR_OVERRIDE = "operator_override"
SANCTIONED_LOCK_SOURCES = frozenset({
SOURCE_LOCK_ISSUE,
SOURCE_LOCK_ADOPTION,
SOURCE_OPERATOR_OVERRIDE,
})
_OPERATOR_OVERRIDE_ENV = "GITEA_ISSUE_LOCK_OPERATOR_OVERRIDE"
_ISSUE_LOCK_PATH_RE = re.compile(
r"(?:/tmp/)?gitea_issue_lock\.json",
re.IGNORECASE,
)
_LOCK_SEED_RE = re.compile(
r"(?:seed(?:ed|ing)?|restor(?:e|ed|ing)|wrote|written|write|programmatically|"
r"hand[- ]forg|manual(?:ly)?).{0,80}gitea_issue_lock",
re.IGNORECASE | re.DOTALL,
)
_LOCK_REMOVE_RE = re.compile(
r"(?:\brm\b|remove|deleted?|unlink).{0,80}gitea_issue_lock",
re.IGNORECASE | re.DOTALL,
)
_LOCK_READ_RE = re.compile(
r"(?:read|loaded?|parsed?).{0,80}gitea_issue_lock",
re.IGNORECASE | re.DOTALL,
)
_EXTERNAL_NONE_RE = re.compile(
r"external[- ]state mutations\s*:\s*none\b",
re.IGNORECASE,
)
_EXTERNAL_FIELD_RE = re.compile(
r"external[- ]state mutations\s*:\s*(.+)$",
re.IGNORECASE | re.MULTILINE,
)
_CLEANUP_ONLY_RE = re.compile(
r"cleanup mutations\s*:\s*(?:none|lock removed|removed issue lock)",
re.IGNORECASE,
)
_PR_CREATED_RE = re.compile(
r"(?:\bgitea_create_pr\b|PR\s*#\s*\d+\s+created|created\s+PR\s*#|opened\s+PR\s*#|"
r"PR\s+creation\s+(?:succeeded|complete))",
re.IGNORECASE,
)
_REVIEW_APPROVE_RE = re.compile(
r"(?:submitted\s+(?:['\"]approve['\"]|approve\s+review)|"
r"review decision\s*:\s*approve|approved\s+PR\s*#|gitea_review_pr.*approve)",
re.IGNORECASE,
)
_OVERRIDE_PROOF_RE = re.compile(
r"operator[- ]override\s+proof\s*:\s*(.+)$",
re.IGNORECASE | re.MULTILINE,
)
def _utc_now_iso() -> str:
return datetime.now(timezone.utc).strftime("%Y-%m-%dT%H:%M:%SZ")
def build_sanctioned_lock_provenance(
*,
tool: str,
source: str = SOURCE_LOCK_ISSUE,
claimant: dict | None = None,
adoption: dict | None = None,
) -> dict:
"""Return provenance metadata stored with a sanctioned lock write."""
entry = {
"source": source,
"written_at": _utc_now_iso(),
"written_by_tool": tool,
"lock_file_path": ISSUE_LOCK_FILE,
}
if claimant:
entry["claimant"] = claimant
if adoption:
entry["adoption"] = adoption
return entry
def operator_override_requested() -> bool:
return os.environ.get(_OPERATOR_OVERRIDE_ENV, "").strip().lower() in {
"1",
"true",
"yes",
}
def build_operator_override_provenance(*, reason: str, claimant: dict | None = None) -> dict:
text = (reason or "").strip()
if not text:
raise ValueError(
"operator override requires a non-empty override reason (fail closed)"
)
entry = build_sanctioned_lock_provenance(
tool="operator_override",
source=SOURCE_OPERATOR_OVERRIDE,
claimant=claimant,
)
entry["override_reason"] = text
return entry
def assess_lock_file_for_create_pr(lock_data: dict | None) -> dict:
"""Fail closed when lock file lacks sanctioned provenance (#447)."""
data = lock_data if isinstance(lock_data, dict) else {}
reasons: list[str] = []
provenance = data.get("lock_provenance")
if not isinstance(provenance, dict):
reasons.append(
"issue lock file lacks sanctioned lock_provenance; manual seeding is "
"not a normal recovery path — call gitea_lock_issue or use #442 adoption"
)
return _provenance_result(False, reasons, provenance)
source = str(provenance.get("source") or "").strip()
if source not in SANCTIONED_LOCK_SOURCES:
reasons.append(
f"issue lock provenance source '{source or '(missing)'}' is not sanctioned"
)
if source == SOURCE_OPERATOR_OVERRIDE and not str(
provenance.get("override_reason") or ""
).strip():
reasons.append(
"operator_override lock provenance requires override_reason proof"
)
if not data.get("work_lease"):
reasons.append("issue lock file missing work_lease metadata")
if not str(provenance.get("written_by_tool") or "").strip():
reasons.append("issue lock provenance missing written_by_tool")
proven = not reasons
return _provenance_result(proven, reasons, provenance)
def _provenance_result(proven: bool, reasons: list[str], provenance: dict | None) -> dict:
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"lock_provenance": provenance,
}
def format_lock_provenance_error(assessment: dict) -> str:
reasons = "; ".join(assessment.get("reasons") or ["unknown lock provenance violation"])
return f"Issue lock provenance guard (#447): {reasons} (fail closed)"
def _lock_activity_detected(text: str) -> dict[str, bool]:
body = text or ""
return {
"seed_or_restore": bool(_LOCK_SEED_RE.search(body)),
"remove": bool(_LOCK_REMOVE_RE.search(body)),
"read": bool(_LOCK_READ_RE.search(body)),
}
def _external_state_discloses_lock(text: str) -> bool:
match = _EXTERNAL_FIELD_RE.search(text or "")
if not match:
return False
value = (match.group(1) or "").strip().lower()
if value in {"", "none", "n/a"}:
return False
return "lock" in value or "gitea_issue_lock" in value or "issue-lock" in value
def assess_issue_lock_external_state_report(report_text: str) -> dict:
"""Require explicit external-state disclosure for issue-lock mutations (#447)."""
text = report_text or ""
activity = _lock_activity_detected(text)
if not any(activity.values()):
return {"proven": True, "block": False, "reasons": [], "activity": activity}
reasons: list[str] = []
disclosed = _external_state_discloses_lock(text)
if activity["seed_or_restore"] and _EXTERNAL_NONE_RE.search(text):
reasons.append(
"report mentions seeding/restoring gitea_issue_lock.json but claims "
"External-state mutations: none"
)
elif activity["seed_or_restore"] and not disclosed:
reasons.append(
"report mentions issue-lock file activity but External-state mutations "
"does not disclose read/write of gitea_issue_lock.json"
)
if activity["remove"]:
if _EXTERNAL_NONE_RE.search(text):
reasons.append(
"report mentions removing gitea_issue_lock.json but claims "
"External-state mutations: none"
)
elif not disclosed and _CLEANUP_ONLY_RE.search(text):
reasons.append(
"report removes issue lock but classifies it as cleanup only; "
"record under External-state mutations"
)
elif not disclosed:
reasons.append(
"report mentions deleting issue lock without External-state "
"mutation disclosure"
)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"activity": activity,
}
def assess_manual_lock_pr_without_override(report_text: str) -> dict:
"""Block reports that created a PR via manual lock seed without override proof."""
text = report_text or ""
seeded = bool(_LOCK_SEED_RE.search(text))
created = bool(_PR_CREATED_RE.search(text))
if not (seeded and created):
return {"proven": True, "block": False, "reasons": []}
if _OVERRIDE_PROOF_RE.search(text):
return {"proven": True, "block": False, "reasons": []}
return {
"proven": False,
"block": True,
"reasons": [
"report created/opened a PR after manual issue-lock seeding without "
"operator override proof"
],
}
def assess_author_reviewer_same_run_report(report_text: str) -> dict:
"""Reviewer handoff must not create and approve the same PR in one run (#447)."""
text = report_text or ""
if not (_PR_CREATED_RE.search(text) and _REVIEW_APPROVE_RE.search(text)):
return {"proven": True, "block": False, "reasons": []}
return {
"proven": False,
"block": True,
"reasons": [
"report mixes author-side PR creation and reviewer approval in one "
"final handoff; split author and reviewer sessions"
],
}
+612
View File
@@ -0,0 +1,612 @@
"""Keyed, persistent issue-lock storage (#443) with flock hardening (#438).
Replaces the single global ``/tmp/gitea_issue_lock.json`` slot with per-issue
lock files under ``GITEA_ISSUE_LOCK_DIR`` (default
``~/.cache/gitea-tools/issue-locks``). Each MCP session binds its active lock
via a per-process pointer file so concurrent repos/issues never clobber each
other. Acquisition is serialized per issue with ``fcntl.flock``.
"""
from __future__ import annotations
import errno
import fcntl
import json
import os
import re
import tempfile
from contextlib import contextmanager
from datetime import datetime, timedelta, timezone
from typing import Any
LOCK_DIR_ENV = "GITEA_ISSUE_LOCK_DIR"
DEFAULT_LOCK_DIR = os.path.expanduser("~/.cache/gitea-tools/issue-locks")
WORK_LEASE_TTL_HOURS = 4
AUTHOR_ISSUE_WORK_LEASE = "author_issue_work"
_SAFE_SEGMENT_RE = re.compile(r"[^A-Za-z0-9._+-]+")
class LockContentionError(RuntimeError):
"""Raised when an exclusive per-issue lock cannot be acquired."""
def default_lock_dir() -> str:
raw = (os.environ.get(LOCK_DIR_ENV) or DEFAULT_LOCK_DIR).strip()
return raw or DEFAULT_LOCK_DIR
def _sanitize_segment(value: str) -> str:
text = (value or "").strip()
if not text:
return "_"
return _SAFE_SEGMENT_RE.sub("_", text)
def lock_key(
*,
remote: str,
org: str,
repo: str,
issue_number: int,
) -> str:
return "-".join(
_sanitize_segment(part)
for part in (remote, org, repo, str(issue_number))
)
def lock_file_path(
*,
remote: str,
org: str,
repo: str,
issue_number: int,
lock_dir: str | None = None,
) -> str:
root = (lock_dir or default_lock_dir()).strip()
return os.path.join(root, f"{lock_key(remote=remote, org=org, repo=repo, issue_number=issue_number)}.json")
def session_pointer_path(lock_dir: str | None = None) -> str:
root = (lock_dir or default_lock_dir()).strip()
return os.path.join(root, f"session-{os.getpid()}.json")
def _ensure_lock_dir(lock_dir: str | None = None) -> str:
root = (lock_dir or default_lock_dir()).strip()
os.makedirs(root, mode=0o700, exist_ok=True)
return root
def flock_path(json_path: str) -> str:
return f"{json_path}.lock"
def is_process_alive(pid: int | None) -> bool:
if not pid or pid <= 0:
return False
try:
os.kill(int(pid), 0)
return True
except OSError as exc:
return exc.errno != errno.ESRCH
except (TypeError, ValueError):
return False
@contextmanager
def _exclusive_file_lock(lock_path: str):
os.makedirs(os.path.dirname(lock_path) or ".", exist_ok=True)
fd = os.open(lock_path, os.O_CREAT | os.O_RDWR, 0o600)
try:
try:
fcntl.flock(fd, fcntl.LOCK_EX | fcntl.LOCK_NB)
except BlockingIOError as exc:
raise LockContentionError(
f"could not acquire exclusive lock on '{lock_path}'"
) from exc
yield fd
finally:
try:
fcntl.flock(fd, fcntl.LOCK_UN)
finally:
os.close(fd)
def read_lock_file(path: str) -> dict[str, Any] | None:
lock_path = (path or "").strip()
if not lock_path or not os.path.exists(lock_path):
return None
try:
with open(lock_path, encoding="utf-8") as handle:
data = json.load(handle)
except (OSError, json.JSONDecodeError):
return None
return data if isinstance(data, dict) else None
def save_lock_file(path: str, data: dict[str, Any]) -> None:
lock_path = (path or "").strip()
if not lock_path:
raise ValueError("lock path is required (fail closed)")
parent = os.path.dirname(lock_path) or "."
os.makedirs(parent, mode=0o700, exist_ok=True)
payload = json.dumps(data, indent=2, sort_keys=True) + "\n"
fd, temp_path = tempfile.mkstemp(prefix=".lock-", suffix=".json", dir=parent)
try:
with os.fdopen(fd, "w", encoding="utf-8") as handle:
handle.write(payload)
handle.flush()
os.fsync(handle.fileno())
os.replace(temp_path, lock_path)
finally:
if os.path.exists(temp_path):
try:
os.remove(temp_path)
except OSError:
pass
def bind_session_lock(lock_data: dict[str, Any], lock_dir: str | None = None) -> str:
"""Persist a keyed lock and bind it to the current process session."""
remote = str(lock_data.get("remote") or "")
org = str(lock_data.get("org") or "")
repo = str(lock_data.get("repo") or "")
issue_number = int(lock_data.get("issue_number") or 0)
if not remote or not org or not repo or issue_number <= 0:
raise ValueError("lock record must include remote, org, repo, and issue_number")
root = _ensure_lock_dir(lock_dir)
path = lock_file_path(
remote=remote,
org=org,
repo=repo,
issue_number=issue_number,
lock_dir=root,
)
record = dict(lock_data)
record["lock_file_path"] = path
record["session_pid"] = os.getpid()
record.setdefault("pid", os.getpid())
pointer = {
"pid": os.getpid(),
"lock_file_path": path,
"issue_number": issue_number,
"branch_name": record.get("branch_name"),
"remote": remote,
"org": org,
"repo": repo,
}
sentinel = flock_path(path)
try:
with _exclusive_file_lock(sentinel):
existing = read_lock_file(path)
overwrite_block = assess_foreign_lock_overwrite(existing, record)
if overwrite_block:
raise RuntimeError(overwrite_block)
lease_block = assess_same_issue_lease_conflict(
existing,
issue_number=issue_number,
branch_name=str(record.get("branch_name") or ""),
worktree_path=str(record.get("worktree_path") or ""),
)
if lease_block:
raise RuntimeError(lease_block)
save_lock_file(path, record)
save_lock_file(session_pointer_path(root), pointer)
except LockContentionError as exc:
competing = read_lock_file(path)
if competing:
owner_pid = competing.get("session_pid") or competing.get("pid")
raise RuntimeError(
f"Issue #{issue_number} lock contention: {exc}; competing owner "
f"pid={owner_pid} (fail closed)"
) from exc
raise RuntimeError(f"Issue #{issue_number} lock contention: {exc} (fail closed)") from exc
return path
def read_session_issue_lock(lock_dir: str | None = None) -> dict[str, Any] | None:
root = (lock_dir or default_lock_dir()).strip()
pointer = read_lock_file(session_pointer_path(root))
if not pointer:
return None
lock_path = str(pointer.get("lock_file_path") or "").strip()
if not lock_path:
return None
return read_lock_file(lock_path)
def load_issue_lock(
*,
remote: str,
org: str,
repo: str,
issue_number: int,
lock_dir: str | None = None,
) -> dict[str, Any] | None:
return read_lock_file(
lock_file_path(
remote=remote,
org=org,
repo=repo,
issue_number=issue_number,
lock_dir=lock_dir,
)
)
def iter_lock_files(lock_dir: str | None = None) -> list[str]:
root = (lock_dir or default_lock_dir()).strip()
if not os.path.isdir(root):
return []
paths: list[str] = []
for name in os.listdir(root):
if not name.endswith(".json") or name.startswith("session-"):
continue
paths.append(os.path.join(root, name))
return sorted(paths)
def find_lock_for_branch(
*,
remote: str,
org: str,
repo: str,
branch_name: str,
lock_dir: str | None = None,
) -> dict[str, Any] | None:
target = (branch_name or "").strip()
if not target:
return None
for path in iter_lock_files(lock_dir):
lock = read_lock_file(path)
if not lock:
continue
if (
str(lock.get("remote") or "") == remote
and str(lock.get("org") or "") == org
and str(lock.get("repo") or "") == repo
and str(lock.get("branch_name") or "").strip() == target
):
lock = dict(lock)
lock.setdefault("lock_file_path", path)
return lock
return None
def _lease_now(now: datetime | None = None) -> datetime:
return now or datetime.now(timezone.utc)
def _parse_lease_timestamp(value: str | None) -> datetime | None:
text = (value or "").strip()
if not text:
return None
try:
return datetime.fromisoformat(text.replace("Z", "+00:00")).astimezone(timezone.utc)
except ValueError:
return None
def lease_expires_at(lock: dict[str, Any] | None) -> datetime | None:
if not lock:
return None
lease = lock.get("work_lease")
if not isinstance(lease, dict):
return None
return _parse_lease_timestamp(lease.get("expires_at"))
def is_lease_expired(lock: dict[str, Any] | None, *, now: datetime | None = None) -> bool:
expires = lease_expires_at(lock)
if expires is None:
return False
return expires <= _lease_now(now)
def is_lease_live(lock: dict[str, Any] | None, *, now: datetime | None = None) -> bool:
return assess_lock_freshness(lock, now=now)["live"]
def assess_lock_freshness(
lock_data: dict[str, Any] | None,
*,
now: datetime | None = None,
) -> dict[str, Any]:
"""Classify a lock as live, expired, stale, or absent."""
current = _lease_now(now)
if not lock_data:
return {
"status": "absent",
"live": False,
"stale": False,
"reason": "no lock record",
}
expires_at = lease_expires_at(lock_data)
lease = lock_data.get("work_lease")
heartbeat_at = _parse_lease_timestamp(lock_data.get("last_heartbeat_at"))
if heartbeat_at is None and isinstance(lease, dict):
heartbeat_at = _parse_lease_timestamp(lease.get("last_heartbeat_at"))
pid = lock_data.get("session_pid")
if pid is None:
pid = lock_data.get("pid")
pid_alive = is_process_alive(pid) if pid is not None else False
if expires_at and expires_at <= current:
return {
"status": "expired",
"live": False,
"stale": True,
"reason": f"lease expired at {expires_at.isoformat()}",
"pid_alive": pid_alive,
}
if pid is not None and not pid_alive:
return {
"status": "stale",
"live": False,
"stale": True,
"reason": f"owner pid {pid} is not alive",
"pid_alive": False,
}
return {
"status": "live",
"live": True,
"stale": False,
"reason": "lock heartbeat and lease are fresh",
"pid_alive": pid_alive,
"heartbeat_at": heartbeat_at.isoformat() if heartbeat_at else None,
"expires_at": expires_at.isoformat() if expires_at else None,
}
def _same_realpath(left: str | None, right: str | None) -> bool:
if not left or not right:
return False
try:
return os.path.realpath(left) == os.path.realpath(right)
except OSError:
return left == right
def assess_same_issue_lease_conflict(
existing_lock: dict[str, Any] | None,
*,
issue_number: int,
branch_name: str,
worktree_path: str,
operation_type: str = AUTHOR_ISSUE_WORK_LEASE,
now: datetime | None = None,
) -> str | None:
"""Return a fail-closed error when a competing live lease blocks acquisition."""
if not existing_lock:
return None
existing_issue = existing_lock.get("issue_number")
lease = existing_lock.get("work_lease")
existing_operation = (
lease.get("operation_type")
if isinstance(lease, dict)
else AUTHOR_ISSUE_WORK_LEASE
)
if existing_issue != issue_number or existing_operation != operation_type:
return None
existing_branch = existing_lock.get("branch_name")
existing_worktree = existing_lock.get("worktree_path")
same_owner = (
existing_branch == branch_name
and _same_realpath(str(existing_worktree or ""), worktree_path)
)
if is_lease_expired(existing_lock, now=now):
return (
f"Issue #{issue_number} has an expired {operation_type} lease on "
f"branch '{existing_branch}' from worktree '{existing_worktree}'. "
"Recovery review is required before takeover (fail closed)"
)
if same_owner:
return None
return (
f"Issue #{issue_number} already has an active {operation_type} lease on "
f"branch '{existing_branch}' from worktree '{existing_worktree}' "
"(fail closed)"
)
def assess_foreign_lock_overwrite(
existing_lock: dict[str, Any] | None,
incoming_lock: dict[str, Any],
*,
now: datetime | None = None,
) -> str | None:
"""Block writes that would clobber an unrelated live lease on the same key."""
if not existing_lock:
return None
same_issue = existing_lock.get("issue_number") == incoming_lock.get("issue_number")
same_branch = existing_lock.get("branch_name") == incoming_lock.get("branch_name")
same_worktree = _same_realpath(
str(existing_lock.get("worktree_path") or ""),
str(incoming_lock.get("worktree_path") or ""),
)
if same_issue and same_branch and same_worktree:
return None
if not is_lease_live(existing_lock, now=now):
return None
return (
"Refusing to overwrite a live foreign issue lock "
f"(issue #{existing_lock.get('issue_number')}, "
f"branch '{existing_lock.get('branch_name')}', "
f"worktree '{existing_lock.get('worktree_path')}') (fail closed)"
)
def find_live_lock_for_branch(
branch_name: str,
lock_dir: str | None = None,
) -> dict[str, Any] | None:
target = (branch_name or "").strip()
if not target:
return None
for path in iter_lock_files(lock_dir):
lock = read_lock_file(path)
if not lock:
continue
if str(lock.get("branch_name") or "").strip() != target:
continue
if not is_lease_live(lock):
continue
record = dict(lock)
record.setdefault("lock_file_path", path)
return record
return None
def resolve_locked_branch_for_session(
branch_name: str | None = None,
lock_dir: str | None = None,
) -> str:
if branch_name:
lock = find_live_lock_for_branch(branch_name, lock_dir)
if lock:
return str(lock.get("branch_name") or "")
lock = read_session_issue_lock(lock_dir)
return str((lock or {}).get("branch_name") or "")
def has_active_issue_lock(
branch: str,
*,
lock_dir: str | None = None,
) -> bool:
target = (branch or "").strip()
if not target:
return False
for path in iter_lock_files(lock_dir):
lock = read_lock_file(path)
if not lock:
continue
if str(lock.get("branch_name") or "").strip() != target:
continue
if is_lease_live(lock):
return True
return False
def verify_lock_for_mutation(
lock_data: dict[str, Any] | None,
*,
issue_number: int | None = None,
branch_name: str | None = None,
worktree_path: str | None = None,
) -> dict[str, Any]:
"""Re-check lock ownership immediately before a mutation (#438)."""
reasons: list[str] = []
if not lock_data:
return {"proven": False, "block": True, "reasons": ["issue lock is missing (fail closed)"]}
freshness = assess_lock_freshness(lock_data)
if not freshness["live"]:
reasons.append(f"issue lock is not live: {freshness['reason']} (fail closed)")
if issue_number is not None and lock_data.get("issue_number") != issue_number:
reasons.append(
f"issue lock targets #{lock_data.get('issue_number')}, expected #{issue_number} (fail closed)"
)
if branch_name is not None and lock_data.get("branch_name") != branch_name:
reasons.append(
f"issue lock branch '{lock_data.get('branch_name')}' does not match "
f"'{branch_name}' (fail closed)"
)
if worktree_path is not None:
locked = os.path.realpath(str(lock_data.get("worktree_path") or ""))
declared = os.path.realpath(worktree_path)
if locked != declared:
reasons.append(
f"issue lock worktree '{locked}' does not match declared '{declared}' (fail closed)"
)
return {
"proven": not reasons,
"block": bool(reasons),
"reasons": reasons,
"freshness": freshness,
"lock_proof": format_lock_proof(lock_data, freshness=freshness),
}
def list_live_locks(
*,
lock_dir: str | None = None,
now: datetime | None = None,
) -> list[dict[str, Any]]:
"""Return live per-issue locks for queue visibility."""
live: list[dict[str, Any]] = []
for path in iter_lock_files(lock_dir):
record = read_lock_file(path)
if not record:
continue
freshness = assess_lock_freshness(record, now=now)
if not freshness["live"]:
continue
live.append(
{
"issue_number": record.get("issue_number"),
"branch_name": record.get("branch_name"),
"remote": record.get("remote"),
"org": record.get("org"),
"repo": record.get("repo"),
"worktree_path": record.get("worktree_path"),
"pid": record.get("session_pid") or record.get("pid"),
"claimant": (
record.get("claimant")
or (record.get("work_lease") or {}).get("claimant")
),
"freshness": freshness,
"lock_path": record.get("lock_file_path") or path,
}
)
return live
def format_lock_proof(
lock_data: dict[str, Any] | None,
*,
freshness: dict[str, Any] | None = None,
competing_live_locks: list[dict[str, Any]] | None = None,
released: bool | None = None,
) -> str:
"""Canonical issue-lock proof string for final reports."""
if not lock_data:
return "issue lock proof: not acquired"
fresh = freshness or assess_lock_freshness(lock_data)
owner = lock_data.get("claimant") or {}
if not owner and isinstance(lock_data.get("work_lease"), dict):
owner = lock_data["work_lease"].get("claimant") or {}
parts = [
"issue lock proof:",
f"acquired issue #{lock_data.get('issue_number')}",
f"branch {lock_data.get('branch_name')}",
f"owner {owner.get('profile') or 'unknown'}",
f"pid {lock_data.get('session_pid') or lock_data.get('pid')}",
f"freshness {fresh.get('status')}",
]
if competing_live_locks is not None:
parts.append(
"no competing live lock"
if not competing_live_locks
else f"competing live locks {len(competing_live_locks)}"
)
if released is True:
parts.append("lock released")
elif released is False:
parts.append("lock retained")
return "; ".join(parts)
+244
View File
@@ -0,0 +1,244 @@
"""Issue-lock worktree validation (#249).
Author issue locks must validate the caller's own scratch clone (or declared
worktree path), not the shared MCP server working directory. A clean scratch at
``master``/``main`` must remain lockable while an unrelated session leaves the
shared dev worktree dirty or on a feature branch.
"""
from __future__ import annotations
import os
import subprocess
from reviewer_worktree import parse_dirty_tracked_files
AUTHOR_WORKTREE_ENV = "GITEA_AUTHOR_WORKTREE"
BASE_BRANCHES = frozenset({"master", "main", "dev"})
def resolve_author_worktree_path(
explicit: str | None,
project_root: str,
) -> str:
"""Resolve the author worktree path for lock/PR gates."""
path = (explicit or "").strip()
if not path:
path = (os.environ.get(AUTHOR_WORKTREE_ENV) or "").strip()
if not path:
path = project_root
return os.path.realpath(os.path.abspath(path))
def read_worktree_git_state(
worktree_path: str,
extra_bases: tuple[str, ...] | list[str] = (),
) -> dict:
"""Read branch name and porcelain status from a git worktree.
``extra_bases`` names additional branches (e.g. an approved stacked base)
that may anchor base-equivalence in addition to master/main/dev. When empty
(the default), only the normal base branches are considered.
"""
path = (worktree_path or "").strip()
if not path:
return {"current_branch": None, "porcelain_status": ""}
branch_res = subprocess.run(
["git", "-C", path, "branch", "--show-current"],
capture_output=True,
text=True,
check=False,
)
current_branch = (branch_res.stdout or "").strip() or None
status_res = subprocess.run(
["git", "-C", path, "status", "--porcelain"],
capture_output=True,
text=True,
check=False,
)
root_res = subprocess.run(
["git", "-C", path, "rev-parse", "--show-toplevel"],
capture_output=True,
text=True,
check=False,
)
head_res = subprocess.run(
["git", "-C", path, "rev-parse", "HEAD"],
capture_output=True,
text=True,
check=False,
)
head_sha = (head_res.stdout or "").strip() if head_res.returncode == 0 else None
base_branch, base_sha = _find_matching_base_ref(path, head_sha, extra_bases)
return {
"current_branch": current_branch,
"porcelain_status": status_res.stdout or "",
"inspected_git_root": (root_res.stdout or "").strip() if root_res.returncode == 0 else None,
"head_sha": head_sha,
"base_branch": base_branch,
"base_sha": base_sha,
"base_equivalent": bool(head_sha and base_sha and head_sha == base_sha),
}
def assess_issue_lock_worktree(
*,
worktree_path: str,
current_branch: str | None,
porcelain_status: str,
base_equivalent: bool | None = None,
inspected_git_root: str | None = None,
base_branch: str | None = None,
base_branches: frozenset[str] | None = None,
) -> dict:
"""Fail closed when lock preconditions are not met on the declared worktree."""
bases = base_branches or BASE_BRANCHES
reasons: list[str] = []
path = (worktree_path or "").strip()
if not path:
reasons.append("worktree path not declared for issue lock; fail closed")
return _assessment(False, reasons, path, None, [])
branch = (current_branch or "").strip()
dirty_files = parse_dirty_tracked_files(porcelain_status)
if dirty_files:
reasons.append(
"tracked file edits exist before issue lock; "
f"lock must precede implementation work in '{path}' "
f"(dirty files: {', '.join(dirty_files)})"
)
if base_equivalent is False:
reasons.append(
"issue lock worktree must be base-equivalent to one of "
f"{_base_list(bases)} before implementation work; inspected "
f"branch '{branch or '(detached)'}' at '{path}'"
)
elif base_equivalent is None:
if not branch:
reasons.append(
"current branch unknown (detached HEAD?); issue lock base-equivalence "
f"to {_base_list(bases)} could not be proven"
)
elif branch not in bases:
reasons.append(
"issue lock worktree base-equivalence could not be proven; "
f"branch '{branch}' is not {_base_list(bases)}"
)
proven = not reasons
return _assessment(
proven,
reasons,
path,
branch or None,
dirty_files,
inspected_git_root=inspected_git_root,
base_branch=base_branch,
base_equivalent=base_equivalent,
)
def format_issue_lock_worktree_error(assessment: dict) -> str:
"""Format a single fail-closed error for ``gitea_lock_issue``."""
reasons = list(assessment.get("reasons") or [])
if not reasons:
reasons = ["issue lock worktree validation failed"]
return "; ".join(reasons) + " (fail closed)"
def verify_pr_worktree_matches_lock(
locked_worktree_path: str | None,
declared_worktree_path: str | None,
project_root: str,
) -> dict:
"""PR creation must use the same worktree the lock was validated against."""
locked = (locked_worktree_path or "").strip()
if not locked:
return {"proven": True, "block": False, "reasons": []}
declared = resolve_author_worktree_path(declared_worktree_path, project_root)
locked_real = os.path.realpath(locked)
declared_real = os.path.realpath(declared)
if locked_real != declared_real:
return {
"proven": False,
"block": True,
"reasons": [
f"PR worktree '{declared_real}' does not match locked worktree "
f"'{locked_real}' (fail closed)"
],
"locked_worktree_path": locked_real,
"declared_worktree_path": declared_real,
}
return {
"proven": True,
"block": False,
"reasons": [],
"locked_worktree_path": locked_real,
"declared_worktree_path": declared_real,
}
def _base_list(bases: frozenset[str]) -> str:
return "/".join(sorted(bases))
def _assessment(
proven: bool,
reasons: list[str],
worktree_path: str,
current_branch: str | None,
dirty_files: list[str],
*,
inspected_git_root: str | None = None,
base_branch: str | None = None,
base_equivalent: bool | None = None,
) -> dict:
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"worktree_path": worktree_path or None,
"inspected_git_root": inspected_git_root,
"current_branch": current_branch,
"dirty_files": dirty_files,
"base_branch": base_branch,
"base_equivalent": base_equivalent,
}
def _find_matching_base_ref(
path: str,
head_sha: str | None,
extra_bases: tuple[str, ...] | list[str] = (),
) -> tuple[str | None, str | None]:
"""Return the stable branch ref whose commit matches HEAD, if any.
Normal base branches (master/main/dev) are always considered. ``extra_bases``
adds explicitly-approved stacked bases; each is checked as a local ref and via
the ``prgs``/``origin`` remotes.
"""
if not head_sha:
return None, None
candidates: list[str] = []
for branch in sorted(BASE_BRANCHES):
candidates.extend((f"origin/{branch}", branch))
for branch in extra_bases:
name = (branch or "").strip()
if name:
candidates.extend((f"prgs/{name}", f"origin/{name}", name))
for ref in candidates:
res = subprocess.run(
["git", "-C", path, "rev-parse", "--verify", ref],
capture_output=True,
text=True,
check=False,
)
sha = (res.stdout or "").strip()
if res.returncode == 0 and sha == head_sha:
return ref, sha
return None, None
+180
View File
@@ -0,0 +1,180 @@
"""Early duplicate-work detection for author work-issue sessions (#400)."""
from __future__ import annotations
from typing import Any
import issue_claim_heartbeat as claim_hb
PHASE_LOCK = "lock_issue"
PHASE_COMMIT = "commit"
PHASE_PUSH = "push"
PHASE_CREATE_PR = "create_pr"
OUTCOME_DUPLICATE_PR_PREVENTED = "duplicate_pr_prevented"
OUTCOME_DUPLICATE_BRANCH_PREVENTED = "duplicate_branch_prevented"
OUTCOME_DUPLICATE_COMMIT_PREVENTED = "duplicate_commit_prevented"
OUTCOME_DUPLICATE_WORK_NOT_PREVENTED = "duplicate_work_not_prevented"
_ACTIVE_CLAIM_STATUSES = frozenset({"active", "awaiting_review"})
def _issue_pattern(issue_number: int) -> str:
return f"issue-{int(issue_number)}"
def _linked_open_pr(issue_number: int, open_prs: list[dict]) -> dict | None:
return claim_hb._linked_open_pr(issue_number, open_prs)
def _matching_branches(
issue_number: int,
branch_names: list[str],
*,
locked_branch: str | None = None,
) -> list[str]:
pattern = _issue_pattern(issue_number)
matches = [
name for name in (branch_names or [])
if pattern in (name or "").lower()
]
if locked_branch:
locked = locked_branch.strip()
matches = [name for name in matches if name != locked]
return matches
def assess_work_issue_duplicate_gate(
issue_number: int,
*,
open_prs: list[dict] | None = None,
branch_names: list[str] | None = None,
claim_entry: dict | None = None,
locked_branch: str | None = None,
phase: str = PHASE_LOCK,
) -> dict[str, Any]:
"""Fail closed when duplicate work is already in flight for an issue."""
reasons: list[str] = []
outcome = OUTCOME_DUPLICATE_WORK_NOT_PREVENTED
prs = list(open_prs or [])
branches = list(branch_names or [])
pattern = _issue_pattern(issue_number)
linked = _linked_open_pr(issue_number, prs)
if linked:
reasons.append(
f"open PR #{linked.get('number')} already covers issue "
f"#{issue_number} (fail closed)"
)
outcome = OUTCOME_DUPLICATE_PR_PREVENTED
conflicting_branches = _matching_branches(
issue_number, branches, locked_branch=locked_branch
)
if conflicting_branches:
names = ", ".join(conflicting_branches[:5])
reasons.append(
f"remote branch(es) already match issue pattern '{pattern}': "
f"{names} (fail closed)"
)
if outcome == OUTCOME_DUPLICATE_WORK_NOT_PREVENTED:
outcome = OUTCOME_DUPLICATE_BRANCH_PREVENTED
entry = claim_entry or {}
if entry.get("linked_open_pr") and not linked:
reasons.append(
f"claim inventory reports open PR #{entry['linked_open_pr']} "
f"for issue #{issue_number} (fail closed)"
)
outcome = OUTCOME_DUPLICATE_PR_PREVENTED
status = (entry.get("status") or "").strip().lower()
if status in _ACTIVE_CLAIM_STATUSES and not linked:
heartbeat = entry.get("latest_heartbeat") or {}
claim_branch = (heartbeat.get("branch") or "").strip()
if locked_branch and claim_branch and claim_branch != locked_branch:
reasons.append(
f"active claim lease on branch '{claim_branch}' blocks "
f"work on '{locked_branch}' for issue #{issue_number} "
"(fail closed)"
)
if outcome == OUTCOME_DUPLICATE_WORK_NOT_PREVENTED:
outcome = OUTCOME_DUPLICATE_BRANCH_PREVENTED
elif not locked_branch and status == "active":
reasons.append(
f"issue #{issue_number} has an active claim lease "
"(fail closed)"
)
if outcome == OUTCOME_DUPLICATE_WORK_NOT_PREVENTED:
outcome = OUTCOME_DUPLICATE_BRANCH_PREVENTED
if phase in {PHASE_COMMIT, PHASE_PUSH} and reasons:
if outcome == OUTCOME_DUPLICATE_PR_PREVENTED:
outcome = OUTCOME_DUPLICATE_COMMIT_PREVENTED
elif outcome == OUTCOME_DUPLICATE_BRANCH_PREVENTED:
outcome = OUTCOME_DUPLICATE_COMMIT_PREVENTED
block = bool(reasons)
return {
"block": block,
"performed": not block,
"issue_number": issue_number,
"phase": phase,
"outcome": outcome,
"linked_open_pr": linked.get("number") if linked else entry.get("linked_open_pr"),
"conflicting_branches": conflicting_branches,
"claim_status": status or None,
"reasons": reasons,
"safe_next_action": (
"stop before mutating; preserve local work and produce a "
"reconciliation handoff if a concurrent PR appeared after push"
if block and phase == PHASE_CREATE_PR
else "stop before mutating; do not commit or push duplicate work"
if block
else "proceed"
),
}
def assess_work_issue_duplicate_report(report_text: str) -> dict[str, Any]:
"""Require explicit duplicate-work outcome wording in work-issue reports."""
text = (report_text or "").lower()
markers = {
OUTCOME_DUPLICATE_PR_PREVENTED: (
"duplicate pr prevented",
"duplicate_pr_prevented",
),
OUTCOME_DUPLICATE_BRANCH_PREVENTED: (
"duplicate branch prevented",
"duplicate_branch_prevented",
),
OUTCOME_DUPLICATE_COMMIT_PREVENTED: (
"duplicate commit prevented",
"duplicate_commit_prevented",
),
OUTCOME_DUPLICATE_WORK_NOT_PREVENTED: (
"duplicate work not prevented",
"duplicate_work_not_prevented",
"no duplicate work",
),
}
matched = [
key for key, phrases in markers.items()
if any(phrase in text for phrase in phrases)
]
if len(matched) != 1:
return {
"complete": False,
"downgraded": True,
"reasons": [
"work-issue report must state exactly one duplicate-work "
"outcome (duplicate PR/branch/commit prevented, or "
"duplicate work not prevented)"
],
}
return {
"complete": True,
"downgraded": False,
"outcome": matched[0],
"reasons": [],
}
+259
View File
@@ -0,0 +1,259 @@
#!/usr/bin/env python3
"""MCP Discoverability Validation Tool for external servers (Issue #155)."""
import os
import sys
import json
import argparse
import subprocess
EXPECTED_JENKINS_TOOLS = {
"jenkins_whoami",
"jenkins_list_jobs",
"jenkins_latest_build",
"jenkins_build_status",
"jenkins_get_build",
}
EXPECTED_GLITCHTIP_TOOLS = {
"glitchtip_whoami",
"glitchtip_list_projects",
"glitchtip_list_unresolved",
"glitchtip_get_issue",
"glitchtip_recent_events",
"glitchtip_search",
}
RELOAD_INSTRUCTIONS = """
=== MCP CLIENT RELOAD/RECONNECT RUNBOOK ===
After registering or changing external MCP servers, reload your client to discover the new tools:
- Codex: Click 'Reload Developer Tools' or restart the editor.
- Gemini / Grok / ChatGPT Desktop: Restart the client or run the reload slash command if available.
- Claude Desktop: Use 'Developer -> Reload' or restart the app.
- General MCP Clients: Restart the process or reload the server config.
===========================================
"""
def parse_gitea_mcp_config(path):
if not path or not os.path.exists(path):
return {}
with open(path, "r", encoding="utf-8") as fh:
try:
return json.load(fh)
except Exception:
return {}
def read_json_rpc_response(proc, req_id):
import time
start_time = time.time()
while time.time() - start_time < 5.0:
line = proc.stdout.readline()
if not line:
break
try:
data = json.loads(line)
if data.get("id") == req_id:
return data
except json.JSONDecodeError:
continue
return None
def query_live_tools(command, args, env):
run_env = os.environ.copy()
if env:
run_env.update(env)
proc = subprocess.Popen(
[command] + args,
stdin=subprocess.PIPE,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
env=run_env,
text=True,
bufsize=1
)
tools = []
try:
# 1. Send initialize
init_req = {
"jsonrpc": "2.0",
"id": 1,
"method": "initialize",
"params": {
"protocolVersion": "2024-11-05",
"capabilities": {},
"clientInfo": {"name": "mcp-discoverability-check", "version": "1.0.0"}
}
}
proc.stdin.write(json.dumps(init_req) + "\n")
proc.stdin.flush()
# Read init response
init_resp = read_json_rpc_response(proc, 1)
if init_resp:
# Send initialized notification
init_notif = {
"jsonrpc": "2.0",
"method": "notifications/initialized"
}
proc.stdin.write(json.dumps(init_notif) + "\n")
proc.stdin.flush()
# 2. Send tools/list
tools_req = {
"jsonrpc": "2.0",
"id": 2,
"method": "tools/list",
"params": {}
}
proc.stdin.write(json.dumps(tools_req) + "\n")
proc.stdin.flush()
tools_resp = read_json_rpc_response(proc, 2)
if tools_resp and "result" in tools_resp and "tools" in tools_resp["result"]:
for t in tools_resp["result"]["tools"]:
tools.append(t["name"])
except Exception as e:
print(f"Error querying live tools: {e}", file=sys.stderr)
finally:
proc.terminate()
try:
proc.wait(timeout=2)
except Exception:
proc.kill()
return set(tools)
def validate_mcp_client_config(client_config_path, gitea_config_path=None, live=False):
if not client_config_path or not os.path.exists(client_config_path):
print(f"SKIPPED: MCP client config not found at '{client_config_path}'", file=sys.stderr)
return True
with open(client_config_path, "r", encoding="utf-8") as fh:
try:
config_data = json.load(fh)
except Exception as e:
print(f"Error parsing client config: {e}", file=sys.stderr)
return False
mcp_servers = config_data.get("mcpServers", {})
# Check for stale server names
stale_names = {"jenkins-readonly", "glitchtip-readonly"}
for name in mcp_servers:
if name in stale_names:
print(f"ERROR: Stale server name '{name}' configured. Use canonical names 'jenkins-mcp' or 'glitchtip-mcp'.", file=sys.stderr)
return False
gitea_data = parse_gitea_mcp_config(gitea_config_path)
enabled_services = set()
contexts = gitea_data.get("contexts", {})
profile_name = os.environ.get("GITEA_MCP_PROFILE")
if profile_name and "profiles" in gitea_data:
profile = gitea_data["profiles"].get(profile_name)
if profile and "context" in profile:
ctx_name = profile["context"]
ctx = contexts.get(ctx_name, {})
if ctx.get("enabled"):
services = ctx.get("services", {})
for s_name, s_data in services.items():
if s_data.get("enabled"):
enabled_services.add(s_name)
else:
for ctx_name, ctx in contexts.items():
if ctx.get("enabled"):
services = ctx.get("services", {})
for s_name, s_data in services.items():
if s_data.get("enabled"):
enabled_services.add(s_name)
if not enabled_services:
print("No external services enabled in Gitea contexts. Discoverability check complete.", file=sys.stderr)
return True
success = True
for service in enabled_services:
canonical_name = f"{service}-mcp"
if canonical_name not in mcp_servers:
stale_match = f"{service}-readonly"
if stale_match in mcp_servers:
print(f"ERROR: Server '{canonical_name}' references stale name '{stale_match}' (fail closed).", file=sys.stderr)
success = False
continue
print(f"ERROR: Enabled service '{service}' is not registered under canonical name '{canonical_name}' in client config.", file=sys.stderr)
success = False
continue
server_conf = mcp_servers[canonical_name]
command = server_conf.get("command")
args = server_conf.get("args") or []
env = server_conf.get("env") or {}
if not command:
print(f"ERROR: Server '{canonical_name}' has no command configured.", file=sys.stderr)
success = False
continue
expected_module = f"{service}_mcp"
if "-m" not in args or expected_module not in args:
print(f"ERROR: Server '{canonical_name}' args do not point to expected module '{expected_module}' (args: {args}).", file=sys.stderr)
success = False
continue
profile_var = f"{service.upper()}_MCP_PROFILE"
config_var = f"{service.upper()}_MCP_CONFIG"
if profile_var not in env or config_var not in env:
print(f"ERROR: Server '{canonical_name}' env is missing required variables '{profile_var}' or '{config_var}'.", file=sys.stderr)
success = False
continue
if live:
tools = query_live_tools(command, args, env)
if not tools:
print("SKIPPED: server enabled but no usable tools visible", file=sys.stdout)
success = False
continue
expected = EXPECTED_JENKINS_TOOLS if service == "jenkins" else EXPECTED_GLITCHTIP_TOOLS
missing = expected - tools
if missing:
print(f"ERROR: Server '{canonical_name}' is missing expected tools: {', '.join(missing)}", file=sys.stderr)
success = False
else:
print(f"SUCCESS: Server '{canonical_name}' discoverability verified.", file=sys.stderr)
else:
print(f"SUCCESS: Server '{canonical_name}' static registration verified.", file=sys.stderr)
return success
def main():
parser = argparse.ArgumentParser(description="MCP client registration discoverability checks.")
parser.add_argument("--client-config", help="Path to MCP client config JSON file.")
parser.add_argument("--gitea-config", help="Path to Gitea MCP config JSON file.")
parser.add_argument("--live", action="store_true", help="Perform live stdio checks on configured servers.")
parser.add_argument("--runbook", action="store_true", help="Print reload/reconnect guide runbook instructions.")
args = parser.parse_args()
if args.runbook:
print(RELOAD_INSTRUCTIONS)
return 0
if not args.client_config:
print("ERROR: --client-config must be specified.", file=sys.stderr)
return 2
ok = validate_mcp_client_config(
client_config_path=args.client_config,
gitea_config_path=args.gitea_config,
live=args.live
)
if not ok:
print(RELOAD_INSTRUCTIONS, file=sys.stderr)
return 1
return 0
if __name__ == "__main__":
sys.exit(main())
+41 -3806
View File
File diff suppressed because it is too large Load Diff
+61
View File
@@ -0,0 +1,61 @@
"""Merge approval must pin the current PR head SHA (#471).
Formal APPROVED reviews that predate the live PR head must not satisfy
``gitea_merge_pr`` eligibility. Pure assessment helpers are isolated here
for hermetic unit tests apart from MCP HTTP calls.
"""
from __future__ import annotations
def assess_merge_approval_head(
*,
current_head_sha: str | None,
latest_by_reviewer: dict,
) -> dict:
"""Return whether a visible approval applies to the live PR head.
Args:
current_head_sha: Current PR head commit SHA.
latest_by_reviewer: Map of reviewer login → review entry dicts with
``verdict``, ``dismissed``, and ``reviewed_head_sha`` keys.
Returns:
dict with ``approval_at_current_head``, ``latest_approved_head_sha``,
and ``stale_approval_block_reason`` (set when merge must fail closed).
"""
current = (current_head_sha or "").strip()
approved_entries = [
entry
for entry in (latest_by_reviewer or {}).values()
if (entry.get("verdict") or "").upper() == "APPROVED"
and not entry.get("dismissed")
]
at_current = any(
(entry.get("reviewed_head_sha") or "").strip() == current
for entry in approved_entries
if current
)
latest_approved = None
if approved_entries:
latest_entry = sorted(
approved_entries,
key=lambda entry: (
entry.get("submitted_at") or "",
entry.get("reviewed_head_sha") or "",
),
)[-1]
latest_approved = (latest_entry.get("reviewed_head_sha") or "").strip() or None
reason = None
if approved_entries and not at_current:
reason = (
f"stale approval: approved SHA '{latest_approved}' does not match "
f"current live PR head SHA '{current or '(unknown)'}' (fail closed); "
"required next action: re-review PR at current head before merge"
)
return {
"approval_at_current_head": at_current,
"latest_approved_head_sha": latest_approved,
"stale_approval_block_reason": reason,
}
+329
View File
@@ -0,0 +1,329 @@
"""Merged PR branch and worktree cleanup reconciliation (#269).
Builds dry-run reports for merged pull requests and optionally executes
remote branch deletion and local worktree removal when every safety gate passes.
"""
from __future__ import annotations
import json
import os
import re
import subprocess
from typing import Any
from reviewer_worktree import parse_dirty_tracked_files
import issue_lock_store
PROTECTED_BRANCHES = frozenset({"master", "main", "dev"})
CLOSES_FIXES_RE = re.compile(r"\b(?:closes|fixes)\s+#(\d+)\b", re.IGNORECASE)
def extract_linked_issue(title: str, body: str) -> int | None:
"""Return the first Closes/Fixes issue number from PR metadata."""
for text in (title or "", body or ""):
match = CLOSES_FIXES_RE.search(text)
if match:
return int(match.group(1))
return None
def branch_worktree_folder(branch: str) -> str:
return (branch or "").replace("/", "-")
def resolve_worktree_path(project_root: str, branch: str) -> str:
return os.path.join(project_root, "branches", branch_worktree_folder(branch))
def read_issue_lock(path: str | None = None) -> dict[str, Any] | None:
if path:
return issue_lock_store.read_lock_file(path.strip())
return issue_lock_store.read_session_issue_lock()
def has_active_issue_lock(branch: str, lock_path: str | None = None) -> bool:
if lock_path:
lock = issue_lock_store.read_lock_file(lock_path.strip())
if not lock:
return False
return (lock.get("branch_name") or "").strip() == (branch or "").strip()
return issue_lock_store.has_active_issue_lock(branch)
def collect_open_pr_heads(open_prs: list[dict[str, Any]]) -> set[str]:
heads: set[str] = set()
for pr in open_prs or []:
head = pr.get("head")
if isinstance(head, dict):
ref = head.get("ref")
else:
ref = head
if ref:
heads.add(str(ref))
return heads
def read_local_worktree_state(worktree_path: str) -> dict[str, Any]:
path = (worktree_path or "").strip()
if not path or not os.path.isdir(path):
return {
"exists": False,
"clean": None,
"current_branch": None,
"head_sha": None,
"dirty_files": [],
}
branch_res = subprocess.run(
["git", "-C", path, "branch", "--show-current"],
capture_output=True,
text=True,
check=False,
)
current_branch = (branch_res.stdout or "").strip() or None
status_res = subprocess.run(
["git", "-C", path, "status", "--porcelain"],
capture_output=True,
text=True,
check=False,
)
dirty_files = parse_dirty_tracked_files(status_res.stdout or "")
head_res = subprocess.run(
["git", "-C", path, "rev-parse", "HEAD"],
capture_output=True,
text=True,
check=False,
)
head_sha = (head_res.stdout or "").strip() if head_res.returncode == 0 else None
return {
"exists": True,
"clean": not dirty_files,
"current_branch": current_branch,
"head_sha": head_sha,
"dirty_files": dirty_files,
}
def is_head_ancestor_of_ref(project_root: str, head_sha: str | None, base_ref: str) -> bool | None:
if not head_sha:
return None
res = subprocess.run(
["git", "-C", project_root, "merge-base", "--is-ancestor", head_sha, base_ref],
capture_output=True,
text=True,
check=False,
)
if res.returncode == 0:
return True
if res.returncode == 1:
return False
return None
def assess_remote_branch_cleanup(
*,
pr_number: int,
head_branch: str,
merged: bool,
remote_branch_exists: bool,
open_pr_heads: set[str],
protected_branches: frozenset[str] | None = None,
head_on_master: bool | None,
delete_capability_allowed: bool,
active_lock: bool,
) -> dict[str, Any]:
protected = protected_branches or PROTECTED_BRANCHES
reasons: list[str] = []
if not merged:
reasons.append("PR is not merged")
if not remote_branch_exists:
reasons.append("remote branch already absent")
if head_branch in protected:
reasons.append(f"branch '{head_branch}' is protected")
if head_branch in open_pr_heads:
reasons.append("an open PR still references this head branch")
if active_lock:
reasons.append("active issue lock references this branch")
if head_on_master is False:
reasons.append("PR head is not an ancestor of master")
if not delete_capability_allowed:
reasons.append("delete_branch capability is not allowed in the active profile")
safe = merged and remote_branch_exists and not reasons
return {
"pr_number": pr_number,
"head_branch": head_branch,
"remote_branch_exists": remote_branch_exists,
"safe_to_delete_remote": safe,
"block_reasons": reasons,
"recommended_action": "delete_remote_branch" if safe else "keep_remote_branch",
}
def assess_local_worktree_cleanup(
*,
pr_number: int,
head_branch: str,
merged: bool,
worktree_state: dict[str, Any],
active_lock: bool,
) -> dict[str, Any]:
reasons: list[str] = []
exists = bool(worktree_state.get("exists"))
if not merged:
reasons.append("PR is not merged")
if not exists:
reasons.append("local worktree not present")
if exists and worktree_state.get("clean") is False:
dirty = worktree_state.get("dirty_files") or []
reasons.append(
"local worktree has tracked edits"
+ (f" ({', '.join(dirty)})" if dirty else "")
)
if exists:
current_branch = worktree_state.get("current_branch")
if current_branch and current_branch != head_branch:
reasons.append(
f"worktree branch '{current_branch}' does not match PR head '{head_branch}'"
)
if active_lock:
reasons.append("active issue lock references this branch")
safe = merged and exists and not reasons
return {
"pr_number": pr_number,
"head_branch": head_branch,
"worktree_path": worktree_state.get("worktree_path"),
"worktree_exists": exists,
"worktree_clean": worktree_state.get("clean"),
"safe_to_remove_worktree": safe,
"block_reasons": reasons,
"recommended_action": "remove_local_worktree" if safe else "keep_local_worktree",
}
def build_pr_cleanup_entry(
*,
pr: dict[str, Any],
project_root: str,
open_pr_heads: set[str],
remote_branch_exists: bool,
head_on_master: bool | None,
delete_capability_allowed: bool,
issue_lock_path: str | None = None,
protected_branches: frozenset[str] | None = None,
) -> dict[str, Any]:
pr_number = int(pr["number"])
head_branch = pr.get("head") or ""
if isinstance(head_branch, dict):
head_branch = head_branch.get("ref") or ""
title = pr.get("title") or ""
body = pr.get("body") or ""
merged = bool(pr.get("merged_at"))
worktree_path = resolve_worktree_path(project_root, head_branch)
worktree_state = read_local_worktree_state(worktree_path)
worktree_state["worktree_path"] = worktree_path
active_lock = has_active_issue_lock(head_branch, issue_lock_path)
remote = assess_remote_branch_cleanup(
pr_number=pr_number,
head_branch=head_branch,
merged=merged,
remote_branch_exists=remote_branch_exists,
open_pr_heads=open_pr_heads,
protected_branches=protected_branches,
head_on_master=head_on_master,
delete_capability_allowed=delete_capability_allowed,
active_lock=active_lock,
)
local = assess_local_worktree_cleanup(
pr_number=pr_number,
head_branch=head_branch,
merged=merged,
worktree_state=worktree_state,
active_lock=active_lock,
)
return {
"pr_number": pr_number,
"issue_number": extract_linked_issue(title, body),
"title": title,
"head_branch": head_branch,
"merge_commit_sha": pr.get("merge_commit_sha"),
"merged_at": pr.get("merged_at"),
"merged": merged,
"remote_branch": remote,
"local_worktree": local,
}
def build_reconciliation_report(
*,
project_root: str,
closed_prs: list[dict[str, Any]],
open_prs: list[dict[str, Any]],
remote_branch_exists: dict[str, bool],
head_on_master: dict[int, bool | None],
delete_capability_allowed: bool,
issue_lock_path: str | None = None,
protected_branches: frozenset[str] | None = None,
) -> dict[str, Any]:
open_heads = collect_open_pr_heads(open_prs)
entries: list[dict[str, Any]] = []
for pr in closed_prs or []:
if not pr.get("merged_at"):
continue
pr_number = int(pr["number"])
head = pr.get("head") or ""
if isinstance(head, dict):
head = head.get("ref") or ""
entries.append(
build_pr_cleanup_entry(
pr=pr,
project_root=project_root,
open_pr_heads=open_heads,
remote_branch_exists=bool(remote_branch_exists.get(head)),
head_on_master=head_on_master.get(pr_number),
delete_capability_allowed=delete_capability_allowed,
issue_lock_path=issue_lock_path,
protected_branches=protected_branches,
)
)
return {
"project_root": os.path.realpath(project_root),
"merged_pr_count": len(entries),
"entries": entries,
"dry_run": True,
"executed": False,
}
def remove_local_worktree(project_root: str, branch: str) -> dict[str, Any]:
worktree_path = resolve_worktree_path(project_root, branch)
if not os.path.isdir(worktree_path):
return {
"success": False,
"performed": False,
"message": f"worktree not found: {worktree_path}",
}
res = subprocess.run(
["git", "-C", project_root, "worktree", "remove", worktree_path],
capture_output=True,
text=True,
check=False,
)
if res.returncode != 0:
return {
"success": False,
"performed": False,
"message": (res.stderr or res.stdout or "worktree remove failed").strip(),
}
return {
"success": True,
"performed": True,
"message": f"removed worktree {worktree_path}",
"worktree_path": worktree_path,
}
+2
View File
@@ -31,6 +31,8 @@ REVIEWER_DEFAULT_FORBIDDEN = ["branch", "commit", "push", "open_pr"]
def infer_role(name, execution_profile):
"""Return the unambiguous role for a legacy profile name, or None."""
haystack = f"{name} {execution_profile or ''}".lower()
if "reconciler" in haystack:
return "reconciler"
has_author = "author" in haystack
has_reviewer = "reviewer" in haystack
if has_author == has_reviewer:
+307
View File
@@ -0,0 +1,307 @@
"""Namespace-scoped MCP workspace binding (#510).
Each role namespace (author, reviewer, merger, reconciler) resolves its own
active task workspace. Foreign role worktree environment variables must not
poison workspace purity checks in another namespace.
"""
from __future__ import annotations
import os
import author_mutation_worktree as amw
ACTIVE_WORKTREE_ENV = amw.ACTIVE_WORKTREE_ENV
AUTHOR_WORKTREE_ENV = amw.AUTHOR_WORKTREE_ENV
REVIEWER_WORKTREE_ENV = "GITEA_REVIEWER_WORKTREE"
MERGER_WORKTREE_ENV = "GITEA_MERGER_WORKTREE"
RECONCILER_WORKTREE_ENV = "GITEA_RECONCILER_WORKTREE"
ROLE_WORKTREE_ENVS: dict[str, str] = {
"author": AUTHOR_WORKTREE_ENV,
"reviewer": REVIEWER_WORKTREE_ENV,
"merger": MERGER_WORKTREE_ENV,
"reconciler": RECONCILER_WORKTREE_ENV,
}
NON_AUTHOR_ROLES = frozenset({"reviewer", "merger", "reconciler"})
def normalize_role_kind(
role_kind: str | None,
*,
profile_name: str | None = None,
) -> str:
"""Map profile/task role to a workspace namespace key."""
role = (role_kind or "author").strip().lower()
profile = (profile_name or "").strip().lower()
if role == "reviewer" and "merger" in profile:
return "merger"
if role in ROLE_WORKTREE_ENVS:
return role
return "author"
def _env_value(env: dict[str, str] | os._Environ, key: str) -> str | None:
text = (env.get(key) or "").strip()
return text or None
def resolve_namespace_workspace(
*,
role_kind: str,
worktree_path: str | None = None,
worktree: str | None = None,
process_project_root: str,
env: dict[str, str] | os._Environ | None = None,
session_lease_worktree: str | None = None,
profile_name: str | None = None,
) -> tuple[str, str]:
"""Return ``(resolved_path, binding_source)`` for *role_kind*."""
env_map = env if env is not None else os.environ
role = normalize_role_kind(role_kind, profile_name=profile_name)
role_env_key = ROLE_WORKTREE_ENVS[role]
for candidate, source in (
(worktree_path, "worktree_path argument"),
(worktree, "worktree argument"),
(_env_value(env_map, ACTIVE_WORKTREE_ENV), f"{ACTIVE_WORKTREE_ENV} environment variable"),
(_env_value(env_map, role_env_key), f"{role_env_key} environment variable"),
(session_lease_worktree if role in {"reviewer", "merger"} else None,
"reviewer PR lease worktree"),
):
text = (candidate or "").strip()
if text:
return os.path.realpath(os.path.abspath(text)), source
return os.path.realpath(process_project_root), "MCP server process root (default)"
def resolve_namespace_mutation_context(
*,
role_kind: str,
worktree_path: str | None,
process_project_root: str,
env: dict[str, str] | os._Environ | None = None,
session_lease_worktree: str | None = None,
worktree: str | None = None,
profile_name: str | None = None,
) -> dict:
"""Shared workspace resolution for runtime_context and mutation guards."""
workspace, binding_source = resolve_namespace_workspace(
role_kind=role_kind,
worktree_path=worktree_path,
worktree=worktree,
process_project_root=process_project_root,
env=env,
session_lease_worktree=session_lease_worktree,
profile_name=profile_name,
)
process_root = os.path.realpath(process_project_root)
role = normalize_role_kind(role_kind, profile_name=profile_name)
pollution = assess_foreign_role_worktree_pollution(
role_kind=role,
resolved_workspace=workspace,
binding_source=binding_source,
env=env,
profile_name=profile_name,
)
canonical_root = amw.resolve_canonical_repo_root(process_root, process_root)
return {
"workspace_path": workspace,
"workspace_binding_source": binding_source,
"workspace_role_kind": role,
"ignored_bindings": pollution.get("ignored_bindings") or [],
"process_project_root": process_root,
"canonical_repo_root": canonical_root,
"roots_aligned": canonical_root == process_root,
}
def assess_foreign_role_worktree_pollution(
*,
role_kind: str,
resolved_workspace: str,
binding_source: str,
env: dict[str, str] | os._Environ | None = None,
profile_name: str | None = None,
) -> dict:
"""Detect when a foreign role env would have hijacked workspace binding."""
env_map = env if env is not None else os.environ
role = normalize_role_kind(role_kind, profile_name=profile_name)
if role == "author":
return {"would_pollute": False, "ignored_bindings": []}
ignored: list[str] = []
author_path = _env_value(env_map, AUTHOR_WORKTREE_ENV)
if author_path:
author_real = os.path.realpath(os.path.abspath(author_path))
resolved_real = os.path.realpath(resolved_workspace)
if author_real != resolved_real and binding_source != f"{AUTHOR_WORKTREE_ENV} environment variable":
ignored.append(
f"{AUTHOR_WORKTREE_ENV}={author_real} (ignored for {role} namespace)"
)
return {
"would_pollute": bool(ignored),
"ignored_bindings": ignored,
}
def assess_metadata_only_worktree_binding(
*,
role_kind: str,
declared_worktree_path: str | None,
mutation_workspace: str,
process_project_root: str,
profile_name: str | None = None,
) -> dict:
"""Fail closed when declared worktree_path would not redirect mutations."""
declared = (declared_worktree_path or "").strip()
process_root = os.path.realpath(process_project_root)
mutation_root = os.path.realpath(mutation_workspace)
role = normalize_role_kind(role_kind, profile_name=profile_name)
if not declared:
return {"block": False, "reasons": [], "metadata_only": False}
declared_root = os.path.realpath(os.path.abspath(declared))
if declared_root == mutation_root:
return {"block": False, "reasons": [], "metadata_only": False}
if declared_root != process_root and mutation_root == process_root:
return {
"block": True,
"metadata_only": True,
"reasons": [
f"worktree_path is metadata-only for {role} mutations: preflight "
f"inspected '{declared_root}' but mutation tools would still "
f"validate MCP server process root '{process_root}'"
],
"declared_worktree_path": declared_root,
"mutation_workspace": mutation_root,
"process_project_root": process_root,
}
return {"block": False, "reasons": [], "metadata_only": False}
def format_namespace_workspace_binding_error(
*,
role_kind: str,
workspace_path: str,
binding_source: str,
reasons: list[str] | None = None,
ignored_bindings: list[str] | None = None,
dirty_files: list[str] | None = None,
) -> str:
"""Canonical error when namespace workspace binding blocks mutations."""
role = normalize_role_kind(role_kind)
workspace = os.path.realpath(workspace_path)
parts = [
f"Namespace workspace binding blocked ({role} namespace, #510): "
f"resolved workspace '{workspace}' via {binding_source}."
]
if ignored_bindings:
parts.append(
"Foreign role bindings ignored: " + "; ".join(ignored_bindings) + "."
)
if dirty_files:
parts.append(
"Dirty tracked files in active task workspace: "
+ ", ".join(dirty_files)
+ "."
)
if reasons:
parts.append("Details: " + "; ".join(reasons) + ".")
parts.append(
"Remediation: reconnect or relaunch the MCP server from a clean dedicated "
f"branches/ {role} worktree, set "
f"{ROLE_WORKTREE_ENVS.get(role, ACTIVE_WORKTREE_ENV)} or {ACTIVE_WORKTREE_ENV} "
"to that path, or pass worktree_path on mutation tools. Do not clean or "
"reset foreign role worktrees to unblock this namespace."
)
return " ".join(parts)
def assess_namespace_mutation_workspace(
*,
role_kind: str,
worktree_path: str | None,
worktree: str | None,
process_project_root: str,
env: dict[str, str] | os._Environ | None = None,
session_lease_worktree: str | None = None,
profile_name: str | None = None,
current_branch: str | None = None,
) -> dict:
"""Evaluate namespace workspace binding before preflight/mutation."""
ctx = resolve_namespace_mutation_context(
role_kind=role_kind,
worktree_path=worktree_path,
worktree=worktree,
process_project_root=process_project_root,
env=env,
session_lease_worktree=session_lease_worktree,
profile_name=profile_name,
)
mutation_workspace = ctx["workspace_path"]
binding_source = ctx["workspace_binding_source"]
role = ctx["workspace_role_kind"]
process_root = ctx["process_project_root"]
metadata = assess_metadata_only_worktree_binding(
role_kind=role,
declared_worktree_path=worktree_path,
mutation_workspace=mutation_workspace,
process_project_root=process_root,
profile_name=profile_name,
)
pollution = assess_foreign_role_worktree_pollution(
role_kind=role,
resolved_workspace=mutation_workspace,
binding_source=binding_source,
env=env,
profile_name=profile_name,
)
reasons = list(metadata.get("reasons") or [])
if role == "author":
branches = amw.assess_author_mutation_worktree(
workspace_path=mutation_workspace,
project_root=ctx["canonical_repo_root"],
current_branch=current_branch,
)
if branches["block"]:
reasons.extend(branches["reasons"])
elif (
role == "reviewer"
and mutation_workspace == process_root
and not amw.is_path_under_branches(mutation_workspace, ctx["canonical_repo_root"])
):
reasons.append(
f"{role} mutation blocked: workspace is the stable control checkout; "
f"create or reconnect to a session-owned worktree under branches/ "
f"or set {ROLE_WORKTREE_ENVS[role]} / {ACTIVE_WORKTREE_ENV}"
)
elif (
role in {"reviewer", "merger"}
and mutation_workspace != process_root
and not amw.is_path_under_branches(mutation_workspace, ctx["canonical_repo_root"])
):
reasons.append(
f"{role} mutation blocked: workspace '{mutation_workspace}' is not under "
f"'{ctx['canonical_repo_root']}/branches/'"
)
block = bool(reasons)
return {
"block": block,
"reasons": reasons,
"mutation_workspace": mutation_workspace,
"workspace_binding_source": binding_source,
"workspace_role_kind": role,
"process_project_root": process_root,
"canonical_repo_root": ctx["canonical_repo_root"],
"metadata_only": metadata.get("metadata_only", False),
"declared_worktree_path": metadata.get("declared_worktree_path"),
"ignored_bindings": pollution.get("ignored_bindings") or [],
}
+369
View File
@@ -0,0 +1,369 @@
"""Native MCP preference gate and shell health circuit breaker (#270).
Gitea mutations must use native MCP tools first. Shell scripts, direct API
calls, browser helpers, and improvised encoders are blocked when MCP is
available unless explicit recovery-mode proof is supplied.
"""
from __future__ import annotations
import re
from typing import Any
from reviewer_fallback import LOCAL_GITEA_SCRIPT_NAMES
# Tasks that must prefer native MCP over shell/API/helper fallbacks.
GITEA_MUTATION_TASKS = frozenset({
"comment_issue",
"mark_issue",
"lock_issue",
"set_issue_labels",
"create_issue",
"close_issue",
"create_branch",
"push_branch",
"create_pr",
"close_pr",
"comment_pr",
"review_pr",
"merge_pr",
"approve_pr",
"request_changes_pr",
"commit_files",
"gitea_commit_files",
"delete_branch",
"address_pr_change_requests",
})
ALLOWED_PATH_KINDS = frozenset({
"mcp_native",
"recovery_fallback",
})
BLOCKED_PATH_KINDS = frozenset({
"shell_script",
"direct_api",
"webfetch",
"playwright",
"helper_script",
"unsafe_helper",
"mcp_server_touch",
})
SHELL_SPAWN_FAILURE_THRESHOLD = 2
TERMINAL_REPORT_HEADING = (
"MCP transport unavailable or shell circuit breaker tripped. "
"No unsafe Gitea fallback performed."
)
_RECOVERY_MODE_RE = re.compile(
r"\b(?:recovery mode|explicit recovery|mcp unavailable|mcp not available|"
r"mcp tools unavailable|no mcp path)\b",
re.IGNORECASE,
)
_LOCAL_SCRIPT_RE = re.compile(
r"(?:^|[\s\"'`/])(?:python3?|bash|sh)?\s*(?:"
+ "|".join(re.escape(name) for name in LOCAL_GITEA_SCRIPT_NAMES)
+ r")",
re.IGNORECASE,
)
_WEBFETCH_RE = re.compile(r"\b(?:webfetch|mcp_web_fetch|fetch\s+url)\b", re.IGNORECASE)
_PLAYWRIGHT_RE = re.compile(r"\b(?:playwright|browser_navigate|browser_click)\b", re.IGNORECASE)
_HELPER_SCRIPT_RE = re.compile(
r"(?:^|[\s\"'`/])(?:_encode_|_emit_|_inline_)[\w-]+\.py",
re.IGNORECASE,
)
_MANUAL_BASE64_RE = re.compile(
r"\b(?:manual\s+base64|llm-generated\s+base64|base64-encode\s+in\s+chat)\b",
re.IGNORECASE,
)
_DIRECT_API_RE = re.compile(
r"\b(?:api_request|urllib\.request|requests\.(?:post|patch|put)|curl\s+-X\s+POST)\b",
re.IGNORECASE,
)
_MCP_SERVER_TOUCH_RE = re.compile(
r"\b(?:kill|pkill|restart|reload|touch|edit|modify|write)\b.{0,40}\b(?:mcp_server|mcp-server|gitea_mcp_server)\b",
re.IGNORECASE,
)
_CLI_AUTH_DIVERGENCE_RE = re.compile(
r"GITEA_MCP_PROFILE\s*=\s*['\"]?([^\s'\";]+)",
re.IGNORECASE,
)
_session_shell: dict[str, Any] = {
"consecutive_spawn_failures": 0,
"shell_unavailable": False,
"hard_stopped": False,
"last_exit_code": None,
}
def _clean(value: str | None) -> str:
return (value or "").strip()
def is_spawn_failure(
*,
exit_code: int | None = None,
stdout: str | None = None,
stderr: str | None = None,
spawn_failure: bool | None = None,
) -> bool:
"""True for executor spawn failures (exit_code -1, empty output)."""
if spawn_failure is True:
return True
if spawn_failure is False:
return False
if exit_code != -1:
return False
return not (_clean(stdout) or _clean(stderr))
def record_shell_spawn_outcome(
*,
exit_code: int | None = None,
stdout: str | None = None,
stderr: str | None = None,
spawn_failure: bool | None = None,
probe_attempted: bool = False,
probe_succeeded: bool | None = None,
) -> dict[str, Any]:
"""Track shell spawn outcomes and trip the session circuit breaker (#270 AC4)."""
global _session_shell
failed = is_spawn_failure(
exit_code=exit_code,
stdout=stdout,
stderr=stderr,
spawn_failure=spawn_failure,
)
if failed:
_session_shell["consecutive_spawn_failures"] = (
int(_session_shell.get("consecutive_spawn_failures") or 0) + 1
)
_session_shell["last_exit_code"] = exit_code
if probe_attempted and probe_succeeded is False:
_session_shell["shell_unavailable"] = True
else:
_session_shell["consecutive_spawn_failures"] = 0
_session_shell["shell_unavailable"] = False
_session_shell["hard_stopped"] = False
_session_shell["last_exit_code"] = exit_code
failures = int(_session_shell["consecutive_spawn_failures"])
if failures >= SHELL_SPAWN_FAILURE_THRESHOLD:
_session_shell["shell_unavailable"] = True
_session_shell["hard_stopped"] = True
return shell_health_status()
def shell_health_status() -> dict[str, Any]:
"""Return the current shell health / circuit-breaker state."""
failures = int(_session_shell.get("consecutive_spawn_failures") or 0)
hard_stopped = bool(_session_shell.get("hard_stopped"))
shell_unavailable = bool(_session_shell.get("shell_unavailable"))
return {
"consecutive_spawn_failures": failures,
"shell_unavailable": shell_unavailable,
"hard_stopped": hard_stopped,
"threshold": SHELL_SPAWN_FAILURE_THRESHOLD,
"shell_use_allowed": not hard_stopped,
"last_exit_code": _session_shell.get("last_exit_code"),
"safe_next_action": (
"emit terminal recovery report; prefer native MCP for remaining Gitea mutations"
if hard_stopped
else (
"probe shell once with echo/pwd; if probe fails mark shell unavailable"
if failures == 1
else "shell healthy"
)
),
}
def clear_shell_health_for_tests() -> None:
"""Reset session shell state (tests only)."""
global _session_shell
_session_shell = {
"consecutive_spawn_failures": 0,
"shell_unavailable": False,
"hard_stopped": False,
"last_exit_code": None,
}
def classify_command_path(command_or_detail: str | None) -> str:
"""Classify a proposed command/detail into a path kind."""
text = command_or_detail or ""
if _MCP_SERVER_TOUCH_RE.search(text):
return "mcp_server_touch"
if _WEBFETCH_RE.search(text):
return "webfetch"
if _PLAYWRIGHT_RE.search(text):
return "playwright"
if _HELPER_SCRIPT_RE.search(text) or _MANUAL_BASE64_RE.search(text):
return "unsafe_helper"
if _LOCAL_SCRIPT_RE.search(text):
return "shell_script"
if _DIRECT_API_RE.search(text):
return "direct_api"
return "mcp_native"
def detect_cli_auth_divergence(
command_or_detail: str | None,
*,
active_profile: str | None,
) -> list[str]:
"""Flag shell commands that override GITEA_MCP_PROFILE away from the session."""
reasons: list[str] = []
text = command_or_detail or ""
active = _clean(active_profile)
match = _CLI_AUTH_DIVERGENCE_RE.search(text)
if not match or not active:
return reasons
requested = _clean(match.group(1))
if requested and requested != active:
reasons.append(
f"CLI auth divergence: command sets GITEA_MCP_PROFILE='{requested}' "
f"but active session profile is '{active}' (fail closed)"
)
return reasons
def format_mcp_unavailable_terminal_report(
*,
task: str | None = None,
reasons: list[str] | None = None,
) -> str:
"""Terminal report when MCP is broken and unsafe recovery is forbidden (#270 AC3)."""
lines = [TERMINAL_REPORT_HEADING, ""]
if task:
lines.append(f"Blocked task: {task}")
if reasons:
lines.extend(["", "Reasons:"])
lines.extend(f"- {reason}" for reason in reasons)
lines.extend([
"",
"Required recovery (no improvised fallback):",
"- Restart the MCP session",
"- Kill hung background terminals holding the shell executor",
"- Retry the native MCP tool once after reconnect",
"- If MCP remains unavailable, stop and hand off to the operator",
"- Do not run local Gitea scripts, WebFetch, Playwright, or manual base64 encoders",
])
return "\n".join(lines)
def assess_gitea_operation_path(
*,
task: str,
path_kind: str | None = None,
command_or_detail: str | None = None,
mcp_available: bool = True,
mcp_tool_visible: bool = True,
recovery_mode: bool = False,
recovery_proof_complete: bool = False,
role: str | None = None,
active_profile: str | None = None,
) -> dict[str, Any]:
"""Fail closed when a Gitea mutation would bypass native MCP (#270)."""
task_name = _clean(task)
resolved_kind = _clean(path_kind) or classify_command_path(command_or_detail)
reasons: list[str] = []
if task_name and task_name not in GITEA_MUTATION_TASKS:
reasons.append(
f"unknown or non-mutation Gitea task '{task_name}'; "
"native MCP preference gate applies only to Gitea mutations"
)
shell = shell_health_status()
if shell["hard_stopped"] and resolved_kind != "mcp_native":
reasons.append(
"shell circuit breaker is hard-stopped after consecutive spawn failures; "
"use native MCP or emit terminal recovery report"
)
recovery_declared = recovery_mode or bool(
_RECOVERY_MODE_RE.search(command_or_detail or "")
)
recovery_allowed = (
recovery_declared
and recovery_proof_complete
and not mcp_available
and resolved_kind in BLOCKED_PATH_KINDS - {"mcp_server_touch"}
)
if resolved_kind in BLOCKED_PATH_KINDS and not recovery_allowed:
reasons.append(f"path kind '{resolved_kind}' is not an approved Gitea mutation path")
if resolved_kind == "mcp_server_touch":
if _clean(role) == "reviewer" or (
active_profile and "reviewer" in active_profile.lower()
):
reasons.append(
"reviewer workflows must not touch, restart, or kill MCP server files/processes"
)
else:
reasons.append(
"MCP server files/processes must not be modified during normal Gitea workflows"
)
reasons.extend(
detect_cli_auth_divergence(command_or_detail, active_profile=active_profile)
)
if (
mcp_available
and mcp_tool_visible
and resolved_kind != "mcp_native"
and not recovery_allowed
):
if not recovery_declared:
reasons.append(
"native MCP tools are available; shell/API/helper fallback is forbidden"
)
elif not recovery_proof_complete:
reasons.append(
"recovery-mode fallback requires complete recovery proof before proceeding"
)
if not mcp_available and resolved_kind != "mcp_native" and not recovery_allowed:
if not recovery_declared:
reasons.append(
"MCP transport unavailable; produce terminal recovery report instead of "
"improvising unsafe fallback"
)
block = bool(reasons)
terminal_report = None
if block and (not mcp_available or shell["hard_stopped"]):
terminal_report = format_mcp_unavailable_terminal_report(
task=task_name or None,
reasons=reasons,
)
return {
"task": task_name or None,
"path_kind": resolved_kind,
"mcp_available": mcp_available,
"mcp_tool_visible": mcp_tool_visible,
"recovery_mode": recovery_declared,
"shell_health": shell,
"block": block,
"allowed": not block,
"reasons": reasons,
"terminal_report": terminal_report,
"safe_next_action": (
"use the native MCP tool for this task"
if mcp_available and mcp_tool_visible and block
else (
terminal_report or "proceed with native MCP"
if block
else "proceed with native MCP"
)
),
}
+239
View File
@@ -0,0 +1,239 @@
"""Post-merge cleanup proof verifier for reviewer final reports (#402)."""
from __future__ import annotations
import re
from typing import Any
CLEANUP_SKIPPED = "CLEANUP_SKIPPED"
CLEANUP_PERFORMED = "CLEANUP_PERFORMED"
_CLEANUP_SECTION_HINT = re.compile(
r"(?:cleanup (?:status|result|mutations)|post-merge cleanup|"
r"gitea_delete_branch|remote branch.*deleted|worktree remove)",
re.IGNORECASE,
)
_CLEANUP_SKIPPED_RE = re.compile(r"\bCLEANUP_SKIPPED\b", re.IGNORECASE)
_CLEANUP_BLOCKER_RE = re.compile(
r"(?:cleanup blocker|cleanup skip(?:ped)? reason)\s*:\s*(.+)",
re.IGNORECASE,
)
_REMOTE_DELETE_CLAIM_RE = re.compile(
r"(?:gitea_delete_branch|remote (?:head )?branch (?:was )?deleted|"
r"deleted remote branch|delete_branch)",
re.IGNORECASE,
)
_WORKTREE_REMOVE_CLAIM_RE = re.compile(
r"(?:git worktree remove|worktree (?:was )?removed|removed (?:local )?worktree|"
r"worktree cleanup performed)",
re.IGNORECASE,
)
_DELETE_CAPABILITY_RE = re.compile(
r"(?:delete[- ]branch capability resolved|gitea\.branch\.delete)\s*:\s*"
r".*(?:gitea\.branch\.delete|delete_branch).*(?:resolved|allowed|proven)|"
r"gitea\.branch\.delete\s+(?:resolved|allowed|proven)",
re.IGNORECASE,
)
_DELETE_TASK_RE = re.compile(
r"(?:delete_branch|cleanup_branch|reconcile_merged_cleanups)",
re.IGNORECASE,
)
_MERGE_RESULT_RE = re.compile(
r"merge result\s*:\s*(?:merged|success|performed)",
re.IGNORECASE,
)
_MERGE_COMMIT_SHA_RE = re.compile(
r"(?:merge commit sha|merged commit sha|merge commit)\s*:\s*([0-9a-f]{7,40})",
re.IGNORECASE,
)
_PR_HEAD_BRANCH_RE = re.compile(
r"(?:merged pr head branch|pr head branch|deleted branch)\s*:\s*(\S+)",
re.IGNORECASE,
)
_BRANCH_NOT_PROTECTED_RE = re.compile(
r"branch (?:is )?not protected|branch protection\s*:\s*(?:none|false|no)",
re.IGNORECASE,
)
_OPEN_PR_INVENTORY_RE = re.compile(
r"(?:no other open pr(?:\s+references)?(?:\s+\S+)?|open pr inventory proof|"
r"open pr references).*(?:none|zero|0|clear|inventory complete)|"
r"no other open pr references branch",
re.IGNORECASE,
)
_ACTIVE_CLAIM_LEASE_RE = re.compile(
r"(?:no active (?:heartbeat|claim|lease)|"
r"(?:active )?(?:heartbeat|claim|lease)(?:/(?:claim|lease))*\s*:\s*none)",
re.IGNORECASE,
)
_SESSION_OWNED_WORKTREE_RE = re.compile(
r"(?:removed worktree path|cleanup worktree path|session-owned worktree)\s*:\s*(\S+)",
re.IGNORECASE,
)
_BRANCHES_PATH_RE = re.compile(r"\bbranches/", re.IGNORECASE)
_CLEAN_TRACKED_RE = re.compile(
r"(?:pre-removal tracked state|tracked state before removal)\s*:\s*clean",
re.IGNORECASE,
)
_CLEAN_UNTRACKED_RE = re.compile(
r"(?:pre-removal untracked state|untracked state before removal)\s*:\s*clean",
re.IGNORECASE,
)
_WORKTREE_LIST_AFTER_RE = re.compile(
r"(?:git worktree list after|post-removal worktree list|worktree list after)",
re.IGNORECASE,
)
_WRONG_BRANCH_RE = re.compile(
r"deleted branch (?:does not match|!=|differs from) (?:merged )?pr head",
re.IGNORECASE,
)
def _claims_remote_delete(text: str) -> bool:
return bool(_REMOTE_DELETE_CLAIM_RE.search(text))
def _claims_worktree_remove(text: str) -> bool:
return bool(_WORKTREE_REMOVE_CLAIM_RE.search(text))
def _branch_safety_fields_present(text: str) -> list[str]:
missing: list[str] = []
if not _DELETE_CAPABILITY_RE.search(text):
missing.append("delete-branch capability resolved (gitea.branch.delete)")
if not _DELETE_TASK_RE.search(text):
missing.append("delete-branch task named (delete_branch or cleanup)")
if not _MERGE_RESULT_RE.search(text):
missing.append("merge result: merged")
if not _MERGE_COMMIT_SHA_RE.search(text):
missing.append("merge commit SHA")
if not _PR_HEAD_BRANCH_RE.search(text):
missing.append("merged PR head branch / deleted branch name")
if not _BRANCH_NOT_PROTECTED_RE.search(text):
missing.append("branch not protected proof")
if not _OPEN_PR_INVENTORY_RE.search(text):
missing.append("open PR inventory proof (no other PR references branch)")
if not _ACTIVE_CLAIM_LEASE_RE.search(text):
missing.append("no active heartbeat/claim/lease proof")
return missing
def _worktree_cleanup_fields_present(text: str) -> list[str]:
missing: list[str] = []
match = _SESSION_OWNED_WORKTREE_RE.search(text)
path = match.group(1).strip() if match else ""
if not path:
missing.append("session-owned worktree path")
elif not _BRANCHES_PATH_RE.search(path.replace("\\", "/")):
missing.append("worktree path under branches/")
if not _CLEAN_TRACKED_RE.search(text):
missing.append("pre-removal tracked state: clean")
if not _CLEAN_UNTRACKED_RE.search(text):
missing.append("pre-removal untracked state: clean")
if not _WORKTREE_LIST_AFTER_RE.search(text):
missing.append("git worktree list after removal")
return missing
def assess_post_merge_cleanup_proof(
report_text: str,
*,
cleanup_session: dict | None = None,
) -> dict[str, Any]:
"""Validate post-merge cleanup claims carry safety-gate proof (#402)."""
text = report_text or ""
session = dict(cleanup_session or {})
reasons: list[str] = []
if _CLEANUP_SKIPPED_RE.search(text) or session.get("outcome") == CLEANUP_SKIPPED:
blocker = (session.get("blocker") or "").strip()
if not blocker:
match = _CLEANUP_BLOCKER_RE.search(text)
blocker = match.group(1).strip() if match else ""
if blocker.upper() == CLEANUP_SKIPPED:
blocker = ""
if not blocker:
reasons.append(
"CLEANUP_SKIPPED requires exact cleanup blocker reason (#402)"
)
return {
"block": bool(reasons),
"proven": not reasons,
"outcome": CLEANUP_SKIPPED,
"remote_delete_claimed": False,
"worktree_remove_claimed": False,
"reasons": reasons,
"safe_next_action": (
"report CLEANUP_SKIPPED with exact blocker; do not claim performed cleanup"
if reasons
else "proceed"
),
}
if not _CLEANUP_SECTION_HINT.search(text) and not session.get("cleanup_claimed"):
return {
"block": False,
"proven": True,
"outcome": None,
"remote_delete_claimed": False,
"worktree_remove_claimed": False,
"reasons": [],
"safe_next_action": "proceed",
}
remote_delete = bool(
session.get("remote_delete_claimed") or _claims_remote_delete(text)
)
worktree_remove = bool(
session.get("worktree_remove_claimed") or _claims_worktree_remove(text)
)
if _WRONG_BRANCH_RE.search(text):
reasons.append(
"cleanup report claims deleted branch that is not the merged PR head branch"
)
if remote_delete:
reasons.extend(
f"remote branch deletion missing {field}"
for field in _branch_safety_fields_present(text)
)
if worktree_remove:
reasons.extend(
f"worktree removal missing {field}"
for field in _worktree_cleanup_fields_present(text)
)
if (remote_delete or worktree_remove) and not (remote_delete or worktree_remove):
pass
if not remote_delete and not worktree_remove:
cleanup_mutations = re.search(
r"cleanup mutations\s*:\s*(?!none\b)\S",
text,
re.IGNORECASE,
)
if cleanup_mutations:
reasons.append(
"cleanup mutations reported without post-merge cleanup proof checklist"
)
outcome = CLEANUP_PERFORMED if (remote_delete or worktree_remove) and not reasons else None
if remote_delete or worktree_remove:
outcome = CLEANUP_PERFORMED if not reasons else "CLEANUP_CLAIMED_UNPROVEN"
block = bool(reasons)
return {
"block": block,
"proven": not block,
"outcome": outcome,
"remote_delete_claimed": remote_delete,
"worktree_remove_claimed": worktree_remove,
"reasons": reasons,
"safe_next_action": (
"report CLEANUP_SKIPPED with exact blocker or include the full cleanup "
"checklist before claiming remote delete or worktree removal"
if reasons
else "proceed"
),
}
+225
View File
@@ -0,0 +1,225 @@
"""PR-only queue cleanup mode gates and report verifier (#390).
Cleanup mode dispatches exactly one canonical review run per PR. It forbids
author-side mutations (issue claiming, branch creation, implementation edits,
issue filing) and enforces the terminal-mutation chain: stop after
``REQUEST_CHANGES``; after ``APPROVED`` continue only to same-PR merge when
merge is explicitly authorized for that PR and merge gates pass; stop after
merge or a merge blocker. The next PR always requires a fresh run.
"""
from __future__ import annotations
import re
CLEANUP_WORKFLOW_PATH = "workflows/pr-queue-cleanup.md"
# Author-side resolver tasks that must never run inside cleanup mode.
CLEANUP_FORBIDDEN_TASKS = frozenset({
"claim_issue",
"mark_issue",
"lock_issue",
"create_issue",
"create_branch",
"push_branch",
"create_pr",
"commit_files",
"gitea_commit_files",
"address_pr_change_requests",
})
# Run-state outcomes for one cleanup dispatch.
STOP_AFTER_REQUEST_CHANGES = "STOP_AFTER_REQUEST_CHANGES"
STOP_AFTER_DECISION = "STOP_AFTER_DECISION"
CONTINUE_TO_SAME_PR_MERGE = "CONTINUE_TO_SAME_PR_MERGE"
STOP_APPROVED_NO_MERGE_AUTH = "STOP_APPROVED_NO_MERGE_AUTH"
STOP_APPROVED_MERGE_GATES_FAILED = "STOP_APPROVED_MERGE_GATES_FAILED"
STOP_AFTER_MERGE = "STOP_AFTER_MERGE"
STOP_AFTER_MERGE_BLOCKER = "STOP_AFTER_MERGE_BLOCKER"
STOP_GATE_NOT_PROVEN = "STOP_GATE_NOT_PROVEN"
_TERMINAL_DECISIONS = frozenset({"approved", "request_changes", "comment", "skip"})
def resolve_cleanup_run_state(
decision: str | None,
*,
merge_authorized_for_pr: bool = False,
merge_gates_passed: bool | None = None,
merge_completed: bool = False,
merge_blocker: bool = False,
) -> dict:
"""Resolve what one cleanup run may do after its terminal review decision.
Fail closed: unknown decisions stop the run with no further mutation.
"""
normalized = (decision or "").strip().lower()
if normalized not in _TERMINAL_DECISIONS:
return {
"outcome": STOP_GATE_NOT_PROVEN,
"further_mutation_allowed": False,
"reasons": [
f"unknown terminal decision {decision!r}; cleanup run stops "
"(fail closed)"
],
}
if normalized == "request_changes":
return {
"outcome": STOP_AFTER_REQUEST_CHANGES,
"further_mutation_allowed": False,
"reasons": ["REQUEST_CHANGES is terminal in cleanup mode"],
}
if normalized in {"comment", "skip"}:
return {
"outcome": STOP_AFTER_DECISION,
"further_mutation_allowed": False,
"reasons": [f"{normalized} decision ends the cleanup run"],
}
# normalized == "approved"
if merge_completed:
return {
"outcome": STOP_AFTER_MERGE,
"further_mutation_allowed": False,
"reasons": ["merge completed; cleanup run stops"],
}
if merge_blocker:
return {
"outcome": STOP_AFTER_MERGE_BLOCKER,
"further_mutation_allowed": False,
"reasons": ["merge blocker recorded; cleanup run stops"],
}
if not merge_authorized_for_pr:
return {
"outcome": STOP_APPROVED_NO_MERGE_AUTH,
"further_mutation_allowed": False,
"reasons": [
"APPROVED without explicit per-PR merge authorization; "
"cleanup run stops"
],
}
if merge_gates_passed is not True:
return {
"outcome": STOP_APPROVED_MERGE_GATES_FAILED,
"further_mutation_allowed": False,
"reasons": [
"merge gates not proven passed; cleanup run stops before merge"
],
}
return {
"outcome": CONTINUE_TO_SAME_PR_MERGE,
"further_mutation_allowed": True,
"reasons": [],
"allowed_mutation": "merge same PR only",
}
def check_cleanup_task_allowed(task: str) -> tuple[bool, list[str]]:
"""Fail closed on any author-side mutation task inside cleanup mode."""
normalized = (task or "").strip().lower()
if normalized in CLEANUP_FORBIDDEN_TASKS:
return False, [
f"task '{normalized}' is forbidden in PR-only cleanup mode: "
"no issue claiming, branch creation, implementation edits, or "
"issue filing"
]
return True, []
_SELECTED_PR_RE = re.compile(
r"^\s*[-*]?\s*selected pr\s*:\s*#?(\d+)", re.IGNORECASE | re.MULTILINE
)
_NEXT_SUGGESTED_RE = re.compile(
r"^\s*[-*]?\s*next suggested pr\s*:", re.IGNORECASE | re.MULTILINE
)
_TERMINAL_MUTATION_RE = re.compile(
r"^\s*[-*]?\s*(?:review (?:decision|verdict)|terminal (?:review )?"
r"(?:decision|mutation))\s*:\s*"
r"(approved|request_changes|request changes|merged|comment)",
re.IGNORECASE | re.MULTILINE,
)
_PAGINATION_HINT_RE = re.compile(
r"inventory_complete|final page|has_more\s*[=:]\s*false|total[_ ]count",
re.IGNORECASE,
)
_MERGE_RESULT_RE = re.compile(
r"^\s*[-*]?\s*merge result\s*:\s*(?!none\b|not attempted\b)\S",
re.IGNORECASE | re.MULTILINE,
)
_MERGE_AUTHORIZED_RE = re.compile(
r"^\s*[-*]?\s*merge authorized(?: for pr)?\s*:\s*true",
re.IGNORECASE | re.MULTILINE,
)
_ISSUE_MUTATION_CLAIM_RE = re.compile(
r"^\s*[-*]?\s*issue mutations\s*:\s*(?!none\b)\S",
re.IGNORECASE | re.MULTILINE,
)
_BRANCH_MUTATION_CLAIM_RE = re.compile(
r"^\s*[-*]?\s*branch mutations\s*:\s*(?!none\b)\S",
re.IGNORECASE | re.MULTILINE,
)
def assess_pr_queue_cleanup_report(report_text: str) -> dict:
"""Validate a PR-only cleanup run report (single dispatch, fail closed)."""
reasons: list[str] = []
text = report_text or ""
if CLEANUP_WORKFLOW_PATH not in text:
reasons.append(
f"report must cite the canonical cleanup workflow "
f"({CLEANUP_WORKFLOW_PATH})"
)
selected = _SELECTED_PR_RE.findall(text)
if len(selected) == 0:
reasons.append("report must name exactly one Selected PR")
elif len(set(selected)) > 1:
reasons.append(
"cleanup run selected multiple PRs "
f"({', '.join(sorted(set(selected)))}); one canonical review per "
"PR per run"
)
terminal = _TERMINAL_MUTATION_RE.findall(text)
if len(terminal) > 1:
reasons.append(
"multiple terminal review mutations reported in one cleanup run"
)
if not _PAGINATION_HINT_RE.search(text):
reasons.append(
"report missing PR inventory pagination proof "
"(inventory_complete / final page / total_count)"
)
if not _NEXT_SUGGESTED_RE.search(text):
reasons.append(
"report must include 'Next suggested PR' (without continuing to it)"
)
if _MERGE_RESULT_RE.search(text) and not _MERGE_AUTHORIZED_RE.search(text):
reasons.append(
"merge reported without explicit per-PR merge authorization "
"('Merge authorized: true')"
)
if _ISSUE_MUTATION_CLAIM_RE.search(text):
reasons.append("issue mutations are forbidden in PR-only cleanup mode")
if _BRANCH_MUTATION_CLAIM_RE.search(text):
reasons.append("branch mutations are forbidden in PR-only cleanup mode")
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"safe_next_action": (
"proceed" if proven else
"fix the cleanup report: one PR, one terminal mutation, pagination "
"proof, next-suggested-PR field, and no issue/branch mutations"
),
}
+482
View File
@@ -0,0 +1,482 @@
"""Conflict-fix and reviewer PR work leases (#399, #407 reader).
Structured PR/issue comments prove exclusive phases so author conflict-fix
pushes cannot race reviewer validation/approval/merge on the same head.
"""
from __future__ import annotations
import re
from datetime import datetime, timedelta, timezone
from typing import Any
REVIEWER_LEASE_MARKER = "<!-- mcp-review-lease:v1 -->"
CONFLICT_FIX_LEASE_MARKER = "<!-- mcp-conflict-fix-lease:v1 -->"
_FULL_SHA = re.compile(r"^[0-9a-f]{40}$", re.IGNORECASE)
_FIELD_RE = re.compile(
r"^\s*([a-z_]+)\s*:\s*(.+?)\s*$",
re.IGNORECASE | re.MULTILINE,
)
_TERMINAL_REVIEWER_PHASES = frozenset({"done", "released", "blocked"})
_ACTIVE_REVIEWER_PHASES = frozenset({
"claimed",
"validating",
"approved",
"request-changes",
"merging",
})
_TERMINAL_CONFLICT_FIX_PHASES = frozenset({"released", "blocked", "done"})
_ACTIVE_CONFLICT_FIX_PHASES = frozenset({"claimed", "pushing", "pushed"})
DEFAULT_CONFLICT_FIX_TTL_MINUTES = 120
DEFAULT_REVIEWER_LEASE_TTL_MINUTES = 120
def _parse_timestamp(value: str | None) -> datetime | None:
if not value:
return None
text = value.strip()
if text.endswith("Z"):
text = text[:-1] + "+00:00"
try:
parsed = datetime.fromisoformat(text)
except ValueError:
return None
if parsed.tzinfo is None:
return parsed.replace(tzinfo=timezone.utc)
return parsed.astimezone(timezone.utc)
def _normalize_sha(value: str | None) -> str | None:
text = (value or "").strip().lower()
if not text:
return None
return text if _FULL_SHA.match(text) else None
def _parse_pr_ref(value: str | None) -> int | None:
digits = re.sub(r"[^\d]", "", value or "")
return int(digits) if digits.isdigit() else None
def _parse_marker_comment(body: str, marker: str) -> dict[str, str] | None:
text = body or ""
if marker not in text:
return None
fields: dict[str, str] = {}
for match in _FIELD_RE.finditer(text):
fields[match.group(1).strip().lower()] = match.group(2).strip()
return fields or None
def parse_reviewer_lease_comment(body: str) -> dict[str, Any] | None:
fields = _parse_marker_comment(body, REVIEWER_LEASE_MARKER)
if not fields:
return None
return {
"lease_kind": "reviewer",
"pr_number": _parse_pr_ref(fields.get("pr")),
"issue_number": _parse_pr_ref(fields.get("issue")),
"reviewer_identity": fields.get("reviewer_identity"),
"profile": fields.get("profile"),
"session_id": fields.get("session_id"),
"worktree": fields.get("worktree"),
"phase": (fields.get("phase") or "").strip().lower() or None,
"candidate_head": _normalize_sha(fields.get("candidate_head")),
"target_branch": fields.get("target_branch"),
"target_branch_sha": _normalize_sha(fields.get("target_branch_sha")),
"last_activity": fields.get("last_activity"),
"expires_at": fields.get("expires_at"),
"blocker": fields.get("blocker"),
"raw_fields": fields,
}
def parse_conflict_fix_lease_comment(body: str) -> dict[str, Any] | None:
fields = _parse_marker_comment(body, CONFLICT_FIX_LEASE_MARKER)
if not fields:
return None
ff = (fields.get("fast_forward") or "").strip().lower()
reviewer_active = (fields.get("reviewer_active") or "").strip().lower()
return {
"lease_kind": "conflict_fix",
"pr_number": _parse_pr_ref(fields.get("pr")),
"branch": fields.get("branch"),
"worktree": fields.get("worktree"),
"profile": fields.get("profile"),
"session_id": fields.get("session_id"),
"phase": (fields.get("phase") or "").strip().lower() or None,
"head_before": _normalize_sha(fields.get("head_before")),
"head_after": _normalize_sha(fields.get("head_after")),
"expires_at": fields.get("expires_at"),
"reviewer_active": reviewer_active in {"yes", "true", "1"},
"fast_forward": ff in {"yes", "true", "1"},
"raw_fields": fields,
}
def _comment_entries(comments: list[dict], *, pr_number: int | None) -> list[dict]:
entries: list[dict] = []
for comment in comments or []:
body = comment.get("body") or ""
for parser in (parse_reviewer_lease_comment, parse_conflict_fix_lease_comment):
parsed = parser(body)
if not parsed:
continue
if pr_number is not None and parsed.get("pr_number") not in (None, pr_number):
continue
entries.append({
**parsed,
"comment_id": comment.get("id"),
"author": (comment.get("user") or {}).get("login") or comment.get("author"),
"created_at": comment.get("created_at"),
"updated_at": comment.get("updated_at"),
})
break
return entries
def _lease_expired(lease: dict, *, now: datetime) -> bool:
expires_at = _parse_timestamp(lease.get("expires_at"))
return bool(expires_at and expires_at <= now)
def _lease_phase_active(lease: dict, *, active_phases: frozenset[str]) -> bool:
phase = (lease.get("phase") or "").strip().lower()
if phase in _TERMINAL_REVIEWER_PHASES or phase in _TERMINAL_CONFLICT_FIX_PHASES:
return False
return phase in active_phases or bool(phase and phase not in (
_TERMINAL_REVIEWER_PHASES | _TERMINAL_CONFLICT_FIX_PHASES
))
def find_active_reviewer_lease(
comments: list[dict],
*,
pr_number: int,
now: datetime | None = None,
) -> dict[str, Any] | None:
"""Return the newest unexpired reviewer lease for *pr_number*, if any."""
now = now or datetime.now(timezone.utc)
candidates = [
entry for entry in _comment_entries(comments, pr_number=pr_number)
if entry.get("lease_kind") == "reviewer"
]
for lease in reversed(candidates):
if _lease_expired(lease, now=now):
continue
phase = (lease.get("phase") or "").strip().lower()
if phase in _TERMINAL_REVIEWER_PHASES:
continue
if phase in _ACTIVE_REVIEWER_PHASES or phase:
return lease
return None
def find_active_conflict_fix_lease(
comments: list[dict],
*,
pr_number: int,
now: datetime | None = None,
) -> dict[str, Any] | None:
"""Return the newest unexpired conflict-fix lease for *pr_number*, if any."""
now = now or datetime.now(timezone.utc)
candidates = [
entry for entry in _comment_entries(comments, pr_number=pr_number)
if entry.get("lease_kind") == "conflict_fix"
]
for lease in reversed(candidates):
if _lease_expired(lease, now=now):
continue
phase = (lease.get("phase") or "").strip().lower()
if phase in _TERMINAL_CONFLICT_FIX_PHASES:
continue
if phase in _ACTIVE_CONFLICT_FIX_PHASES or phase:
return lease
return None
def format_conflict_fix_lease_body(
*,
pr_number: int,
branch: str,
worktree: str,
profile: str,
head_before: str,
phase: str = "claimed",
session_id: str = "unknown",
expires_at: datetime | None = None,
reviewer_active: bool = False,
) -> str:
expires = expires_at or (
datetime.now(timezone.utc) + timedelta(minutes=DEFAULT_CONFLICT_FIX_TTL_MINUTES)
)
expires_text = expires.astimezone(timezone.utc).replace(microsecond=0).isoformat().replace(
"+00:00", "Z"
)
lines = [
CONFLICT_FIX_LEASE_MARKER,
f"pr: #{pr_number}",
f"branch: {branch}",
f"worktree: {worktree}",
f"profile: {profile}",
f"session_id: {session_id}",
f"phase: {phase}",
f"head_before: {head_before}",
f"expires_at: {expires_text}",
f"reviewer_active: {'yes' if reviewer_active else 'no'}",
]
return "\n".join(lines)
def assess_head_sha_equality(
reviewed_head_sha: str | None,
live_head_sha: str | None,
) -> dict[str, Any]:
"""Fail closed when reviewed and live PR heads differ."""
reviewed = _normalize_sha(reviewed_head_sha)
live = _normalize_sha(live_head_sha)
reasons: list[str] = []
if not reviewed or not live:
reasons.append(
"reviewed/live head SHA missing or not full 40-hex; fail closed"
)
elif reviewed != live:
reasons.append(
"PR head changed after validation; re-pin and re-validate before "
"approval or merge"
)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"reviewed_head_sha": reviewed,
"live_head_sha": live,
"head_changed": bool(reviewed and live and reviewed != live),
}
def assess_conflict_fix_push(
*,
pr_number: int,
comments: list[dict],
branch_head_before: str | None,
branch_head_after: str | None,
worktree_path: str | None,
push_cwd: str | None,
is_fast_forward: bool | None,
now: datetime | None = None,
) -> dict[str, Any]:
"""Author pre-push gate: block when a reviewer holds an active lease."""
now = now or datetime.now(timezone.utc)
reasons: list[str] = []
reviewer_lease = find_active_reviewer_lease(comments, pr_number=pr_number, now=now)
conflict_lease = find_active_conflict_fix_lease(comments, pr_number=pr_number, now=now)
if reviewer_lease:
reasons.append(
f"active reviewer lease on PR #{pr_number} "
f"(phase={reviewer_lease.get('phase')}); author push blocked"
)
head_before = _normalize_sha(branch_head_before)
head_after = _normalize_sha(branch_head_after)
if not head_before:
reasons.append("branch head before push missing or invalid SHA")
if head_after and head_before and head_before == head_after:
reasons.append("branch head unchanged; no push to perform")
worktree = (worktree_path or "").strip()
cwd = (push_cwd or "").strip()
if not worktree:
reasons.append("worktree path required for conflict-fix push proof")
elif cwd and worktree and not cwd.rstrip("/").endswith(worktree.rstrip("/").split("/")[-1]):
if worktree not in cwd:
reasons.append(
f"push cwd '{cwd}' does not match session worktree '{worktree}'"
)
if is_fast_forward is False:
reasons.append("non-fast-forward push rejected for conflict-fix (fail closed)")
if conflict_lease and conflict_lease.get("phase") == "pushing":
owner = conflict_lease.get("worktree")
if owner and worktree and owner != worktree:
reasons.append(
f"sibling conflict-fix lease active from worktree '{owner}'"
)
push_allowed = not reasons
return {
"push_allowed": push_allowed,
"block": not push_allowed,
"reasons": reasons,
"active_reviewer_lease": reviewer_lease,
"active_conflict_fix_lease": conflict_lease,
"branch_head_before": head_before,
"branch_head_after": head_after,
"reviewer_was_active": bool(reviewer_lease),
"fast_forward": is_fast_forward,
}
def assess_reviewer_mutation_blocked(
*,
pr_number: int,
comments: list[dict],
reviewed_head_sha: str | None,
live_head_sha: str | None,
mutation: str,
now: datetime | None = None,
) -> dict[str, Any]:
"""Reviewer gate: block when conflict-fix lease active or head moved."""
now = now or datetime.now(timezone.utc)
reasons: list[str] = []
conflict_lease = find_active_conflict_fix_lease(comments, pr_number=pr_number, now=now)
if conflict_lease and (conflict_lease.get("phase") or "") in _ACTIVE_CONFLICT_FIX_PHASES:
reasons.append(
f"active conflict-fix lease on PR #{pr_number} "
f"(phase={conflict_lease.get('phase')}); reviewer {mutation} blocked"
)
head_check = assess_head_sha_equality(reviewed_head_sha, live_head_sha)
if head_check["block"]:
reasons.extend(head_check["reasons"])
if not _normalize_sha(reviewed_head_sha):
reasons.append(
f"reviewed head SHA required before reviewer {mutation} (fail closed)"
)
allowed = not reasons
return {
"mutation_allowed": allowed,
"block": not allowed,
"reasons": reasons,
"active_conflict_fix_lease": conflict_lease,
"head_check": head_check,
"reviewed_head_sha": head_check.get("reviewed_head_sha"),
"live_head_sha": head_check.get("live_head_sha"),
"push_during_validation": bool(
conflict_lease and conflict_lease.get("phase") in {"pushing", "pushed"}
),
}
_REVIEWED_HEAD_RE = re.compile(
r"reviewed head sha\s*:\s*([0-9a-f]{40})",
re.IGNORECASE,
)
_LIVE_HEAD_BEFORE_APPROVAL_RE = re.compile(
r"(?:live head sha before approval|final live head sha before approval)\s*:\s*([0-9a-f]{40})",
re.IGNORECASE,
)
_LIVE_HEAD_BEFORE_MERGE_RE = re.compile(
r"(?:live head sha before merge|final live head sha before merge)\s*:\s*([0-9a-f]{40})",
re.IGNORECASE,
)
_PUSH_DURING_VALIDATION_RE = re.compile(
r"push(?:es)? occurred during validation\s*:\s*(yes|no|true|false)",
re.IGNORECASE,
)
_CONFLICT_HEAD_BEFORE_RE = re.compile(
r"branch head before push\s*:\s*([0-9a-f]{40})",
re.IGNORECASE,
)
_CONFLICT_HEAD_AFTER_RE = re.compile(
r"branch head after push\s*:\s*([0-9a-f]{40})",
re.IGNORECASE,
)
_REVIEWER_LEASE_STATUS_RE = re.compile(
r"active reviewer lease status\s*:\s*(.+)$",
re.IGNORECASE | re.MULTILINE,
)
_FAST_FORWARD_RE = re.compile(
r"whether push was fast-forward\s*:\s*(yes|no|true|false)",
re.IGNORECASE,
)
_REVIEWER_ACTIVE_RE = re.compile(
r"whether any reviewer was active\s*:\s*(yes|no|true|false)",
re.IGNORECASE,
)
def assess_reviewer_stale_head_final_report(report_text: str) -> dict[str, Any]:
"""Final-report proof for reviewed vs live head SHAs (#399 AC 6)."""
text = report_text or ""
reasons: list[str] = []
reviewed = _normalize_sha(_REVIEWED_HEAD_RE.search(text).group(1) if _REVIEWED_HEAD_RE.search(text) else None)
live_approval = _normalize_sha(
_LIVE_HEAD_BEFORE_APPROVAL_RE.search(text).group(1)
if _LIVE_HEAD_BEFORE_APPROVAL_RE.search(text)
else None
)
live_merge = _normalize_sha(
_LIVE_HEAD_BEFORE_MERGE_RE.search(text).group(1)
if _LIVE_HEAD_BEFORE_MERGE_RE.search(text)
else None
)
push_during = _PUSH_DURING_VALIDATION_RE.search(text)
if not reviewed:
reasons.append("reviewed head SHA not stated in final report")
if not live_approval:
reasons.append("final live head SHA before approval not stated")
if not live_merge:
reasons.append("final live head SHA before merge not stated")
if not push_during:
reasons.append("whether push occurred during validation not stated")
elif reviewed and live_approval and reviewed != live_approval:
reasons.append("live head before approval differs from reviewed head SHA")
elif reviewed and live_merge and reviewed != live_merge:
reasons.append("live head before merge differs from reviewed head SHA")
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"reviewed_head_sha": reviewed,
"live_head_sha_before_approval": live_approval,
"live_head_sha_before_merge": live_merge,
"push_during_validation": (push_during.group(1).lower() if push_during else None),
}
def assess_conflict_fix_final_report(report_text: str) -> dict[str, Any]:
"""Final-report proof for conflict-fix push sessions (#399 AC 7)."""
text = report_text or ""
reasons: list[str] = []
head_before = _normalize_sha(
_CONFLICT_HEAD_BEFORE_RE.search(text).group(1)
if _CONFLICT_HEAD_BEFORE_RE.search(text)
else None
)
head_after = _normalize_sha(
_CONFLICT_HEAD_AFTER_RE.search(text).group(1)
if _CONFLICT_HEAD_AFTER_RE.search(text)
else None
)
if not head_before:
reasons.append("branch head before push not stated")
if not head_after:
reasons.append("branch head after push not stated")
if not _REVIEWER_LEASE_STATUS_RE.search(text):
reasons.append("active reviewer lease status not stated")
if not _FAST_FORWARD_RE.search(text):
reasons.append("whether push was fast-forward not stated")
if not _REVIEWER_ACTIVE_RE.search(text):
reasons.append("whether any reviewer was active not stated")
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"branch_head_before": head_before,
"branch_head_after": head_after,
}
+94
View File
@@ -0,0 +1,94 @@
"""Reconciler profile model for already-landed PR closure (#304).
Defines the narrowly scoped operation set for a dedicated reconciler profile
such as ``prgs-reconciler``. Close MCP tooling and ancestry gates ship in the
#310 stack; this module validates profile shape only.
"""
from __future__ import annotations
import gitea_config
RECONCILER_REQUIRED_OPERATIONS = (
"gitea.read",
"gitea.pr.close",
)
RECONCILER_RECOMMENDED_OPERATIONS = (
"gitea.pr.comment",
"gitea.issue.comment",
"gitea.issue.close",
)
RECONCILER_FORBIDDEN_OPERATIONS = (
"gitea.pr.approve",
"gitea.pr.merge",
"gitea.pr.review",
"gitea.pr.create",
"gitea.branch.push",
"gitea.repo.commit",
)
def _normalized_allowed(allowed: list[str]) -> set[str]:
normalized: set[str] = set()
for entry in allowed or []:
try:
normalized.add(gitea_config.normalize_operation(entry))
except gitea_config.ConfigError:
continue
return normalized
def _forbidden_in_allowed(allowed: list[str], ops: tuple[str, ...]) -> list[str]:
"""Return forbidden ops that are explicitly listed in *allowed*."""
allowed_n = _normalized_allowed(allowed)
present: list[str] = []
for op in ops:
try:
if gitea_config.normalize_operation(op) in allowed_n:
present.append(op)
except gitea_config.ConfigError:
continue
return present
def is_reconciler_profile(allowed: list[str], forbidden: list[str]) -> bool:
"""Return True when *allowed*/*forbidden* describe a reconciler profile."""
def can(op: str) -> bool:
return gitea_config.check_operation(op, allowed, forbidden)[0]
if not can("gitea.pr.close"):
return False
if _forbidden_in_allowed(allowed, RECONCILER_FORBIDDEN_OPERATIONS):
return False
return True
def assess_reconciler_profile(allowed: list[str], forbidden: list[str]) -> dict:
"""Validate reconciler profile operations (read-only, fail closed)."""
allowed = list(allowed or [])
forbidden = list(forbidden or [])
reasons: list[str] = []
for op in RECONCILER_REQUIRED_OPERATIONS:
ok, _ = gitea_config.check_operation(op, allowed, forbidden)
if not ok:
reasons.append(f"missing required operation {op}")
for op in _forbidden_in_allowed(allowed, RECONCILER_FORBIDDEN_OPERATIONS):
reasons.append(f"forbidden operation must not be allowed: {op}")
missing_recommended = [
op for op in RECONCILER_RECOMMENDED_OPERATIONS
if not gitea_config.check_operation(op, allowed, forbidden)[0]
]
return {
"is_reconciler_profile": is_reconciler_profile(allowed, forbidden),
"valid": not reasons and is_reconciler_profile(allowed, forbidden),
"reasons": reasons,
"missing_recommended_operations": missing_recommended,
"required_operations": list(RECONCILER_REQUIRED_OPERATIONS),
"forbidden_operations": list(RECONCILER_FORBIDDEN_OPERATIONS),
}
+292
View File
@@ -0,0 +1,292 @@
"""Already-landed PR reconciliation workflow helpers (#301).
Read-only assessment and capability planning for reconciling open PRs whose
heads are already ancestors of the target branch. Does not invoke review or
merge paths.
"""
from __future__ import annotations
import subprocess
from typing import Any
import gitea_config
from merged_cleanup_reconcile import extract_linked_issue, is_head_ancestor_of_ref
ELIGIBILITY_ALREADY_LANDED = "ALREADY_LANDED_RECONCILE_REQUIRED"
ELIGIBILITY_NOT_LANDED = "NOT_ALREADY_LANDED"
ELIGIBILITY_STALE_TARGET = "TARGET_BRANCH_UNVERIFIED"
ELIGIBILITY_PR_NOT_OPEN = "PR_NOT_OPEN"
OUTCOME_FULL_RECONCILE = "FULL_RECONCILE_CLOSE_ALLOWED"
OUTCOME_PARTIAL_COMMENT = "PARTIAL_RECONCILE_COMMENT_THEN_STOP"
OUTCOME_RECOVERY_HANDOFF = "RECOVERY_HANDOFF_ONLY"
OUTCOME_NOT_LANDED = "NOT_LANDED_NO_ACTION"
OUTCOME_GATE_NOT_PROVEN = "GATE_NOT_PROVEN"
RECONCILE_WORKFLOW_MARKERS = (
"workflows/reconcile-landed-pr.md",
"reconcile-landed-pr.md",
)
RECONCILE_TASK_MARKERS = (
"reconcile-landed-pr",
"reconcile already-landed",
"reconcile_already_landed",
)
def fetch_target_branch(project_root: str, remote: str, branch: str) -> dict[str, Any]:
"""Fetch *branch* from *remote* and return the resolved SHA."""
ref = f"{remote}/{branch}"
fetch = subprocess.run(
["git", "-C", project_root, "fetch", remote, branch],
capture_output=True,
text=True,
check=False,
)
if fetch.returncode != 0:
return {
"success": False,
"target_branch": branch,
"target_ref": ref,
"target_branch_sha": None,
"reasons": [
f"git fetch {remote} {branch} failed: "
f"{(fetch.stderr or fetch.stdout or '').strip()}"
],
}
rev = subprocess.run(
["git", "-C", project_root, "rev-parse", ref],
capture_output=True,
text=True,
check=False,
)
if rev.returncode != 0:
return {
"success": False,
"target_branch": branch,
"target_ref": ref,
"target_branch_sha": None,
"reasons": [
f"git rev-parse {ref} failed: "
f"{(rev.stderr or rev.stdout or '').strip()}"
],
}
return {
"success": True,
"target_branch": branch,
"target_ref": ref,
"target_branch_sha": (rev.stdout or "").strip(),
"reasons": [],
"git_fetch_command": f"git fetch {remote} {branch}",
}
def profile_reconciliation_capabilities(
allowed: list[str] | None,
forbidden: list[str] | None,
) -> dict[str, bool]:
"""Map active profile operations to reconciliation capabilities."""
allowed = allowed or []
forbidden = forbidden or []
def can(op: str) -> bool:
return gitea_config.check_operation(op, allowed, forbidden)[0]
return {
"read": can("gitea.read"),
"comment_pr": can("gitea.pr.comment"),
"comment_issue": can("gitea.issue.comment"),
"close_pr": can("gitea.pr.close"),
"close_issue": can("gitea.issue.close"),
"review_pr": can("gitea.pr.review") or can("gitea.pr.approve"),
"merge_pr": can("gitea.pr.merge"),
}
def assess_open_pr_reconciliation(
*,
pr: dict[str, Any],
project_root: str,
remote: str,
target_branch: str,
target_fetch: dict[str, Any] | None = None,
) -> dict[str, Any]:
"""Assess whether an open PR is eligible for already-landed reconciliation."""
pr_number = int(pr.get("number") or 0)
pr_state = (pr.get("state") or "").strip().lower()
head = pr.get("head") or {}
head_sha = head.get("sha") if isinstance(head, dict) else None
head_ref = head.get("ref") if isinstance(head, dict) else None
base = pr.get("base") or {}
base_ref = base.get("ref") if isinstance(base, dict) else None
title = pr.get("title") or ""
body = pr.get("body") or ""
fetch_result = target_fetch or fetch_target_branch(
project_root, remote, target_branch
)
linked_issue = extract_linked_issue(title, body)
result: dict[str, Any] = {
"pr_number": pr_number,
"pr_state": pr_state,
"candidate_head_sha": head_sha,
"head_ref": head_ref,
"base_ref": base_ref or target_branch,
"target_branch": target_branch,
"target_branch_sha": fetch_result.get("target_branch_sha"),
"linked_issue": linked_issue,
"git_ref_mutations": [],
"reasons": [],
"review_merge_allowed": False,
}
if fetch_result.get("git_fetch_command"):
result["git_ref_mutations"].append(fetch_result["git_fetch_command"])
if pr_state != "open":
result["eligibility_class"] = ELIGIBILITY_PR_NOT_OPEN
result["ancestor_proof"] = None
result["reconciliation_allowed"] = False
result["reasons"].append(f"PR #{pr_number} state is {pr_state!r}, not open")
return result
if not fetch_result.get("success"):
result["eligibility_class"] = ELIGIBILITY_STALE_TARGET
result["ancestor_proof"] = None
result["reconciliation_allowed"] = False
result["reasons"].extend(fetch_result.get("reasons") or [])
return result
target_ref = fetch_result.get("target_ref") or f"{remote}/{target_branch}"
ancestor = is_head_ancestor_of_ref(project_root, head_sha, target_ref)
result["ancestor_proof"] = ancestor
if ancestor is None:
result["eligibility_class"] = ELIGIBILITY_STALE_TARGET
result["reconciliation_allowed"] = False
result["reasons"].append(
f"ancestor check failed for head {head_sha!r} against {target_ref}"
)
return result
if ancestor:
result["eligibility_class"] = ELIGIBILITY_ALREADY_LANDED
result["reconciliation_allowed"] = True
return result
result["eligibility_class"] = ELIGIBILITY_NOT_LANDED
result["reconciliation_allowed"] = False
result["reasons"].append(
f"PR head {head_sha} is not an ancestor of {target_ref}"
)
return result
def resolve_reconciliation_plan(
*,
assessment: dict[str, Any],
capabilities: dict[str, bool] | None,
) -> dict[str, Any]:
"""Map eligibility + profile capabilities to a reconciliation outcome."""
caps = capabilities or {}
reasons: list[str] = []
if not assessment.get("reconciliation_allowed"):
outcome = OUTCOME_NOT_LANDED
if assessment.get("eligibility_class") in {
ELIGIBILITY_STALE_TARGET,
ELIGIBILITY_PR_NOT_OPEN,
}:
outcome = OUTCOME_GATE_NOT_PROVEN
reasons.extend(assessment.get("reasons") or [])
return {
"outcome": outcome,
"close_pr_allowed": False,
"comment_pr_allowed": False,
"close_issue_allowed": False,
"comment_issue_allowed": False,
"review_merge_allowed": False,
"missing_capabilities": _missing_reconciliation_capabilities(caps),
"safe_next_action": (
"Do not reconcile via review/merge; repair missing proof first."
),
"reasons": reasons,
}
close_pr = bool(caps.get("close_pr"))
comment_pr = bool(caps.get("comment_pr"))
close_issue = bool(caps.get("close_issue"))
comment_issue = bool(caps.get("comment_issue"))
if caps.get("review_pr") or caps.get("merge_pr"):
reasons.append(
"reconciliation path must not use review/merge capabilities"
)
if close_pr:
outcome = OUTCOME_FULL_RECONCILE
safe_next_action = (
"Run reconciliation close via exact gitea.pr.close after proof; "
"do not approve or merge."
)
elif comment_pr:
outcome = OUTCOME_PARTIAL_COMMENT
safe_next_action = (
"Post one reconciliation comment with ancestor proof, then stop "
"for an authorized close profile."
)
else:
outcome = OUTCOME_RECOVERY_HANDOFF
safe_next_action = (
"Produce a recovery handoff naming missing gitea.pr.close and/or "
"gitea.pr.comment; do not loop through review/merge."
)
reasons.append("gitea.pr.close is not available in the active profile")
return {
"outcome": outcome,
"close_pr_allowed": close_pr,
"comment_pr_allowed": comment_pr,
"close_issue_allowed": close_issue,
"comment_issue_allowed": comment_issue,
"review_merge_allowed": False,
"missing_capabilities": _missing_reconciliation_capabilities(caps),
"safe_next_action": safe_next_action,
"reasons": reasons,
}
def _missing_reconciliation_capabilities(caps: dict[str, bool]) -> list[str]:
missing = []
if not caps.get("read"):
missing.append("gitea.read")
if not caps.get("close_pr"):
missing.append("gitea.pr.close")
if not caps.get("comment_pr"):
missing.append("gitea.pr.comment")
return missing
def assess_reconcile_workflow_source(report_text: str) -> dict[str, Any]:
"""Reconciliation reports must cite the canonical workflow file."""
lower = (report_text or "").lower()
reasons = []
if not any(marker in lower for marker in RECONCILE_WORKFLOW_MARKERS):
reasons.append(
"reconciliation report missing workflow source "
"(workflows/reconcile-landed-pr.md)"
)
if not any(marker in lower for marker in RECONCILE_TASK_MARKERS):
reasons.append(
"reconciliation report missing task mode declaration "
"(reconcile-landed-pr)"
)
return {
"complete": not reasons,
"downgraded": bool(reasons),
"reasons": reasons,
}
+321
View File
@@ -0,0 +1,321 @@
"""Reviewer final-report schema verification before session output (#391).
Composes the composable ``assess_final_report_validator`` (#327) with
review-specific schema gates required before a reviewer session completes.
"""
from __future__ import annotations
import re
from typing import Any
from final_report_validator import (
assess_final_report_validator,
validator_finding,
_handoff_fields,
)
_LEGACY_STALE_FIELDS = (
"pinned reviewed head",
"scratch worktree used",
"workspace mutations",
)
_REVIEWED_HEAD_RE = re.compile(
r"(?:pinned reviewed head|reviewed head sha)\s*:\s*([0-9a-f]{7,40})",
re.IGNORECASE,
)
_VALIDATION_PASS_RE = re.compile(
r"validation\s*:\s*(?:pass|passed|strong|ok|green)",
re.IGNORECASE,
)
_MERGED_CLAIM_RE = re.compile(
r"\b(?:merged|merge result\s*:\s*merged|pr\s+#\d+\s+merged)\b",
re.IGNORECASE,
)
_MERGE_RESULT_RE = re.compile(r"merge result\s*:", re.IGNORECASE)
_ANCESTRY_PROOF_RE = re.compile(
r"(?:ancestor proof|target branch sha|merge commit)",
re.IGNORECASE,
)
_ISSUE_CLOSED_RE = re.compile(
r"(?:linked issue(?:\s+live)?\s+status\s*:\s*closed|issue\s+#\d+\s+closed)",
re.IGNORECASE,
)
_LIVE_ISSUE_PROOF_RE = re.compile(
r"gitea_view_issue|(?:issue|linked issue).{0,40}fetched live|live fetch proof",
re.IGNORECASE,
)
_FULL_SUITE_PASS_RE = re.compile(
r"(?:full suite passed|full test suite passed|\d+\s+passed,\s*0\s+failed)",
re.IGNORECASE,
)
_TESTS_IGNORED_RE = re.compile(
r"(?:ignored tests|tests?\s+ignored|skipped tests|not run|tests?\s+skipped)",
re.IGNORECASE,
)
_BLOCKED_HANDOFF_RE = re.compile(
r"(?:blocker\s*:|recovery handoff|infra_stop|capability stop|mutation blocked)",
re.IGNORECASE,
)
_REPLAY_COMMAND_RE = re.compile(
r"(?:gitea_submit_pr_review|gitea_merge_pr|gitea_review_pr|"
r"submitted\s+['\"]approve['\"]|replay approve|replay merge)",
re.IGNORECASE,
)
_PENDING_RE = re.compile(r"\bPENDING\b", re.IGNORECASE)
_APPROVED_RE = re.compile(
r"(?:review decision\s*:\s*approve|submitted\s+approve|official review submitted)",
re.IGNORECASE,
)
_DRY_RUN_RE = re.compile(r"dry[- ]run", re.IGNORECASE)
_FINDING_READY_RE = re.compile(
r"(?:finding ready|ready to submit|not yet submitted)",
re.IGNORECASE,
)
_NARRATIVE_APPROVE_RE = re.compile(
r"(?:^|\n)\s*(?:##\s+)?(?:summary|verdict|recommendation)\b[^\n]*\bapprove\b",
re.IGNORECASE | re.MULTILINE,
)
_HANDOFF_DECISION_RE = re.compile(
r"review decision\s*:\s*(\w+)",
re.IGNORECASE,
)
def _rule_legacy_stale_fields(report_text: str) -> list[dict[str, str]]:
fields = _handoff_fields(report_text)
findings: list[dict[str, str]] = []
for stale in _LEGACY_STALE_FIELDS:
if stale in fields and fields[stale].lower() not in {"", "none", "n/a"}:
findings.append(
validator_finding(
"reviewer.legacy_stale_field",
"block",
stale.title(),
f"legacy or stale handoff field '{stale}' must not appear in "
"canonical reviewer final reports",
"remove stale fields; use Candidate head SHA and precise "
"mutation categories instead",
)
)
return findings
def _rule_reviewed_head_without_validation(report_text: str) -> list[dict[str, str]]:
text = report_text or ""
if not _REVIEWED_HEAD_RE.search(text):
return []
if _VALIDATION_PASS_RE.search(text):
return []
return [
validator_finding(
"reviewer.reviewed_head_without_validation",
"block",
"Pinned reviewed head",
"reviewed/pinned head SHA reported without validation pass proof",
"document validation command, cwd, and pass result before pinning head SHA",
)
]
def _rule_merged_without_proof(report_text: str) -> list[dict[str, str]]:
text = report_text or ""
if not _MERGED_CLAIM_RE.search(text):
return []
if _MERGE_RESULT_RE.search(text) and _ANCESTRY_PROOF_RE.search(text):
return []
return [
validator_finding(
"reviewer.merged_without_proof",
"block",
"Merge result",
"merged outcome claimed without merge result and target ancestry proof",
"include Merge result, target branch SHA, and ancestor proof before claiming merged",
)
]
def _rule_issue_closed_without_live_proof(report_text: str) -> list[dict[str, str]]:
text = report_text or ""
if not _ISSUE_CLOSED_RE.search(text):
return []
fields = _handoff_fields(text)
status_value = fields.get("linked issue live status", "")
readonly_value = fields.get("read-only diagnostics", "")
proof_blob = f"{status_value} {readonly_value}".lower()
if _LIVE_ISSUE_PROOF_RE.search(proof_blob):
return []
return [
validator_finding(
"reviewer.issue_closed_without_live_proof",
"block",
"Linked issue",
"issue closed claimed without live linked-issue verification proof",
"fetch linked issue live (gitea_view_issue) and record Linked issue live status",
)
]
def _rule_full_suite_pass_ignored_tests(report_text: str) -> list[dict[str, str]]:
text = report_text or ""
if not (_FULL_SUITE_PASS_RE.search(text) and _TESTS_IGNORED_RE.search(text)):
return []
if re.search(r"explicitly\s+(?:ignored|skipped)", text, re.IGNORECASE):
return []
return [
validator_finding(
"reviewer.full_suite_pass_ignored_tests",
"block",
"Validation",
"full suite pass claimed while tests were ignored or skipped without "
"explicit disclosure",
"state which tests were ignored/skipped or downgrade validation verdict",
)
]
def _rule_blocked_handoff_replay(report_text: str) -> list[dict[str, str]]:
text = report_text or ""
if not _BLOCKED_HANDOFF_RE.search(text):
return []
if not _REPLAY_COMMAND_RE.search(text):
return []
return [
validator_finding(
"reviewer.blocked_handoff_replay",
"block",
"Safe next action",
"blocked recovery handoff includes direct approve or merge replay commands",
"restart the full workflow after the blocker clears; do not replay mutations",
)
]
def _rule_narrative_handoff_drift(report_text: str) -> list[dict[str, str]]:
text = report_text or ""
narrative_approve = bool(_NARRATIVE_APPROVE_RE.search(text))
decision_match = _HANDOFF_DECISION_RE.search(text)
if not narrative_approve or not decision_match:
return []
handoff_decision = decision_match.group(1).strip().lower()
if handoff_decision in {"approve", "approved"}:
return []
return [
validator_finding(
"reviewer.narrative_handoff_drift",
"block",
"Review decision",
f"narrative approves PR but controller handoff says '{handoff_decision}'",
"align narrative summary with controller handoff Review decision field",
)
]
def _rule_review_state_ambiguous(report_text: str) -> list[dict[str, str]]:
text = report_text or ""
findings: list[dict[str, str]] = []
if _PENDING_RE.search(text) and _APPROVED_RE.search(text):
findings.append(
validator_finding(
"reviewer.pending_vs_approved",
"block",
"Review decision",
"report conflates PENDING review state with APPROVED outcome",
"distinguish official submitted review from pending or ready-to-submit state",
)
)
if _DRY_RUN_RE.search(text) and _APPROVED_RE.search(text):
if "dry-run only" not in text.lower():
findings.append(
validator_finding(
"reviewer.dry_run_vs_submitted",
"block",
"Review mutations",
"dry-run language mixed with official approve/submitted claims",
"mark dry-run explicitly or document the live review mutation proof",
)
)
if _FINDING_READY_RE.search(text) and _APPROVED_RE.search(text):
findings.append(
validator_finding(
"reviewer.finding_ready_vs_submitted",
"downgrade",
"Review decision",
"finding-ready wording combined with submitted-approve claims",
"state whether review was submitted live or only prepared for submission",
)
)
return findings
_SCHEMA_RULES = (
_rule_legacy_stale_fields,
_rule_reviewed_head_without_validation,
_rule_merged_without_proof,
_rule_issue_closed_without_live_proof,
_rule_full_suite_pass_ignored_tests,
_rule_blocked_handoff_replay,
_rule_narrative_handoff_drift,
_rule_review_state_ambiguous,
)
def _merge_validator_results(
base: dict[str, Any],
extra_findings: list[dict[str, str]],
) -> dict[str, Any]:
findings = list(base.get("findings") or []) + extra_findings
blocked = base.get("blocked") or any(f["severity"] == "block" for f in extra_findings)
downgraded = base.get("downgraded") or any(
f["severity"] == "downgrade" for f in extra_findings
)
grade = base.get("grade", "A")
if blocked:
grade = "blocked"
elif downgraded and grade == "A":
grade = "downgraded"
reasons = [f"{f['rule_id']}: {f['reason']}" for f in findings]
safe_next = base.get("safe_next_action") or "none"
if extra_findings:
safe_next = extra_findings[0].get("safe_next_action", safe_next)
return {
**base,
"grade": grade,
"blocked": blocked,
"downgraded": downgraded,
"findings": findings,
"reasons": reasons,
"safe_next_action": safe_next,
"complete": grade == "A",
"schema_rules_applied": len(_SCHEMA_RULES),
}
def assess_review_final_report_schema(
report_text: str,
*,
review_decision_lock: dict | None = None,
linked_issue_lock: dict | None = None,
validation_report: dict | None = None,
action_log: list[dict] | None = None,
mutations_observed: bool = False,
local_edits: bool = False,
validation_session: dict | None = None,
) -> dict[str, Any]:
"""Validate reviewer final report text before session completion (#391)."""
base = assess_final_report_validator(
report_text,
"review_pr",
review_decision_lock=review_decision_lock,
linked_issue_lock=linked_issue_lock,
validation_report=validation_report,
action_log=action_log,
mutations_observed=mutations_observed,
local_edits=local_edits,
validation_session=validation_session,
)
extra: list[dict[str, str]] = []
for rule in _SCHEMA_RULES:
extra.extend(rule(report_text))
return _merge_validator_results(base, extra)
+332
View File
@@ -0,0 +1,332 @@
"""Enforced PR review/merge workflow state machine (#290).
Review and merge must advance through explicit states. Any failed upstream
gate forbids downstream approve/merge mutations until the full workflow is
restarted after blockers clear.
"""
from __future__ import annotations
import re
from typing import Any
REVIEW_MERGE_STATES: tuple[str, ...] = (
"PRECHECK",
"INVENTORY",
"SELECT_PR",
"PIN_HEAD_SHA",
"CREATE_BRANCHES_WORKTREE",
"VALIDATE",
"REVIEW_DECISION",
"APPROVE_OR_REQUEST_CHANGES",
"PRE_MERGE_RECHECK",
"MERGE",
"POST_MERGE_REPORT",
)
TERMINAL_BLOCKED_HEADING = (
"PR review/merge workflow blocked. Restart the full workflow after blockers clear."
)
_FORBIDDEN_RECOVERY_REPLAY_RE = re.compile(
r"\b(?:approve(?:\s+pr)?\s*#?\d+|merge(?:\s+pr)?\s*#?\d+|gitea_merge_pr|"
r"gitea_submit_pr_review|submit\s+approve|run\s+merge)\b",
re.IGNORECASE,
)
_RESTART_WORKFLOW_RE = re.compile(
r"restart(?:\s+the)?\s+full\s+workflow|rerun\s+the\s+full\s+workflow",
re.IGNORECASE,
)
_READY_TO_MERGE_RE = re.compile(
r"\bready\s+to\s+merge\b",
re.IGNORECASE,
)
_REVIEWED_VALIDATED_RE = re.compile(
r"\b(?:reviewed|validated|approval\s+submitted)\b",
re.IGNORECASE,
)
_PRE_MERGE_REQUIRED_GATES = (
"whoami_verified",
"profile_runtime_verified",
"merge_capability_verified",
"pr_refetched",
"reviewed_head_sha_unchanged",
"pr_mergeable",
"checks_passed",
"reviewer_not_author",
"worktree_clean",
)
def _clean(value: str | None) -> str:
return (value or "").strip()
def state_index(state: str) -> int:
name = _clean(state).upper()
try:
return REVIEW_MERGE_STATES.index(name)
except ValueError as exc:
raise ValueError(f"unknown review/merge state '{state}' (fail closed)") from exc
def downstream_states(from_state: str) -> list[str]:
idx = state_index(from_state)
return list(REVIEW_MERGE_STATES[idx + 1 :])
def _completed_through(state_completion: dict[str, bool], state: str) -> bool:
return bool(state_completion.get(state))
def _first_incomplete_state(state_completion: dict[str, bool]) -> str | None:
for state in REVIEW_MERGE_STATES:
if not _completed_through(state_completion, state):
return state
return None
def assess_workflow_blockers(
*,
infra_stop: bool = False,
capability_blocked: bool = False,
mcp_reconnect_failed: bool = False,
stale_capability_state: bool = False,
) -> dict[str, Any]:
"""Return hard blockers that forbid all PR queue work (#290 AC3)."""
reasons: list[str] = []
if infra_stop:
reasons.append("infra_stop is active; PR selection/review/merge is forbidden")
if capability_blocked:
reasons.append("gitea_resolve_task_capability returned blocked/stop_required")
if mcp_reconnect_failed:
reasons.append("MCP reconnect failed; stale session state cannot be reused")
if stale_capability_state:
reasons.append("stale MCP capability state detected after reconnect failure")
return {
"block": bool(reasons),
"reasons": reasons,
"forbidden_states": list(REVIEW_MERGE_STATES) if reasons else [],
"safe_next_action": (
TERMINAL_BLOCKED_HEADING if reasons else "proceed with PRECHECK"
),
}
def assess_state_advancement(
state_completion: dict[str, bool] | None,
*,
target_state: str,
infra_stop: bool = False,
capability_blocked: bool = False,
) -> dict[str, Any]:
"""Fail closed when *target_state* is requested before upstream gates pass."""
completion = dict(state_completion or {})
target = _clean(target_state).upper()
blockers = assess_workflow_blockers(
infra_stop=infra_stop,
capability_blocked=capability_blocked,
)
reasons = list(blockers["reasons"])
try:
target_idx = state_index(target)
except ValueError as exc:
return {
"target_state": target,
"allowed": False,
"block": True,
"reasons": [str(exc)],
"next_allowed_state": None,
"safe_next_action": TERMINAL_BLOCKED_HEADING,
}
if blockers["block"]:
return {
"target_state": target,
"allowed": False,
"block": True,
"reasons": reasons,
"next_allowed_state": None,
"safe_next_action": blockers["safe_next_action"],
}
next_allowed = _first_incomplete_state(completion)
if next_allowed is None:
allowed = target_idx == len(REVIEW_MERGE_STATES) - 1
if not allowed:
reasons.append("workflow already completed through POST_MERGE_REPORT")
else:
allowed = state_index(next_allowed) >= target_idx
if not allowed:
reasons.append(
f"state '{target}' is forbidden until '{next_allowed}' completes"
)
return {
"target_state": target,
"allowed": allowed and not reasons,
"block": bool(reasons),
"reasons": reasons,
"next_allowed_state": next_allowed,
"completed_states": [s for s in REVIEW_MERGE_STATES if completion.get(s)],
"safe_next_action": (
f"complete state '{next_allowed}' before advancing"
if next_allowed and not allowed
else (
TERMINAL_BLOCKED_HEADING
if reasons
else f"advance to {target}"
)
),
}
def can_approve(state_completion: dict[str, bool] | None, **blocker_kwargs) -> dict[str, Any]:
"""Approval requires all states through REVIEW_DECISION (#290 AC)."""
required = REVIEW_MERGE_STATES[: REVIEW_MERGE_STATES.index("REVIEW_DECISION") + 1]
completion = dict(state_completion or {})
advance = assess_state_advancement(
completion,
target_state="APPROVE_OR_REQUEST_CHANGES",
**blocker_kwargs,
)
missing = [s for s in required if not completion.get(s)]
reasons = list(advance["reasons"])
if missing:
reasons.append(
"approval blocked: incomplete states: " + ", ".join(missing)
)
block = bool(reasons) or advance["block"]
return {
"allowed": not block,
"block": block,
"reasons": reasons,
"missing_states": missing,
"safe_next_action": advance["safe_next_action"],
}
def can_merge(
state_completion: dict[str, bool] | None,
*,
pre_merge_gates: dict[str, bool] | None = None,
**blocker_kwargs,
) -> dict[str, Any]:
"""Merge requires approve path plus fresh PRE_MERGE_RECHECK gates (#290 AC6)."""
completion = dict(state_completion or {})
approve = can_approve(completion, **blocker_kwargs)
reasons = list(approve["reasons"])
if not completion.get("APPROVE_OR_REQUEST_CHANGES"):
reasons.append("merge blocked: APPROVE_OR_REQUEST_CHANGES not completed")
gate_map = dict(pre_merge_gates or {})
missing_gates = [g for g in _PRE_MERGE_REQUIRED_GATES if not gate_map.get(g)]
if missing_gates:
reasons.append(
"merge blocked: pre-merge gates incomplete: " + ", ".join(missing_gates)
)
advance = assess_state_advancement(
completion,
target_state="MERGE",
**blocker_kwargs,
)
reasons.extend(advance["reasons"])
block = bool(reasons)
return {
"allowed": not block,
"block": block,
"reasons": reasons,
"missing_pre_merge_gates": missing_gates,
"safe_next_action": (
"complete PRE_MERGE_RECHECK with fresh whoami/capability/PR re-fetch "
"before merge"
if block
else "merge allowed"
),
}
def assess_blocked_recovery_handoff(report_text: str) -> dict[str, Any]:
"""Blocked handoffs must not replay approve/merge commands (#290 AC4)."""
text = report_text or ""
reasons: list[str] = []
if _FORBIDDEN_RECOVERY_REPLAY_RE.search(text):
reasons.append(
"blocked recovery handoff contains direct approve/merge replay command"
)
if reasons and not _RESTART_WORKFLOW_RE.search(text):
reasons.append(
"blocked recovery handoff must direct operator to restart full workflow"
)
return {
"block": bool(reasons),
"allowed": not reasons,
"reasons": reasons,
"safe_next_action": (
"remove approve/merge replay commands and say to restart full workflow "
"after blockers clear"
if reasons
else "recovery handoff wording acceptable"
),
}
def assess_final_report_state_claims(
report_text: str,
*,
state_completion: dict[str, bool] | None = None,
approve_completed: bool = False,
merge_completed: bool = False,
) -> dict[str, Any]:
"""Reports must not claim reviewed/ready-to-merge without gate proof (#290 AC10)."""
text = report_text or ""
completion = dict(state_completion or {})
reasons: list[str] = []
if _READY_TO_MERGE_RE.search(text) and not (
merge_completed or completion.get("PRE_MERGE_RECHECK")
):
reasons.append(
"report claims ready-to-merge without PRE_MERGE_RECHECK completion"
)
if _REVIEWED_VALIDATED_RE.search(text):
if not (approve_completed or completion.get("VALIDATE")):
reasons.append(
"report claims reviewed/validated without VALIDATE/APPROVE proof"
)
return {
"block": bool(reasons),
"allowed": not reasons,
"reasons": reasons,
"safe_next_action": (
"remove stale reviewed/ready-to-merge claims unless gates passed"
if reasons
else "final report state claims consistent with workflow"
),
}
def workflow_status(
state_completion: dict[str, bool] | None,
**blocker_kwargs,
) -> dict[str, Any]:
"""Summarize current state-machine position for MCP/runtime reporting."""
completion = dict(state_completion or {})
blockers = assess_workflow_blockers(**blocker_kwargs)
next_state = _first_incomplete_state(completion)
return {
"states": list(REVIEW_MERGE_STATES),
"completed_states": [s for s in REVIEW_MERGE_STATES if completion.get(s)],
"next_required_state": next_state,
"workflow_complete": next_state is None,
"approve_allowed": can_approve(completion, **blocker_kwargs)["allowed"],
"merge_allowed": can_merge(completion, **blocker_kwargs)["allowed"],
"blockers": blockers,
}
+16 -85
View File
@@ -7,16 +7,16 @@ making any API call. Merge is handled solely by the gated `gitea_merge_pr` MCP
workflow (#16), which enforces identity/profile/eligibility, explicit
confirmation, expected head SHA checking, and self-merge protection.
Usage (review only):
Live review submission is also disabled (#211): use the gated
``gitea_submit_pr_review`` MCP workflow, which enforces validation-phase
dry-run, final decision marking, and single-terminal review mutation rules.
Usage (review only — disabled):
review_pr.py --pr-number 12 --event APPROVE --body "Approved and signed off"
"""
import os
import sys
import json
import base64
import argparse
import urllib.request
import urllib.error
# Auto-execute using the project's local virtual environment Python
PROJECT_ROOT = os.path.dirname(os.path.abspath(__file__))
@@ -24,7 +24,7 @@ venv_python = os.path.join(PROJECT_ROOT, "venv", "bin", "python3")
if os.path.exists(venv_python) and sys.executable != venv_python:
os.execv(venv_python, [venv_python] + sys.argv)
from gitea_auth import get_auth_header, resolve_remote, add_remote_args, api_request, repo_api_url, get_profile
from gitea_auth import add_remote_args
def main(argv=None):
@@ -42,95 +42,26 @@ def main(argv=None):
help="Ignored — CLI merge is disabled (see --merge).")
args = parser.parse_args(argv)
# Fail closed: direct CLI merge is disabled (#16). LLM automations were
# using this flag as an ungated merge bypass. Merge is only available via
# the gated `gitea_merge_pr` MCP workflow, which enforces
# identity/profile/eligibility, explicit confirmation, expected head SHA,
# and self-merge protection. No API call is made here.
if args.merge:
print(
"Direct CLI merge is disabled. Merge is only available through the "
"gated #16 workflow (MCP tool 'gitea_merge_pr'), which enforces "
"identity/profile/eligibility, explicit confirmation, expected head "
"SHA checking, and self-merge protection. Re-run without --merge to "
"submit a review only.",
"see the review-submission guard message.",
file=sys.stderr,
)
return 2
host, org, repo = resolve_remote(args)
# ── Reviewer mutation side-channel wall (#199, refs #194) ──
# The launching MCP session exports GITEA_SESSION_PROFILE_LOCK with the
# profile it was started with; child processes inherit it. If this CLI
# resolves a different profile — e.g. an ad-hoc GITEA_MCP_PROFILE
# override escalating an author-bound session to reviewer — refuse
# before any API call. No lock in the environment means no session
# context (direct operator CLI use), which stays allowed. Unlike a /tmp
# lock file, the environment is per-process-tree: other sessions cannot
# spoof it and it cannot go stale across sessions.
session_lock = (os.environ.get("GITEA_SESSION_PROFILE_LOCK") or "").strip()
if session_lock:
try:
cli_profile = (get_profile().get("profile_name") or "").strip()
except Exception as e:
print(f"Mutation authority check failed: {e}", file=sys.stderr)
return 3
if cli_profile != session_lock:
print(
f"Mismatched active profile vs session profile lock "
f"(CLI override rejected): CLI profile '{cli_profile}' does "
f"not match locked session profile '{session_lock}' "
f"(fail closed)",
file=sys.stderr,
)
return 3
body = args.body
if args.body_file:
if args.body_file == "-":
body = sys.stdin.read()
else:
with open(args.body_file, "r", encoding="utf-8") as fh:
body = fh.read()
auth = get_auth_header(host)
if not auth:
print(f"Could not get credentials or token for {host}.", file=sys.stderr)
return 1
# 1. Fetch PR to get the latest head commit SHA (required for review validation)
pr_url = f"{repo_api_url(host, org, repo)}/pulls/{args.pr_number}"
try:
pr_data = api_request("GET", pr_url, auth)
except Exception as e:
print(f"Error fetching PR #{args.pr_number}: {e}", file=sys.stderr)
return 1
commit_sha = pr_data.get("head", {}).get("sha")
if not commit_sha:
print(f"Could not find head commit SHA for PR #{args.pr_number}.", file=sys.stderr)
return 1
# 2. Submit the PR review
review_url = f"{repo_api_url(host, org, repo)}/pulls/{args.pr_number}/reviews"
payload = {
"body": body,
"event": args.event,
"commit_id": commit_sha
}
try:
api_request("POST", review_url, auth, payload)
print(f"Successfully submitted review for PR #{args.pr_number}: event={args.event}")
except Exception as e:
print(f"Error submitting review: {e}", file=sys.stderr)
return 1
# Merge is intentionally not performed here — see the fail-closed guard
# above. Use the gated `gitea_merge_pr` MCP workflow (#16) to merge.
return 0
print(
"Direct CLI review submission is disabled (#211). Use the gated "
"'gitea_submit_pr_review' MCP workflow, which enforces validation-phase "
"dry-run, gitea_mark_final_review_decision, and single-terminal review "
"mutation rules.",
file=sys.stderr,
)
return 2
if __name__ == "__main__":
sys.exit(main())
sys.exit(main())
+4489 -8
View File
File diff suppressed because it is too large Load Diff
+143
View File
@@ -0,0 +1,143 @@
"""Already-landed PR classification verifier for reviewer reports (#295)."""
from __future__ import annotations
import re
from typing import Any
ALREADY_LANDED_ELIGIBILITY_CLASS = "ALREADY_LANDED_RECONCILE_REQUIRED"
_LANDED_CONTEXT_RE = re.compile(
r"already[- ]landed|ancestor proof\s*:\s*passed|"
r"eligibility class\s*:\s*already_landed",
re.IGNORECASE,
)
_ELIGIBILITY_CLASS_LINE_RE = re.compile(
r"eligibility class\s*:\s*(.+)$",
re.IGNORECASE | re.MULTILINE,
)
_INELIGIBLE_SELECTION_RE = re.compile(
r"\b(?:next|oldest)\s+eligible\s+pr\b",
re.IGNORECASE,
)
_REVIEW_ELIGIBLE_RE = re.compile(
r"\beligible for (?:review|merge)\b|\bready for (?:review|merge)\b",
re.IGNORECASE,
)
_PINNED_REVIEWED_HEAD_RE = re.compile(
r"\bpinned reviewed head\s*:",
re.IGNORECASE,
)
_CANDIDATE_HEAD_RE = re.compile(
r"\bcandidate head sha\s*:",
re.IGNORECASE,
)
_REVIEWED_HEAD_NONE_RE = re.compile(
r"\breviewed head sha\s*:\s*none\b",
re.IGNORECASE,
)
_VALIDATION_PROOF_RE = re.compile(
r"(?:validation passed|pytest.*passed|diff review passed|"
r"validated on pinned head)",
re.IGNORECASE,
)
_QUEUE_BLOCKED_RE = re.compile(
r"queue blocked by already[- ]landed|"
r"reconciliation(?:-only)?\s+(?:next step|required)",
re.IGNORECASE,
)
def _landed_context(report_text: str, eligibility_class: str | None) -> bool:
if (eligibility_class or "").strip().upper() == ALREADY_LANDED_ELIGIBILITY_CLASS:
return True
return bool(_LANDED_CONTEXT_RE.search(report_text or ""))
def assess_already_landed_classification_report(
report_text: str,
*,
eligibility_class: str | None = None,
selected_pr_already_landed: bool | None = None,
) -> dict[str, Any]:
"""#295: already-landed PRs are reconciliation-only, not review eligible."""
text = report_text or ""
reasons: list[str] = []
landed = _landed_context(text, eligibility_class)
if selected_pr_already_landed is True:
landed = True
if not landed:
return {
"proven": True,
"block": False,
"landed_context": False,
"reasons": [],
"safe_next_action": "proceed",
}
if _INELIGIBLE_SELECTION_RE.search(text):
reasons.append(
"already-landed PR must not be described as oldest/next eligible PR; "
"use 'Oldest open PR requiring action' and "
f"'Eligibility class: {ALREADY_LANDED_ELIGIBILITY_CLASS}'"
)
if _REVIEW_ELIGIBLE_RE.search(text):
reasons.append(
"already-landed PR must not be described as eligible for review or merge"
)
class_match = _ELIGIBILITY_CLASS_LINE_RE.search(text)
if class_match:
declared = class_match.group(1).strip().upper()
if declared != ALREADY_LANDED_ELIGIBILITY_CLASS:
reasons.append(
"already-landed PR eligibility class must be "
f"{ALREADY_LANDED_ELIGIBILITY_CLASS}"
)
elif selected_pr_already_landed is True:
reasons.append(
f"already-landed selected PR must declare Eligibility class: "
f"{ALREADY_LANDED_ELIGIBILITY_CLASS}"
)
if _PINNED_REVIEWED_HEAD_RE.search(text):
if not _VALIDATION_PROOF_RE.search(text):
reasons.append(
"already-landed pre-review report must use Candidate head SHA, "
"not Pinned reviewed head, unless validation and diff review passed"
)
if selected_pr_already_landed is True and not _CANDIDATE_HEAD_RE.search(text):
if _PINNED_REVIEWED_HEAD_RE.search(text) or "head sha" in text.lower():
reasons.append(
"already-landed ancestry check must report Candidate head SHA"
)
if selected_pr_already_landed is True and not _REVIEWED_HEAD_NONE_RE.search(text):
if _PINNED_REVIEWED_HEAD_RE.search(text) and not _VALIDATION_PROOF_RE.search(text):
reasons.append(
"already-landed report must state 'Reviewed head SHA: none' "
"when validation did not run"
)
if selected_pr_already_landed is True and not _QUEUE_BLOCKED_RE.search(text):
reasons.append(
"already-landed queue blocker must state reconciliation requirement "
"or queue blocked wording"
)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"landed_context": True,
"reasons": reasons,
"safe_next_action": (
"classify as ALREADY_LANDED_RECONCILE_REQUIRED, use Candidate head SHA, "
"and stop normal review"
if not proven
else "proceed"
),
}
+216
View File
@@ -0,0 +1,216 @@
"""Already-landed controller handoff consistency verifier (#299)."""
from __future__ import annotations
import re
from typing import Any
from review_proofs import git_ref_mutating_commands
ALREADY_LANDED_STATE = "ALREADY_LANDED_RECONCILE_REQUIRED"
_HANDOFF_SECTION_RE = re.compile(r"^##\s*Controller Handoff\s*$", re.I | re.M)
_STALE_HANDOFF_FIELDS = (
"pinned reviewed head",
"scratch worktree used",
)
_LEGACY_MUTATIONS_NONE_RE = re.compile(
r"^\s*[-*]?\s*mutations\s*:\s*none\s*$",
re.I | re.M,
)
_LEGACY_WORKSPACE_NONE_RE = re.compile(
r"^\s*[-*]?\s*workspace\s+mutations\s*:\s*none\s*$",
re.I | re.M,
)
_FORBIDDEN_REVIEW_STATES_RE = re.compile(
r"review decision\s*:\s*approved?\b|"
r"merge result\s*:\s*merged\b|"
r"\bready[_ ]to[_ ]merge\b",
re.I,
)
_NARRATIVE_ELIGIBILITY_RE = re.compile(
r"eligibility class\s*:\s*([^\n]+)",
re.I,
)
_NARRATIVE_REVIEWED_SHA_RE = re.compile(
r"reviewed head sha\s*:\s*([^\n]+)",
re.I,
)
_NARRATIVE_WORKTREE_RE = re.compile(
r"review worktree used\s*:\s*([^\n]+)",
re.I,
)
def _handoff_field_map(report_text: str) -> dict[str, str]:
text = report_text or ""
match = _HANDOFF_SECTION_RE.search(text)
if not match:
return {}
fields: dict[str, str] = {}
for line in text[match.end() :].splitlines():
stripped = line.strip().lstrip("-*").strip()
if ":" not in stripped:
continue
key, value = stripped.split(":", 1)
fields[key.strip().lower()] = value.strip()
return fields
def _narrative_before_handoff(report_text: str) -> str:
text = report_text or ""
match = _HANDOFF_SECTION_RE.search(text)
if not match:
return text
return text[: match.start()]
def _is_truthy(value: str) -> bool:
lowered = (value or "").strip().lower()
return lowered not in {"", "none", "n/a", "not applicable", "false", "no", "", "-"}
def _gate_active(text: str, fields: dict[str, str], session: dict) -> bool:
if session.get("gate_fired") or session.get("already_landed_gate_fired"):
return True
eligibility = (fields.get("eligibility class") or "").upper()
if ALREADY_LANDED_STATE in eligibility:
return True
return ALREADY_LANDED_STATE.lower() in text.lower()
def assess_already_landed_handoff_report(
report_text: str,
*,
handoff_session: dict | None = None,
command_log: list | None = None,
) -> dict[str, Any]:
"""Reject stale handoff fields and narrative/handoff drift after the gate (#299)."""
text = report_text or ""
session = dict(handoff_session or {})
fields = _handoff_field_map(text)
reasons: list[str] = []
if not _gate_active(text, fields, session):
return {
"proven": True,
"block": False,
"reasons": [],
"gate_active": False,
"safe_next_action": "proceed",
}
for stale_field in _STALE_HANDOFF_FIELDS:
if stale_field in fields:
reasons.append(
f"already-landed handoff must not include stale field "
f"'{stale_field.title()}' (#299)"
)
if _LEGACY_WORKSPACE_NONE_RE.search(text):
reasons.append(
"already-landed handoff must not include legacy "
"'Workspace mutations: None' (#299)"
)
ref_commands = git_ref_mutating_commands(command_log or session.get("command_log"))
mutations_observed = bool(
ref_commands
or session.get("mutations_observed")
or session.get("mcp_mutations")
or session.get("review_mutations")
)
if mutations_observed and _LEGACY_MUTATIONS_NONE_RE.search(text):
reasons.append(
"already-landed handoff must not claim 'Mutations: None' when "
"git ref, MCP, review, merge, or cleanup mutations occurred (#299)"
)
git_ref_value = fields.get("git ref mutations", "").strip().lower()
if ref_commands and (not git_ref_value or git_ref_value == "none"):
reasons.append(
"git fetch/ref updates must be reported under Git ref mutations (#299)"
)
reviewed_sha = fields.get("reviewed head sha", "")
if reviewed_sha and _is_truthy(reviewed_sha):
reasons.append(
"already-landed handoff must use 'Reviewed head SHA: none' (#299)"
)
worktree_used = fields.get("review worktree used", "")
if worktree_used and _is_truthy(worktree_used):
reasons.append(
"already-landed handoff must set Review worktree used: false (#299)"
)
candidate_sha = fields.get("candidate head sha", "")
if not candidate_sha or candidate_sha.lower() in {"none", "n/a", "unknown"}:
reasons.append(
"already-landed handoff must include Candidate head SHA (#299)"
)
if _FORBIDDEN_REVIEW_STATES_RE.search(text):
reasons.append(
"already-landed handoff must not claim approved/merged/ready-to-merge (#299)"
)
narrative = _narrative_before_handoff(text)
handoff_eligibility = (fields.get("eligibility class") or "").strip()
narrative_eligibility = (
_NARRATIVE_ELIGIBILITY_RE.search(narrative).group(1).strip()
if _NARRATIVE_ELIGIBILITY_RE.search(narrative)
else ""
)
if handoff_eligibility and narrative_eligibility:
if handoff_eligibility.upper() != narrative_eligibility.upper():
reasons.append(
"narrative Eligibility class disagrees with controller handoff (#299)"
)
narrative_reviewed = (
_NARRATIVE_REVIEWED_SHA_RE.search(narrative).group(1).strip()
if _NARRATIVE_REVIEWED_SHA_RE.search(narrative)
else ""
)
if narrative_reviewed and reviewed_sha:
if narrative_reviewed.lower() != reviewed_sha.lower():
reasons.append(
"narrative Reviewed head SHA disagrees with controller handoff (#299)"
)
narrative_worktree = (
_NARRATIVE_WORKTREE_RE.search(narrative).group(1).strip()
if _NARRATIVE_WORKTREE_RE.search(narrative)
else ""
)
if narrative_worktree and worktree_used:
if narrative_worktree.lower() != worktree_used.lower():
reasons.append(
"narrative Review worktree used disagrees with controller handoff (#299)"
)
git_ref_narrative = "git ref mutations" in narrative.lower()
if git_ref_narrative and git_ref_value:
if "none" in git_ref_value and ref_commands:
reasons.append(
"narrative and handoff disagree on git ref mutation reporting (#299)"
)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": list(dict.fromkeys(reasons)),
"gate_active": True,
"safe_next_action": (
"emit canonical ALREADY_LANDED_RECONCILE_REQUIRED handoff without "
"stale review/merge fields; align narrative and controller handoff"
if reasons
else "proceed"
),
}
+164
View File
@@ -0,0 +1,164 @@
"""Prior-blocker skip proof verifier for reviewer queue reports (#318)."""
from __future__ import annotations
import re
from typing import Any
_FULL_SHA = re.compile(r"\b[0-9a-f]{40}\b", re.IGNORECASE)
_SHORT_SHA = re.compile(r"\b[0-9a-f]{7,39}\b", re.IGNORECASE)
_PR_NUMBER_RE = re.compile(r"(?:\bPR\s*#?|#)(\d+)\b", re.IGNORECASE)
_BLOCKING_DECISION_RE = re.compile(
r"(?:blocking review decision|review decision|blocking decision)\s*:\s*request[_ ]changes|"
r"request[_ ]changes",
re.IGNORECASE,
)
_BLOCKING_HEAD_RE = re.compile(
r"(?:blocking review head(?:\s+sha)?|blocker head(?:\s+sha)?|review head at blocker)",
re.IGNORECASE,
)
_HEAD_CHANGED_RE = re.compile(
r"(?:head (?:changed|unchanged)|head sha (?:changed|unchanged)|"
r"head changed (?:after|since) (?:the )?blocker|head unchanged since blocker)",
re.IGNORECASE,
)
_BLOCKER_REASON_RE = re.compile(
r"(?:reason (?:it )?remains blocked|remains blocked because|blocking category)",
re.IGNORECASE,
)
_LIVE_BLOCKER_PROOF_RE = re.compile(
r"(?:blocker revalidated live|live proof|gitea_get_pr_review_feedback|"
r"gitea_view_pr|review feedback fetched|current review state)",
re.IGNORECASE,
)
_BLOCKER_UNVERIFIED_RE = re.compile(r"\bBLOCKER_STATUS_UNVERIFIED\b")
def _pr_section(text: str, pr_number: int) -> str:
lines = (text or "").splitlines()
chunks: list[str] = []
capture = False
token = f"#{pr_number}"
for line in lines:
lower = line.lower()
if token in lower or f"pr {pr_number}" in lower or f"pr#{pr_number}" in lower.replace(" ", ""):
capture = True
chunks.append(line)
continue
if capture:
if _PR_NUMBER_RE.search(line) and token not in line.lower():
break
if line.strip() == "" and len(chunks) > 3:
break
chunks.append(line)
if chunks:
return "\n".join(chunks)
return text or ""
def _has_head_sha(text: str, head_sha: str | None) -> bool:
if not head_sha:
return bool(_FULL_SHA.search(text) or _SHORT_SHA.search(text))
head = head_sha.strip().lower()
if head in text.lower():
return True
if len(head) >= 7 and head[:7] in text.lower():
return True
return bool(_FULL_SHA.search(text))
def assess_prior_blocker_skip_proof(
report_text: str,
*,
skipped_prs: list[dict] | None = None,
) -> dict[str, Any]:
"""Validate live blocker proof for skipped earlier open PRs (#318)."""
text = report_text or ""
reasons: list[str] = []
assessments: list[dict[str, Any]] = []
for entry in skipped_prs or []:
pr_number = entry.get("pr_number")
if pr_number is None:
reasons.append("skipped PR entry missing pr_number")
continue
pr_number = int(pr_number)
section = _pr_section(text, pr_number)
verified = entry.get("blocker_verified")
head_changed = entry.get("head_changed_since_blocker")
blocking_head = (entry.get("blocking_review_head_sha") or "").strip() or None
current_head = (entry.get("head_sha") or "").strip() or None
skip_reason = (entry.get("skip_reason") or entry.get("blocking_decision") or "").lower()
item_reasons: list[str] = []
if head_changed is True:
item_reasons.append(
f"PR #{pr_number} head changed after blocker; it cannot be skipped "
"based on stale REQUEST_CHANGES"
)
if f"#{pr_number}" not in text.lower() and f"pr {pr_number}" not in text.lower():
item_reasons.append(f"skipped PR #{pr_number} not documented in final report")
if "request_changes" in skip_reason or entry.get("blocking_decision") == "request_changes":
if verified is False:
if not _BLOCKER_UNVERIFIED_RE.search(section):
item_reasons.append(
f"PR #{pr_number} blocker proof unavailable; report must classify "
"BLOCKER_STATUS_UNVERIFIED"
)
else:
if not _BLOCKING_DECISION_RE.search(section):
item_reasons.append(
f"PR #{pr_number} skip missing blocking review decision proof"
)
if not _has_head_sha(section, current_head):
item_reasons.append(
f"PR #{pr_number} skip missing current head SHA"
)
if blocking_head and not _BLOCKING_HEAD_RE.search(section):
if blocking_head.lower() not in section.lower():
item_reasons.append(
f"PR #{pr_number} skip missing blocking review head SHA"
)
if head_changed is False and not _HEAD_CHANGED_RE.search(section):
item_reasons.append(
f"PR #{pr_number} skip missing head-changed-since-blocker proof"
)
if not _BLOCKER_REASON_RE.search(section):
item_reasons.append(
f"PR #{pr_number} skip missing reason-it-remains-blocked"
)
if not _LIVE_BLOCKER_PROOF_RE.search(section):
item_reasons.append(
f"PR #{pr_number} skip missing live blocker proof for this session"
)
proven = not item_reasons
assessments.append({
"pr_number": pr_number,
"proven": proven,
"classification": (
"BLOCKER_STATUS_UNVERIFIED"
if verified is False
else "BLOCKED_SKIPPED"
),
"reasons": item_reasons,
})
reasons.extend(item_reasons)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"downgraded": False,
"assessments": assessments,
"reasons": reasons,
"safe_next_action": (
"fetch live review feedback and document blocker proof for each skipped PR, "
"or classify BLOCKER_STATUS_UNVERIFIED"
if reasons
else "proceed"
),
}
+213
View File
@@ -0,0 +1,213 @@
"""Reviewer local-fallback detection for normal PR review workflows (#324).
Normal reviewer runs must use MCP tools. Reading profile secret files or
running local Gitea helper scripts while MCP is available is a fail-closed
violation. Explicit recovery mode may use local fallback only with full proof.
"""
from __future__ import annotations
import re
from typing import Any
LOCAL_GITEA_SCRIPT_NAMES = (
"list_prs.py",
"view_pr.py",
"create_pr.py",
"merge_pr.py",
"edit_pr.py",
"list_issues.py",
"delete_branch.py",
"create_issue.py",
"close_issue.py",
"review_pr.py",
"mark_issue.py",
"mirror_refs.sh",
)
PROFILE_SECRET_MARKERS = (
"profiles.json",
".config/gitea-tools/profiles",
"gitea_auth.py",
"gitea_config.py",
"keychain",
"credential fill",
"token store",
)
_RECOVERY_MODE_RE = re.compile(
r"\b(?:recovery mode|explicit recovery|mcp unavailable|mcp not available|"
r"mcp tools unavailable|no mcp path)\b",
re.IGNORECASE,
)
_FALLBACK_CLASSIFICATION_RE = re.compile(
r"\b(?:local fallback|fallback mode|used local gitea|ran local script)\b",
re.IGNORECASE,
)
_MCP_TOOL_USE_RE = re.compile(
r"\b(?:gitea_list_prs|gitea_view_pr|gitea_review_pr|gitea_merge_pr|"
r"gitea_resolve_task_capability|gitea_whoami|mcp tool)\b",
re.IGNORECASE,
)
_LOCAL_SCRIPT_RE = re.compile(
r"(?:^|[\s\"'`/])(?:" + "|".join(re.escape(name) for name in LOCAL_GITEA_SCRIPT_NAMES) + r")",
re.IGNORECASE,
)
_PROFILE_ACCESS_RE = re.compile(
r"(?:read|open|inspect|cat|view|access(?:ed)?|loaded?)\s+(?:file\s+)?[`'\"]?"
r"[^`'\"]*profiles\.json|profiles\.json[`'\"]?\s+(?:read|opened|inspected|accessed)|"
r"~/?\.config/gitea-tools/profiles",
re.IGNORECASE,
)
_LOCAL_ENV_SCRIPT_RE = re.compile(
r"GITEA_MCP_(?:CONFIG|PROFILE)=.*\b(?:python\s+)?(?:list_prs|view_pr|create_pr|merge_pr)\.py",
re.IGNORECASE,
)
_RECOVERY_PROOF_FIELDS = (
("identity proof", ("identity proof", "exact identity proof")),
("profile proof", ("profile proof", "exact profile proof")),
("repo proof", ("repo proof", "exact repo proof")),
("capability proof", ("capability proof", "exact capability proof")),
("mcp unavailable reason", (
"why mcp was unavailable",
"mcp unavailable",
"mcp unavailability",
)),
)
def _collect_observed_text(
report_text: str,
action_log: list[dict] | None,
) -> str:
chunks = [report_text or ""]
for entry in action_log or []:
for key in ("command", "action", "detail", "path", "script"):
value = entry.get(key)
if value:
chunks.append(str(value))
return "\n".join(chunks)
def _detect_profile_secret_access(text: str) -> list[str]:
reasons = []
lower = text.lower()
if _PROFILE_ACCESS_RE.search(text):
reasons.append("report or action log shows profiles.json/profile secret access")
for marker in PROFILE_SECRET_MARKERS:
if marker == "profiles.json":
continue
if marker in lower and "do not" not in lower and "must not" not in lower:
if any(verb in lower for verb in ("read ", "open ", "inspect ", "cat ", "loaded ")):
reasons.append(f"profile secret surface '{marker}' accessed during review")
return reasons
def _detect_local_script_use(text: str) -> list[str]:
reasons = []
match = _LOCAL_SCRIPT_RE.search(text)
if match:
reasons.append(
f"local Gitea helper script invoked ({match.group(0).strip()})"
)
if _LOCAL_ENV_SCRIPT_RE.search(text):
reasons.append(
"local Gitea script run with GITEA_MCP_CONFIG/GITEA_MCP_PROFILE env overrides"
)
return reasons
def _recovery_proof_missing(text: str) -> list[str]:
lower = text.lower()
missing = []
for label, aliases in _RECOVERY_PROOF_FIELDS:
if not any(alias in lower for alias in aliases):
missing.append(label)
if not _FALLBACK_CLASSIFICATION_RE.search(text):
missing.append("fallback classification")
if not _RECOVERY_MODE_RE.search(text):
missing.append("recovery mode declaration")
return missing
def assess_reviewer_fallback_report(
report_text: str,
*,
action_log: list[dict] | None = None,
recovery_mode: bool = False,
mcp_available: bool = True,
mcp_tools_used: bool | None = None,
) -> dict[str, Any]:
"""Fail closed when normal reviewer workflows use local Gitea fallbacks (#324)."""
text = _collect_observed_text(report_text, action_log)
lower = (report_text or "").lower()
profile_violations = _detect_profile_secret_access(text)
script_violations = _detect_local_script_use(text)
violations = profile_violations + script_violations
if mcp_tools_used is None:
mcp_tools_used = bool(_MCP_TOOL_USE_RE.search(report_text or ""))
elif mcp_tools_used is False and _MCP_TOOL_USE_RE.search(report_text or ""):
mcp_tools_used = True
reasons: list[str] = []
recovery_declared = recovery_mode or bool(_RECOVERY_MODE_RE.search(report_text or ""))
if violations and mcp_available and not recovery_declared:
reasons.extend(violations)
if mcp_tools_used:
reasons.append(
"MCP tools were available and used; local fallback/profile access is forbidden"
)
else:
reasons.append(
"MCP path is available; normal review must not use local Gitea fallbacks"
)
if violations and recovery_declared:
missing = _recovery_proof_missing(report_text or "")
if missing:
reasons.append(
"recovery-mode fallback missing proof fields: "
+ ", ".join(missing)
)
if (
not violations
and recovery_declared
and _FALLBACK_CLASSIFICATION_RE.search(report_text or "")
and not recovery_mode
):
missing = _recovery_proof_missing(report_text or "")
if missing:
reasons.append(
"report claims local fallback but recovery proof is incomplete: "
+ ", ".join(missing)
)
proven = not reasons
blocked = bool(reasons) and not recovery_declared or any(
"recovery-mode fallback missing" in r or "recovery proof is incomplete" in r
for r in reasons
)
return {
"proven": proven,
"block": blocked or (bool(reasons) and mcp_available and not recovery_declared),
"downgraded": bool(reasons) and not blocked,
"recovery_mode": recovery_declared,
"mcp_available": mcp_available,
"mcp_tools_used": mcp_tools_used,
"profile_violations": profile_violations,
"script_violations": script_violations,
"reasons": reasons,
"safe_next_action": (
"use MCP reviewer tools only; do not read profiles.json or run local Gitea scripts"
if reasons and not recovery_declared
else (
"complete recovery-mode fallback proof before claiming local fallback"
if reasons
else "proceed with MCP tools"
)
),
}
+162
View File
@@ -0,0 +1,162 @@
"""Infra-stop repair handoff verifier for blocked reviewer workflows (#289)."""
from __future__ import annotations
import re
from typing import Any
from capability_stop_terminal import (
TERMINAL_REPORT_HEADING,
assess_capability_stop_report,
)
_REPAIR_HANDOFF_RE = re.compile(
r"repair handoff|control-checkout repair mode",
re.IGNORECASE,
)
_CONTROL_REPAIR_RE = re.compile(r"control-checkout repair mode", re.IGNORECASE)
_PR_QUEUE_ADVANCE_RE = re.compile(
r"(?:selected pr|pr #\d+ (?:to review|selected)|eligible pr|"
r"next (?:eligible )?pr(?: to review)?|oldest eligible pr|pinned (?:review )?head)",
re.IGNORECASE,
)
_REVIEW_MUTATION_RE = re.compile(
r"(?:gitea_review_pr|gitea_merge_pr|request_changes|submitted\s+approve|"
r"merge result|review decision\s*:\s*approve)",
re.IGNORECASE,
)
_BACKGROUND_TOOL_RE = re.compile(r"\b(?:schedule|manage_task)\b", re.IGNORECASE)
_PR_REVIEW_REPORT_RE = re.compile(
r"(?:review summary|merge recommendation|validation result\s*:\s*pass|"
r"queue status report with selected pr)",
re.IGNORECASE,
)
_DIAGNOSTIC_FIELDS = {
"mcp process root": re.compile(r"mcp process root\s*:", re.IGNORECASE),
"inspected git root": re.compile(r"inspected git root\s*:", re.IGNORECASE),
"conflict marker path": re.compile(
r"(?:conflict marker path|exact conflict marker path)\s*:",
re.IGNORECASE,
),
"merge/rebase control path": re.compile(
r"(?:merge/rebase control path|exact merge/rebase control path)\s*:",
re.IGNORECASE,
),
"safe next repair action": re.compile(
r"safe next (?:repair )?action\s*:",
re.IGNORECASE,
),
}
def assess_infra_stop_handoff_report(
report_text: str,
*,
stop_session: dict | None = None,
) -> dict[str, Any]:
"""Validate repair handoff purity when infra_stop blocks reviewer capability (#289)."""
text = report_text or ""
session = dict(stop_session or {})
reasons: list[str] = []
infra_stop = bool(
session.get("infra_stop")
or session.get("infra_stop_blocked")
or session.get("route_result") == "infra_stop"
)
if not infra_stop:
return {
"proven": True,
"block": False,
"reasons": [],
"infra_stop": False,
"safe_next_action": "proceed",
}
lower = text.lower()
repair_handoff = bool(
_REPAIR_HANDOFF_RE.search(text)
or TERMINAL_REPORT_HEADING.lower() in lower
)
if not repair_handoff:
reasons.append(
"infra_stop requires a repair handoff, not a PR review report"
)
if _PR_REVIEW_REPORT_RE.search(text) and not _REPAIR_HANDOFF_RE.search(text):
reasons.append(
"final output must be a repair handoff instead of stale PR review state"
)
if _PR_QUEUE_ADVANCE_RE.search(text):
reasons.append(
"infra_stop blocks PR queue advancement; do not select or advance next PR"
)
if _REVIEW_MUTATION_RE.search(text):
reasons.append(
"review, approval, request-changes, merge, or comment mutations "
"forbidden while infra_stop is active"
)
if session.get("pinned_head_sha") or re.search(
r"pinned (?:review )?head sha\s*:", text, re.IGNORECASE
):
if session.get("capability_cleared") is not True:
reasons.append(
"cannot pin PR head SHA while review capability never cleared"
)
if session.get("main_checkout_diagnostics") and not _CONTROL_REPAIR_RE.search(text):
reasons.append(
"main-checkout diagnostics must be labeled CONTROL-CHECKOUT REPAIR MODE"
)
if _BACKGROUND_TOOL_RE.search(text):
reasons.append(
"background schedule/manage_task tools must not be used during "
"blocked infra_stop recovery"
)
assessment = session.get("infra_stop_assessment") or {}
if assessment.get("infra_stop") or infra_stop:
missing = [
label
for label, pattern in _DIAGNOSTIC_FIELDS.items()
if not pattern.search(text)
]
if missing and session.get("require_infra_diagnostics", True):
reasons.append(
"infra_stop repair handoff missing diagnostic fields: "
+ ", ".join(missing)
)
conflict_file = assessment.get("conflict_file")
if conflict_file and conflict_file.lower() not in lower:
reasons.append(
f"infra_stop assessment conflict file {conflict_file!r} "
"not reflected in repair handoff"
)
capability = assess_capability_stop_report(
text,
trust_gate_status=session.get("trust_gate_status"),
capability_denied=True,
)
if not capability.get("pure"):
reasons.extend(capability.get("reasons") or [])
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": list(dict.fromkeys(reasons)),
"infra_stop": True,
"repair_handoff": repair_handoff,
"safe_next_action": (
"stop PR queue work; emit CONTROL-CHECKOUT REPAIR MODE handoff "
"with infra diagnostics and safe next repair action"
if reasons
else "proceed"
),
}
+308
View File
@@ -0,0 +1,308 @@
"""Reviewer inventory completeness and worktree state verifier (#293)."""
from __future__ import annotations
import re
from typing import Any
_PAGINATION_FINALITY_EVIDENCE = re.compile(
r"pagination_complete\s*:\s*true|inventory_complete\s*:\s*true|"
r"is_final_page\s*:\s*true|pages_fetched|has_more\s*:\s*false|"
r"no next page|final[- ]page|pr_inventory_trust_gate\.status|"
r"total_count\s*:|pagination.*(?:final|complete)",
re.I,
)
_INCOMPLETE_PAGINATION_PROOF = re.compile(
r"first page only|partial page|page 1 only|truncated|"
r"returned \d+ open prs?(?:\s*$|\s*[,;])",
re.I,
)
_DEFAULT_PAGE_SIZE_ASSUMPTION = re.compile(
r"(?:less|fewer) than (?:the )?(?:default )?(?:gitea )?page[- ]?(?:size|limit)|"
r"default (?:gitea )?page[- ]?size|under (?:the )?50[- ]?(?:item )?limit|"
r"(?:complete|exhaustive).*(?:default )?page[- ]?size|"
r"page[- ]?size assumption|assumed complete",
re.I,
)
_ELIGIBILITY_CLAIM_RE = re.compile(
r"\b(?:oldest eligible pr|next eligible pr|next pr to review|"
r"inventory (?:is )?complete|inventory exhaustive|exhaustive inventory)\b",
re.I,
)
_EXACT_PAGE_LIMIT_RE = re.compile(
r"(?:returned|listed|fetched)\s+(\d+)\s+open prs?|"
r"open pr count\s*:\s*(\d+)|"
r"page[- ]?size\s*(?:=|:)\s*(\d+)",
re.I,
)
_LEGACY_SCRATCH_FALSE_RE = re.compile(
r"scratch worktree used\s*:\s*false",
re.I,
)
_BRANCHES_WORKTREE_RE = re.compile(r"\bbranches/", re.I)
_CONTRADICTORY_HEAD_RE = re.compile(
r"detached head\s*/\s*branch\s+master|"
r"branch\s+master\s*/\s*detached head|"
r"detached head.*branch\s+(?:master|main|dev)\b.*detached|"
r"checkout\s*:\s*detached head\s*/\s*branch",
re.I,
)
_MAIN_CHECKOUT_BRANCH_RE = re.compile(
r"main checkout branch\s*:\s*(\S+)",
re.I,
)
_REVIEW_HEAD_STATE_RE = re.compile(
r"review worktree head state\s*:\s*(\S+)",
re.I,
)
_HANDOFF_SECTION_RE = re.compile(
r"^##\s*Controller Handoff\s*$",
re.I | re.M,
)
def _handoff_field_map(report_text: str) -> dict[str, str]:
"""Parse ``- Field: value`` lines from the Controller Handoff section."""
text = report_text or ""
match = _HANDOFF_SECTION_RE.search(text)
if not match:
return {}
section = text[match.end() :]
fields: dict[str, str] = {}
for line in section.splitlines():
stripped = line.strip().lstrip("-*").strip()
if ":" not in stripped:
continue
key, value = stripped.split(":", 1)
fields[key.strip().lower()] = value.strip()
return fields
def _truthy_field(value: str) -> bool:
lowered = (value or "").strip().lower()
if not lowered:
return False
if lowered in {"false", "no", "none", "not applicable", "n/a", "", "-"}:
return False
return True
def _field_proves_pagination(value: str) -> bool:
proof = (value or "").strip()
if not proof or proof.lower() in {"none", "n/a", "not applicable", "unknown"}:
return False
if _DEFAULT_PAGE_SIZE_ASSUMPTION.search(proof):
return False
if _INCOMPLETE_PAGINATION_PROOF.search(proof):
return False
return bool(_PAGINATION_FINALITY_EVIDENCE.search(proof))
def _pagination_proven(text: str, session: dict, *, pagination_field: str = "") -> bool:
if session.get("pagination_complete") or session.get("inventory_complete"):
return True
session_proof = session.get("inventory_pagination_proof") or ""
if _field_proves_pagination(str(session_proof)):
return True
if _field_proves_pagination(pagination_field):
return True
if _PAGINATION_FINALITY_EVIDENCE.search(text):
return True
return False
def _exact_page_limit_unproven(
text: str, session: dict, *, pagination_proven: bool = False
) -> bool:
"""True when a full page was returned without final-page proof."""
if pagination_proven:
return False
requested = session.get("requested_page_size")
returned = session.get("returned_page_size")
if isinstance(requested, int) and isinstance(returned, int):
if returned >= requested > 0:
return True
for match in _EXACT_PAGE_LIMIT_RE.finditer(text):
count = next((g for g in match.groups() if g), None)
if not count:
continue
try:
n = int(count)
except ValueError:
continue
if n in {10, 20, 50}:
return True
return False
def assess_inventory_worktree_report(
report_text: str,
*,
inventory_session: dict | None = None,
) -> dict[str, Any]:
"""Validate PR inventory pagination proof and worktree field consistency (#293)."""
text = report_text or ""
session = dict(inventory_session or {})
reasons: list[str] = []
fields = _handoff_field_map(text)
pagination_proof_field = fields.get("inventory pagination proof", "")
pagination_proven = _pagination_proven(
text, session, pagination_field=pagination_proof_field
)
if _ELIGIBILITY_CLAIM_RE.search(text):
if not pagination_proven:
reasons.append(
"oldest/next eligible PR or inventory-complete claim requires "
"final-page/no-next-page/pagination_complete proof"
)
if _DEFAULT_PAGE_SIZE_ASSUMPTION.search(text):
if not pagination_proven:
reasons.append(
"inventory pagination assumed from default page size without "
"final-page/no-next-page/total-count/traversal proof"
)
if pagination_proof_field:
if _DEFAULT_PAGE_SIZE_ASSUMPTION.search(pagination_proof_field):
reasons.append(
"Inventory pagination proof field relies on page-size assumption"
)
elif _INCOMPLETE_PAGINATION_PROOF.search(pagination_proof_field):
reasons.append(
"Inventory pagination proof field does not prove final page"
)
elif not _field_proves_pagination(pagination_proof_field):
if _ELIGIBILITY_CLAIM_RE.search(text) or _EXACT_PAGE_LIMIT_RE.search(text):
reasons.append(
"Inventory pagination proof field missing final-page metadata"
)
if _exact_page_limit_unproven(text, session, pagination_proven=pagination_proven):
reasons.append(
"exactly-full first page returned without final-page or "
"no-next-page proof"
)
review_worktree_used = fields.get("review worktree used", "")
review_worktree_path = fields.get("review worktree path", "")
scratch_used = fields.get("scratch worktree used", "")
inside_branches = fields.get("review worktree inside branches:", "") or fields.get(
"review worktree inside branches", ""
)
branches_path_in_text = bool(
_BRANCHES_WORKTREE_RE.search(review_worktree_path)
or _BRANCHES_WORKTREE_RE.search(text)
)
if branches_path_in_text:
if review_worktree_used and not _truthy_field(review_worktree_used):
reasons.append(
"reports using a branches/ review worktree must set "
"Review worktree used: true"
)
if scratch_used and re.search(r"\bfalse\b", scratch_used, re.I):
reasons.append(
"Scratch worktree used: false rejected when a branches/ "
"review worktree was created or used"
)
if _LEGACY_SCRATCH_FALSE_RE.search(text) and not _truthy_field(
review_worktree_used
):
reasons.append(
"legacy Scratch worktree used: false contradicts branches/ "
"review worktree usage"
)
if review_worktree_path and _truthy_field(review_worktree_used):
if "branches/" not in review_worktree_path.replace("\\", "/").lower():
reasons.append(
"Review worktree path must be under branches/ when "
"Review worktree used is true"
)
if inside_branches and re.search(r"\bfalse\b", inside_branches, re.I):
reasons.append(
"Review worktree inside branches must be true when path is "
"under branches/"
)
main_branch = (
fields.get("main checkout branch", "")
or _MAIN_CHECKOUT_BRANCH_RE.search(text).group(1)
if _MAIN_CHECKOUT_BRANCH_RE.search(text)
else ""
)
head_state = (
fields.get("review worktree head state", "")
or (
_REVIEW_HEAD_STATE_RE.search(text).group(1)
if _REVIEW_HEAD_STATE_RE.search(text)
else ""
)
)
if main_branch and head_state:
main_norm = main_branch.strip().lower()
head_norm = head_state.strip().lower()
if head_norm in {"branch", "on branch"} and main_norm in {
"master",
"main",
"dev",
}:
if "detached" in text.lower() and "review worktree" in text.lower():
reasons.append(
"review worktree HEAD state must not reuse main checkout "
"branch wording"
)
if _CONTRADICTORY_HEAD_RE.search(text):
reasons.append(
"contradictory checkout wording such as 'Detached HEAD / Branch master'"
)
if review_worktree_used and _truthy_field(review_worktree_used):
if not review_worktree_path or review_worktree_path.lower() in {
"none",
"n/a",
"not applicable",
}:
reasons.append(
"Review worktree used: true requires Review worktree path"
)
if not head_state or head_state.lower() in {
"none",
"n/a",
"not applicable",
"unknown",
}:
reasons.append(
"Review worktree used: true requires Review worktree HEAD state"
)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": list(dict.fromkeys(reasons)),
"pagination_proven": pagination_proven,
"branches_worktree_used": branches_path_in_text,
"safe_next_action": (
"prove inventory pagination with final-page metadata; align "
"Review worktree used/path/HEAD state with branches/ usage"
if reasons
else "proceed"
),
}
+148
View File
@@ -0,0 +1,148 @@
"""Merge-simulation worktree mutation verifier for reviewer reports (#317)."""
from __future__ import annotations
import re
from typing import Any
_WORKTREE_INDEX_FIELD_RE = re.compile(
r"^\s*worktree/index mutations\s*:\s*(.+?)\s*$",
re.IGNORECASE | re.MULTILINE,
)
_READONLY_DIAG_RE = re.compile(
r"read[- ]only diagnostics\s*:\s*(.+)$",
re.IGNORECASE | re.MULTILINE,
)
_WORKTREE_PATH_RE = re.compile(
r"(?:worktree path|diagnostic worktree|simulation worktree)\s*:",
re.IGNORECASE,
)
_PRE_CLEAN_RE = re.compile(
r"(?:pre[- ]simulation|before simulation|clean before).*(?:clean|status)",
re.IGNORECASE,
)
_MERGE_RESULT_RE = re.compile(
r"(?:merge result|simulation result|conflict status|merge conflict)",
re.IGNORECASE,
)
_ABORT_RE = re.compile(
r"(?:merge --abort|abort/reset command|abort command|git merge --abort)",
re.IGNORECASE,
)
_POST_CLEAN_RE = re.compile(
r"(?:post[- ]abort|after abort|clean after).*(?:clean|status)",
re.IGNORECASE,
)
def _command_text(entry: Any) -> str:
if isinstance(entry, dict):
return str(entry.get("command") or "").strip()
return str(entry or "").strip()
def _git_subcommand_line(command: str) -> str | None:
tokens = command.split()
if not tokens or tokens[0] != "git":
return None
return " ".join(tokens[1:])
def merge_simulation_commands(command_log: list | None) -> list[str]:
"""Return logged commands that perform local merge simulation (#317)."""
out: list[str] = []
for entry in command_log or []:
command = _command_text(entry)
rest = _git_subcommand_line(command)
if rest is None:
continue
lower = command.lower()
if rest.startswith("merge"):
if "--no-commit" in lower or (
"--no-ff" in lower and "merge" in lower and "--commit" not in lower
):
out.append(command)
elif "--abort" in lower:
out.append(command)
return out
def assess_merge_simulation_report(
report_text: str,
*,
command_log: list | None = None,
) -> dict[str, Any]:
"""Reject merge simulation classified as read-only; require worktree proof (#317)."""
text = report_text or ""
lower = text.lower()
reasons: list[str] = []
simulation_commands = merge_simulation_commands(command_log)
has_abort = any("--abort" in c.lower() for c in simulation_commands)
has_simulation = any(
"--no-commit" in c.lower() or (
"--no-ff" in c.lower() and "--abort" not in c.lower()
)
for c in simulation_commands
)
if not simulation_commands:
return {
"proven": True,
"block": False,
"reasons": [],
"simulation_commands": [],
"safe_next_action": "proceed",
}
field = _WORKTREE_INDEX_FIELD_RE.search(text)
value = (field.group(1).strip().lower() if field else "") or ""
if not value or value in {"none", "n/a", "no"}:
reasons.append(
"merge simulation commands ran but report lacks "
"'Worktree/index mutations' entry"
)
elif "merge" not in value and "simulation" not in value:
if not any(cmd.lower() in lower for cmd in simulation_commands):
reasons.append(
"Worktree/index mutations must document merge simulation commands"
)
readonly = _READONLY_DIAG_RE.search(text)
if readonly:
readonly_value = readonly.group(1)
for command in simulation_commands:
if command.lower() in readonly_value.lower():
reasons.append(
"merge simulation may not be listed under read-only diagnostics"
)
if has_simulation and "merge simulation" in readonly_value.lower():
reasons.append(
"merge simulation may not be classified as read-only diagnostics"
)
if has_simulation:
if not _WORKTREE_PATH_RE.search(text):
reasons.append("merge simulation report missing worktree path")
if not _PRE_CLEAN_RE.search(text):
reasons.append("merge simulation report missing pre-simulation clean status")
if not _MERGE_RESULT_RE.search(text):
reasons.append("merge simulation report missing merge result/conflict status")
if has_abort or has_simulation:
if has_abort and not _ABORT_RE.search(text):
reasons.append("merge simulation report missing abort/reset command proof")
if has_abort and not _POST_CLEAN_RE.search(text):
reasons.append("merge simulation report missing post-abort clean status")
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"simulation_commands": simulation_commands,
"safe_next_action": (
"classify merge simulation under Worktree/index mutations with full "
"pre/merge/abort/post-clean proof; never list as read-only diagnostics"
if reasons
else "proceed"
),
}
+159
View File
@@ -0,0 +1,159 @@
"""Non-mergeable PR skip conflict-proof verifier for reviewer reports (#322)."""
from __future__ import annotations
import re
from typing import Any
_FULL_SHA = re.compile(r"\b[0-9a-f]{40}\b", re.IGNORECASE)
_SHORT_SHA = re.compile(r"\b[0-9a-f]{7,39}\b", re.IGNORECASE)
_PR_NUMBER_RE = re.compile(r"(?:\bPR\s*#?|#)(\d+)\b", re.IGNORECASE)
_MERGEABILITY_FALSE_RE = re.compile(
r"(?:mergeable\s*:\s*false|not mergeable|non[- ]mergeable|mergeability\s*:\s*false|"
r"mergeability result\s*:\s*false)",
re.IGNORECASE,
)
_CONFLICT_PROOF_RE = re.compile(
r"(?:merge-tree|git merge-tree|gitea_view_pr|gitea_check_pr_eligibility|"
r"conflict proof|mergeability tool|merge simulation|conflicting files?)",
re.IGNORECASE,
)
_CONFLICTING_FILES_RE = re.compile(
r"(?:conflicting files?|conflict files?)\s*:\s*(.+)$",
re.IGNORECASE | re.MULTILINE,
)
_HEAD_CHANGED_RE = re.compile(
r"(?:head (?:changed|unchanged)|head sha (?:changed|unchanged)|"
r"head changed since|head unchanged since)",
re.IGNORECASE,
)
_MERGEABILITY_UNVERIFIED_RE = re.compile(
r"\bMERGEABILITY_UNVERIFIED\b",
)
def _pr_section(text: str, pr_number: int) -> str:
"""Return report lines likely describing one skipped PR."""
lines = (text or "").splitlines()
chunks: list[str] = []
capture = False
token = f"#{pr_number}"
for line in lines:
lower = line.lower()
if token in lower or f"pr {pr_number}" in lower or f"pr#{pr_number}" in lower.replace(" ", ""):
capture = True
chunks.append(line)
continue
if capture:
if _PR_NUMBER_RE.search(line) and token not in line.lower():
break
if line.strip() == "" and len(chunks) > 3:
break
chunks.append(line)
if chunks:
return "\n".join(chunks)
return text or ""
def _has_head_sha(text: str, head_sha: str | None) -> bool:
if not head_sha:
return bool(_FULL_SHA.search(text) or _SHORT_SHA.search(text))
head = head_sha.strip().lower()
if head in text.lower():
return True
if len(head) >= 7 and head[:7] in text.lower():
return True
return bool(_FULL_SHA.search(text))
def assess_non_mergeable_skip_proof(
report_text: str,
*,
skipped_prs: list[dict] | None = None,
) -> dict[str, Any]:
"""Validate skipped non-mergeable PR documentation in reviewer reports (#322)."""
text = report_text or ""
reasons: list[str] = []
assessments: list[dict[str, Any]] = []
for entry in skipped_prs or []:
pr_number = entry.get("pr_number")
if pr_number is None:
reasons.append("skipped PR entry missing pr_number")
continue
pr_number = int(pr_number)
section = _pr_section(text, pr_number)
mergeable = entry.get("mergeable")
verified = entry.get("mergeability_verified")
classification = (entry.get("classification") or "").strip().upper()
head_sha = (entry.get("head_sha") or "").strip() or None
item_reasons: list[str] = []
if f"#{pr_number}" not in text.lower() and f"pr {pr_number}" not in text.lower():
item_reasons.append(f"skipped PR #{pr_number} not documented in final report")
if mergeable is False or verified is False:
if not _MERGEABILITY_UNVERIFIED_RE.search(section) and verified is False:
if "MERGEABILITY_UNVERIFIED" not in classification:
item_reasons.append(
f"PR #{pr_number} conflict proof unavailable; report must classify "
"MERGEABILITY_UNVERIFIED"
)
elif verified is not False:
if not _MERGEABILITY_FALSE_RE.search(section):
item_reasons.append(
f"PR #{pr_number} skip missing mergeability:false proof"
)
if not _has_head_sha(section, head_sha):
item_reasons.append(
f"PR #{pr_number} skip missing current head SHA"
)
if not _CONFLICT_PROOF_RE.search(section):
item_reasons.append(
f"PR #{pr_number} skip missing conflict proof command/tool"
)
if entry.get("head_changed_since_prior_blocker") is not None:
if not _HEAD_CHANGED_RE.search(section):
item_reasons.append(
f"PR #{pr_number} skip missing head-changed-since-blocker proof"
)
if (
entry.get("conflicting_files")
and not _CONFLICTING_FILES_RE.search(section)
and not any(
path.lower() in section.lower()
for path in (entry.get("conflicting_files") or [])
)
):
item_reasons.append(
f"PR #{pr_number} has conflicting files in session proof but "
"report omits file-level conflict proof"
)
proven = not item_reasons
assessments.append({
"pr_number": pr_number,
"proven": proven,
"classification": classification or (
"MERGEABILITY_UNVERIFIED" if verified is False else "NON_MERGEABLE_SKIPPED"
),
"reasons": item_reasons,
})
reasons.extend(item_reasons)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"downgraded": False,
"assessments": assessments,
"reasons": reasons,
"safe_next_action": (
"document each skipped non-mergeable PR with head SHA, mergeability result, "
"conflict proof command, conflicting files, and head-change status"
if reasons
else "proceed"
),
}
+130
View File
@@ -0,0 +1,130 @@
"""Precise mutation category verifier for reviewer controller handoffs (#319)."""
from __future__ import annotations
import re
from typing import Any
_CONTROLLER_HANDOFF_RE = re.compile(r"controller handoff", re.IGNORECASE)
_WORKSPACE_MUTATIONS_RE = re.compile(
r"^\s*[-*]?\s*workspace\s+mutations\s*:",
re.IGNORECASE | re.MULTILINE,
)
_VAGUE_MUTATIONS_NONE_RE = re.compile(
r"^\s*[-*]?\s*mutations\s*:\s*none\s*$",
re.IGNORECASE | re.MULTILINE,
)
_REQUIRED_CATEGORIES = (
"file edits by reviewer",
"worktree/index mutations",
"git ref mutations",
)
_WORKTREE_FIELD_ALIASES = (
"worktree/index mutations",
"worktree mutations",
)
def _has_category(text: str, category: str) -> bool:
pattern = re.compile(
rf"^\s*[-*]?\s*{re.escape(category)}\s*:",
re.IGNORECASE | re.MULTILINE,
)
return bool(pattern.search(text))
def _has_worktree_category(text: str) -> bool:
return any(_has_category(text, alias) for alias in _WORKTREE_FIELD_ALIASES)
def _normalize_for_consistency_check(text: str) -> str:
"""Map #319 precise labels to #313 field names for consistency delegation."""
normalized = text
if _has_category(text, "worktree/index mutations") and not _has_category(
text, "worktree mutations"
):
normalized = re.sub(
r"(worktree/index mutations\s*:)",
"Worktree mutations:",
normalized,
flags=re.IGNORECASE,
)
return normalized
def assess_mutation_categories_report(
report_text: str,
*,
handoff_session: dict | None = None,
observed_commands: list | None = None,
) -> dict[str, Any]:
"""Require precise mutation categories in reviewer controller handoffs (#319)."""
text = report_text or ""
session = dict(handoff_session or {})
reasons: list[str] = []
lower = text.lower()
is_controller = bool(
session.get("controller_handoff")
or _CONTROLLER_HANDOFF_RE.search(text)
)
if not is_controller and not session.get("require_categories"):
return {
"proven": True,
"block": False,
"reasons": [],
"controller_handoff": False,
"safe_next_action": "proceed",
}
if _WORKSPACE_MUTATIONS_RE.search(text):
reasons.append(
"controller handoffs must not use legacy 'Workspace mutations'; "
"use precise mutation category fields"
)
if _VAGUE_MUTATIONS_NONE_RE.search(text):
reasons.append(
"controller handoffs must not use vague 'Mutations: none'; "
"report each precise category separately"
)
missing = [
category
for category in _REQUIRED_CATEGORIES
if category != "worktree/index mutations" and not _has_category(text, category)
]
if not _has_worktree_category(text):
missing.append("worktree/index mutations")
if missing:
reasons.append(
"controller handoff missing precise mutation categories: "
+ ", ".join(missing)
)
commands = observed_commands or session.get("observed_commands")
if commands:
from review_proofs import assess_workspace_mutation_consistency
consistency = assess_workspace_mutation_consistency(
_normalize_for_consistency_check(text),
commands,
)
if not consistency.get("complete"):
reasons.extend(consistency.get("reasons") or [])
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"controller_handoff": True,
"safe_next_action": (
"replace Workspace mutations with file edits, worktree/index, git ref, "
"and other precise mutation category fields"
if reasons
else "proceed"
),
}
+382
View File
@@ -0,0 +1,382 @@
"""Per-PR reviewer leases for safe parallel review sessions (#407)."""
from __future__ import annotations
import os
import re
import uuid
from datetime import datetime, timedelta, timezone
from typing import Any
MARKER = "<!-- mcp-review-lease:v1 -->"
_FIELD_RE = re.compile(
r"^\s*([a-z_]+)\s*:\s*(.+?)\s*$",
re.IGNORECASE | re.MULTILINE,
)
_FULL_SHA = re.compile(r"^[0-9a-f]{40}$", re.IGNORECASE)
_TERMINAL_PHASES = frozenset({"done", "released", "blocked"})
_ACTIVE_PHASES = frozenset({
"claimed",
"validating",
"approved",
"request-changes",
"merging",
})
DEFAULT_LEASE_TTL_MINUTES = 120
STALE_WARNING_MINUTES = 30
RECLAIMABLE_MINUTES = 60
_SESSION_LEASE: dict[str, Any] | None = None
def _parse_timestamp(value: str | None) -> datetime | None:
if not value:
return None
text = value.strip()
if text.endswith("Z"):
text = text[:-1] + "+00:00"
try:
parsed = datetime.fromisoformat(text)
except ValueError:
return None
if parsed.tzinfo is None:
return parsed.replace(tzinfo=timezone.utc)
return parsed.astimezone(timezone.utc)
def _normalize_sha(value: str | None) -> str | None:
text = (value or "").strip().lower()
return text if text and _FULL_SHA.match(text) else None
def _parse_pr_ref(value: str | None) -> int | None:
digits = re.sub(r"[^\d]", "", value or "")
return int(digits) if digits.isdigit() else None
def new_session_id() -> str:
return f"{os.getpid()}-{uuid.uuid4().hex[:12]}"
def format_lease_body(
*,
repo: str,
pr_number: int,
issue_number: int | None,
reviewer_identity: str,
profile: str,
session_id: str,
worktree: str,
phase: str,
candidate_head: str | None,
target_branch: str,
target_branch_sha: str | None,
last_activity: datetime | None = None,
expires_at: datetime | None = None,
blocker: str = "none",
) -> str:
now = last_activity or datetime.now(timezone.utc)
expires = expires_at or (now + timedelta(minutes=DEFAULT_LEASE_TTL_MINUTES))
last_text = now.astimezone(timezone.utc).replace(microsecond=0).isoformat().replace(
"+00:00", "Z"
)
expires_text = expires.astimezone(timezone.utc).replace(microsecond=0).isoformat().replace(
"+00:00", "Z"
)
issue_text = f"#{issue_number}" if issue_number else "none"
lines = [
MARKER,
f"repo: {repo}",
f"pr: #{pr_number}",
f"issue: {issue_text}",
f"reviewer_identity: {reviewer_identity}",
f"profile: {profile}",
f"session_id: {session_id}",
f"worktree: {worktree}",
f"phase: {phase}",
f"candidate_head: {candidate_head or 'none'}",
f"target_branch: {target_branch}",
f"target_branch_sha: {target_branch_sha or 'none'}",
f"last_activity: {last_text}",
f"expires_at: {expires_text}",
f"blocker: {blocker}",
]
return "\n".join(lines)
def parse_lease_comment(body: str) -> dict[str, Any] | None:
text = body or ""
if MARKER not in text:
return None
fields: dict[str, str] = {}
for match in _FIELD_RE.finditer(text):
fields[match.group(1).strip().lower()] = match.group(2).strip()
if not fields:
return None
return {
"repo": fields.get("repo"),
"pr_number": _parse_pr_ref(fields.get("pr")),
"issue_number": _parse_pr_ref(fields.get("issue")),
"reviewer_identity": fields.get("reviewer_identity"),
"profile": fields.get("profile"),
"session_id": fields.get("session_id"),
"worktree": fields.get("worktree"),
"phase": (fields.get("phase") or "").strip().lower() or None,
"candidate_head": _normalize_sha(fields.get("candidate_head")),
"target_branch": fields.get("target_branch"),
"target_branch_sha": _normalize_sha(fields.get("target_branch_sha")),
"last_activity": fields.get("last_activity"),
"expires_at": fields.get("expires_at"),
"blocker": fields.get("blocker"),
"raw_fields": fields,
}
def _lease_entries(comments: list[dict], *, pr_number: int) -> list[dict]:
entries: list[dict] = []
for comment in comments or []:
parsed = parse_lease_comment(comment.get("body") or "")
if not parsed:
continue
if parsed.get("pr_number") not in (None, pr_number):
continue
entries.append({
**parsed,
"comment_id": comment.get("id"),
"author": (comment.get("user") or {}).get("login") or comment.get("author"),
"created_at": comment.get("created_at"),
"updated_at": comment.get("updated_at"),
})
return entries
def _lease_expired(lease: dict, *, now: datetime) -> bool:
expires_at = _parse_timestamp(lease.get("expires_at"))
return bool(expires_at and expires_at <= now)
def _minutes_since_activity(lease: dict, *, now: datetime) -> float | None:
last = _parse_timestamp(lease.get("last_activity"))
if not last:
return None
return (now - last).total_seconds() / 60.0
def classify_lease_freshness(lease: dict, *, now: datetime | None = None) -> str:
"""Return active, stale_warning, reclaimable, expired, or terminal."""
now = now or datetime.now(timezone.utc)
phase = (lease.get("phase") or "").strip().lower()
if phase in _TERMINAL_PHASES:
return "terminal"
if _lease_expired(lease, now=now):
return "expired"
minutes = _minutes_since_activity(lease, now=now)
if minutes is None:
return "active"
if minutes >= RECLAIMABLE_MINUTES:
return "reclaimable"
if minutes >= STALE_WARNING_MINUTES:
return "stale_warning"
return "active"
def find_active_reviewer_lease(
comments: list[dict],
*,
pr_number: int,
now: datetime | None = None,
) -> dict[str, Any] | None:
"""Newest non-terminal, unexpired lease for *pr_number*."""
now = now or datetime.now(timezone.utc)
for lease in reversed(_lease_entries(comments, pr_number=pr_number)):
phase = (lease.get("phase") or "").strip().lower()
if phase in _TERMINAL_PHASES:
continue
if _lease_expired(lease, now=now):
continue
if phase in _ACTIVE_PHASES or phase:
lease = dict(lease)
lease["freshness"] = classify_lease_freshness(lease, now=now)
return lease
return None
def assess_acquire_lease(
comments: list[dict],
*,
pr_number: int,
reviewer_identity: str,
profile: str,
session_id: str,
repo: str,
issue_number: int | None,
worktree: str,
candidate_head: str | None,
target_branch: str,
target_branch_sha: str | None,
now: datetime | None = None,
) -> dict[str, Any]:
"""Fail closed when another session holds an active lease."""
now = now or datetime.now(timezone.utc)
reasons: list[str] = []
existing = find_active_reviewer_lease(comments, pr_number=pr_number, now=now)
if existing:
owner_session = (existing.get("session_id") or "").strip()
freshness = existing.get("freshness") or classify_lease_freshness(existing, now=now)
if owner_session and owner_session != session_id and freshness in {
"active", "stale_warning"
}:
reasons.append(
f"PR #{pr_number} already has active reviewer lease "
f"(session_id={owner_session}, phase={existing.get('phase')})"
)
elif owner_session and owner_session != session_id and freshness == "reclaimable":
reasons.append(
f"PR #{pr_number} lease is reclaimable but still held by "
f"session_id={owner_session}; explicit reclaim not implemented "
"(fail closed)"
)
if not (reviewer_identity or "").strip():
reasons.append("reviewer identity required for lease acquisition")
if not (session_id or "").strip():
reasons.append("session_id required for lease acquisition")
if not (worktree or "").strip():
reasons.append("worktree path required for lease acquisition")
allowed = not reasons
body = None
if allowed:
body = format_lease_body(
repo=repo,
pr_number=pr_number,
issue_number=issue_number,
reviewer_identity=reviewer_identity,
profile=profile,
session_id=session_id,
worktree=worktree,
phase="claimed",
candidate_head=candidate_head,
target_branch=target_branch,
target_branch_sha=target_branch_sha,
last_activity=now,
)
return {
"acquire_allowed": allowed,
"reasons": reasons,
"existing_lease": existing,
"lease_body": body,
"session_id": session_id,
}
def record_session_lease(lease: dict[str, Any]) -> dict[str, Any]:
global _SESSION_LEASE
_SESSION_LEASE = dict(lease)
return dict(_SESSION_LEASE)
def clear_session_lease() -> None:
global _SESSION_LEASE
_SESSION_LEASE = None
def get_session_lease() -> dict[str, Any] | None:
return dict(_SESSION_LEASE) if _SESSION_LEASE else None
def assess_mutation_lease_gate(
*,
pr_number: int,
comments: list[dict],
reviewer_identity: str,
session_id: str | None,
mutation: str,
live_head_sha: str | None,
pinned_head_sha: str | None,
now: datetime | None = None,
) -> dict[str, Any]:
"""Reviewer mutations require an owned, current PR lease."""
now = now or datetime.now(timezone.utc)
reasons: list[str] = []
session = get_session_lease()
active = find_active_reviewer_lease(comments, pr_number=pr_number, now=now)
if not session:
reasons.append(
f"no in-session reviewer lease recorded; acquire via "
f"gitea_acquire_reviewer_pr_lease before {mutation}"
)
elif session.get("pr_number") != pr_number:
reasons.append(
f"session lease is for PR #{session.get('pr_number')}, not #{pr_number}"
)
elif (session.get("session_id") or "") != (session_id or session.get("session_id")):
reasons.append("session lease session_id mismatch (fail closed)")
if active:
owner = (active.get("session_id") or "").strip()
if owner and session_id and owner != session_id:
reasons.append(
f"active PR lease owned by session_id={owner}; current session "
f"cannot {mutation}"
)
pinned = _normalize_sha(pinned_head_sha)
live = _normalize_sha(live_head_sha)
lease_head = active.get("candidate_head")
if pinned and live and pinned != live:
reasons.append(
"PR head changed during lease; stop and re-validate before "
f"reviewer {mutation}"
)
if lease_head and live and lease_head != live:
reasons.append(
"live PR head differs from lease candidate_head; refresh lease "
f"before {mutation}"
)
freshness = active.get("freshness") or classify_lease_freshness(active, now=now)
if freshness in {"expired", "reclaimable"}:
reasons.append(f"reviewer lease freshness is '{freshness}' (fail closed)")
else:
reasons.append(f"no active reviewer lease found on PR #{pr_number}")
allowed = not reasons
return {
"mutation_allowed": allowed,
"block": not allowed,
"reasons": reasons,
"active_lease": active,
"session_lease": session,
}
def assess_lease_inventory(
comments_by_pr: dict[int, list[dict]],
*,
now: datetime | None = None,
) -> dict[str, Any]:
"""Summarize lease states across PR comment threads."""
now = now or datetime.now(timezone.utc)
active: list[dict] = []
stale: list[dict] = []
reclaimable: list[dict] = []
for pr_number, comments in (comments_by_pr or {}).items():
lease = find_active_reviewer_lease(comments, pr_number=pr_number, now=now)
if not lease:
continue
freshness = lease.get("freshness") or classify_lease_freshness(lease, now=now)
entry = {"pr_number": pr_number, "session_id": lease.get("session_id"), "freshness": freshness}
if freshness == "stale_warning":
stale.append(entry)
elif freshness == "reclaimable":
reclaimable.append(entry)
else:
active.append(entry)
return {
"active_review_leases": active,
"stale_review_leases": stale,
"reclaimable_review_leases": reclaimable,
}
+217
View File
@@ -0,0 +1,217 @@
"""Proof-backed reviewer handoff claim verifier (#395)."""
from __future__ import annotations
import re
from typing import Any
_PAGINATION_FINALITY = re.compile(
r"has_more\s*:\s*false|is_final_page\s*:\s*true|"
r"inventory_complete\s*:\s*true|pages_fetched|total_count\s*:",
re.I,
)
_PAGE_SIZE_ONLY = re.compile(
r"(?:less|fewer) than (?:the )?(?:default )?page[- ]?size|"
r"under (?:the )?50[- ]?(?:item )?limit|"
r"returned \d+ (?:open )?prs?.*less than 50",
re.I,
)
_INVENTORY_COMPLETE_CLAIM = re.compile(
r"\b(?:inventory (?:is )?complete|inventory exhaustive|"
r"complete (?:pr )?inventory)\b",
re.I,
)
_SKIP_CLAIM = re.compile(
r"\b(?:earlier prs? skipped|skipped (?:earlier )?pr|"
r"skip(?:ped)? pr #?\d+|non-mergeable|prior request.changes)\b",
re.I,
)
_CONFLICT_PROOF = re.compile(
r"merge simulation|git merge --no-commit|conflicting files|"
r"conflict proof|merge_exit|non-mergeable",
re.I,
)
_BASELINE_CLAIM = re.compile(
r"\b(?:baseline (?:validation|worktree|comparison)|"
r"same as master|pre-existing (?:on )?master|"
r"failure signatures match)\b",
re.I,
)
_BASELINE_PATH = re.compile(r"baseline worktree path\s*:\s*(\S+)", re.I)
_BASELINE_SHA = re.compile(r"baseline (?:target )?sha\s*:\s*([0-9a-f]{7,40})", re.I)
_BASELINE_DIRTY_BEFORE = re.compile(
r"baseline.*dirty before|dirty before.*baseline", re.I
)
_BASELINE_DIRTY_AFTER = re.compile(
r"baseline.*dirty after|dirty after.*baseline", re.I
)
_BASELINE_COMMAND = re.compile(
r"baseline.*(?:validation )?command|pytest.*baseline", re.I
)
_BASELINE_RESULT = re.compile(
r"baseline.*(?:validation )?result|baseline_exit|baseline failures", re.I
)
_MASTER_INTEGRATION = re.compile(
r"\b(?:merged master into|merge(?:d)? (?:remote[- ]tracking )?branch.*master|"
r"master integration|integrated master|rebase.*master)\b",
re.I,
)
_CLEANUP_CLAIM = re.compile(
r"\b(?:worktree(?:s)? (?:were )?cleaned|cleanup (?:result|mutations)|"
r"removed (?:session[- ]owned )?worktree|git worktree remove)\b",
re.I,
)
_WORKTREE_LIST = re.compile(r"git worktree list|worktree list proof", re.I)
_PROOF_SOURCE = re.compile(
r"proof source\s*:\s*(command|mcp metadata|prior blocker|not checked)",
re.I,
)
_HANDOFF_SECTION = re.compile(r"^##\s*Controller Handoff\s*$", re.I | re.M)
def _handoff_fields(report_text: str) -> dict[str, str]:
text = report_text or ""
match = _HANDOFF_SECTION.search(text)
if not match:
return {}
fields: dict[str, str] = {}
for line in text[match.end() :].splitlines():
stripped = line.strip().lstrip("-*").strip()
if ":" not in stripped:
continue
key, value = stripped.split(":", 1)
fields[key.strip().lower()] = value.strip()
return fields
def _command_text(entry: Any) -> str:
if isinstance(entry, dict):
return str(entry.get("command") or entry.get("tool") or "").strip()
return str(entry or "").strip()
def _action_log_has(action_log: list | None, pattern: re.Pattern[str]) -> bool:
for entry in action_log or []:
blob = " ".join(
filter(
None,
[
_command_text(entry),
str(entry.get("result") or "") if isinstance(entry, dict) else "",
str(entry.get("reason") or "") if isinstance(entry, dict) else "",
],
)
)
if pattern.search(blob):
return True
return False
def assess_proof_backed_handoff_report(
report_text: str,
*,
action_log: list | None = None,
inventory_session: dict | None = None,
baseline_session: dict | None = None,
skip_session: list[dict] | None = None,
) -> dict[str, Any]:
"""Require explicit command/tool evidence for proof-sensitive review claims (#395)."""
text = report_text or ""
fields = _handoff_fields(text)
reasons: list[str] = []
inventory = dict(inventory_session or {})
baseline = dict(baseline_session or {})
skips = list(skip_session or [])
pagination_field = fields.get("inventory pagination proof", "")
if _INVENTORY_COMPLETE_CLAIM.search(text) or _PAGE_SIZE_ONLY.search(text):
has_meta = bool(
inventory.get("inventory_complete")
or inventory.get("pagination_complete")
or _PAGINATION_FINALITY.search(pagination_field)
or _PAGINATION_FINALITY.search(text)
or _action_log_has(action_log, re.compile(r"gitea_list_prs", re.I))
)
if _PAGE_SIZE_ONLY.search(text) and not has_meta:
reasons.append(
"inventory completeness claimed from page-size assumption only; "
"require has_more=false, is_final_page=true, or inventory_complete=true"
)
elif _INVENTORY_COMPLETE_CLAIM.search(text) and not has_meta:
reasons.append(
"inventory complete claim lacks explicit pagination metadata proof"
)
if _SKIP_CLAIM.search(text) or fields.get("earlier prs skipped", "").lower() not in {
"",
"none",
"n/a",
}:
skip_proof = bool(
skips
or _CONFLICT_PROOF.search(text)
or _action_log_has(
action_log, re.compile(r"git merge --no-commit|gitea_get_pr_review_feedback", re.I)
)
)
if not skip_proof:
reasons.append(
"earlier PR skip claim lacks command/tool conflict or blocker proof"
)
if _BASELINE_CLAIM.search(text) or fields.get("baseline worktree used", "").lower() == "true":
baseline_ok = bool(
baseline.get("complete")
or (
_BASELINE_PATH.search(text)
and _BASELINE_SHA.search(text)
and (_BASELINE_DIRTY_BEFORE.search(text) or baseline.get("clean_before"))
and (_BASELINE_COMMAND.search(text) or baseline.get("command"))
and (_BASELINE_RESULT.search(text) or baseline.get("result"))
and (_BASELINE_DIRTY_AFTER.search(text) or baseline.get("clean_after"))
)
)
if not baseline_ok:
reasons.append(
"baseline validation claim missing worktree path, target SHA, "
"dirty-before/after status, command, or result proof"
)
if _MASTER_INTEGRATION.search(text):
if not _action_log_has(
action_log,
re.compile(r"git merge.*master|git rebase.*master", re.I),
) and not re.search(r"merge_exit\s*=\s*\d+|merge simulation", text, re.I):
reasons.append(
"master integration claim lacks exact merge/rebase command and result"
)
if _CLEANUP_CLAIM.search(text) or fields.get("cleanup mutations", "").lower() not in {
"",
"none",
"n/a",
}:
if not (_WORKTREE_LIST.search(text) or _action_log_has(action_log, _WORKTREE_LIST)):
reasons.append(
"cleanup claim lacks final git worktree list or equivalent proof"
)
if re.search(r"\blive proof\b", text, re.I) and not _PROOF_SOURCE.search(text):
if not action_log:
reasons.append(
"live proof wording requires proof source classification "
"(command, MCP metadata, prior blocker, or not checked)"
)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"safe_next_action": (
"cite explicit command/tool evidence or structured MCP pagination metadata "
"for each proof-sensitive claim"
if not proven
else "proceed"
),
}
+162
View File
@@ -0,0 +1,162 @@
"""Reconciliation PR inventory pagination proof verifier (#308)."""
from __future__ import annotations
import re
from typing import Any
_COMPLETE_SCAN_CLAIM_RE = re.compile(
r"\b(?:all already[- ]landed(?:\s+prs?)?(?:\s+were)?\s+found|"
r"complete queue scan|all open prs? (?:were )?(?:found|checked|inventoried|listed)|"
r"inventory (?:is )?complete|exhaustive (?:pr )?inventory|"
r"no additional already[- ]landed|scanned (?:the )?(?:full )?open pr queue)\b",
re.I,
)
_FINALITY_RE = re.compile(
r"is_final_page\s*:\s*true|has_more\s*:\s*false|no next page|final[- ]page|"
r"pagination_complete\s*:\s*true|inventory pagination proof\s*:\s*(?!assumed|none\b).+",
re.I,
)
_REQUESTED_PAGE_SIZE_RE = re.compile(
r"requested page size\s*:\s*(\d+)|page[- ]?size\s*(?:=|:)\s*(\d+)",
re.I,
)
_PAGES_FETCHED_RE = re.compile(
r"pages? fetched\s*:\s*(\d+)|page count\s*:\s*(\d+)",
re.I,
)
_RETURNED_PER_PAGE_RE = re.compile(
r"returned pr count(?: per page)?\s*:|open pr count per page|"
r"returned \d+ open prs? per page|per[- ]page counts?\s*:",
re.I,
)
_EXACT_LIMIT_RETURN_RE = re.compile(
r"returned\s+(\d+)\s+open prs?|listed\s+(\d+)\s+open prs?",
re.I,
)
_DEFAULT_PAGE_SIZE_ASSUMPTION = re.compile(
r"(?:less|fewer) than (?:the )?(?:default )?(?:gitea )?page[- ]?(?:size|limit)|"
r"default (?:gitea )?page[- ]?size|assumed complete",
re.I,
)
def _int_from_groups(match: re.Match[str] | None) -> int | None:
if not match:
return None
for group in match.groups():
if group:
return int(group)
return None
def _pagination_proven(text: str, session: dict) -> bool:
if session.get("pagination_complete") or session.get("inventory_complete"):
return True
if _FINALITY_RE.search(text):
return True
pages = session.get("pages_fetched")
if isinstance(pages, int) and pages >= 1 and session.get("is_final_page"):
return True
return False
def _metadata_present(text: str, session: dict) -> list[str]:
missing: list[str] = []
has_page_size = bool(
_REQUESTED_PAGE_SIZE_RE.search(text)
or session.get("requested_page_size")
)
has_pages_fetched = bool(
_PAGES_FETCHED_RE.search(text) or session.get("pages_fetched")
)
has_returned_counts = bool(
_RETURNED_PER_PAGE_RE.search(text) or session.get("returned_per_page")
)
if not has_page_size:
missing.append("requested page size")
if not has_pages_fetched:
missing.append("page count fetched")
if not has_returned_counts:
missing.append("returned PR count per page")
if not _pagination_proven(text, session):
missing.append("final-page/no-next-page proof")
return missing
def assess_reconcile_inventory_report(
report_text: str,
*,
inventory_session: dict | None = None,
) -> dict[str, Any]:
"""Validate PR inventory pagination proof in reconciliation reports (#308)."""
text = report_text or ""
session = dict(inventory_session or {})
reasons: list[str] = []
claims_scan = bool(_COMPLETE_SCAN_CLAIM_RE.search(text))
lists_open_prs = bool(
re.search(r"open prs? listed|listed open prs?|pr inventory", text, re.I)
)
if not claims_scan and not lists_open_prs and not session.get("inventory_required"):
return {
"proven": True,
"block": False,
"reasons": [],
"inventory_claimed": False,
"safe_next_action": "proceed",
}
if _DEFAULT_PAGE_SIZE_ASSUMPTION.search(text) and not _pagination_proven(text, session):
reasons.append(
"reconciliation inventory assumed complete from page-size guess "
"without final-page proof (#308)"
)
if claims_scan and not _pagination_proven(text, session):
reasons.append(
"complete queue scan or all already-landed claim requires "
"pagination finality proof (#308)"
)
missing = _metadata_present(text, session)
if (claims_scan or lists_open_prs) and missing:
reasons.append(
"reconciliation inventory missing pagination metadata: "
+ ", ".join(missing)
)
requested = session.get("requested_page_size")
returned = session.get("returned_page_size")
if isinstance(requested, int) and isinstance(returned, int):
if returned >= requested > 0 and not _pagination_proven(text, session):
reasons.append(
"exactly-full first reconciliation page without final-page proof (#308)"
)
else:
for match in _EXACT_LIMIT_RETURN_RE.finditer(text):
count = _int_from_groups(match)
if count in {10, 20, 50} and not _pagination_proven(text, session):
reasons.append(
"exactly-full first reconciliation page without final-page proof (#308)"
)
break
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": list(dict.fromkeys(reasons)),
"inventory_claimed": claims_scan or lists_open_prs,
"pagination_proven": _pagination_proven(text, session),
"safe_next_action": (
"include requested page size, pages fetched, per-page counts, and "
"final-page/no-next-page proof before claiming complete scan"
if reasons
else "proceed"
),
}
+191
View File
@@ -0,0 +1,191 @@
"""Reconciliation linked-issue live proof verifier (#300)."""
from __future__ import annotations
import re
from typing import Any
_HANDOFF_SECTION_RE = re.compile(r"^##\s*Controller Handoff\s*$", re.I | re.M)
_LINKED_ISSUE_FIELD_RE = re.compile(
r"linked issue\s*:\s*(.+)$",
re.I | re.M,
)
_LINKED_ISSUE_STATUS_RE = re.compile(
r"linked issue(?:\s+live)?\s+status\s*:\s*(.+)$",
re.I | re.M,
)
_STATUS_CLAIM_RE = re.compile(
r"\b(?:issue\s+#?\d+\s*\()?(open|closed|resolved)\b|"
r"issue\s+(?:is\s+)?(open|closed|resolved)\b|"
r"linked issue.*\b(open|closed|resolved)\b",
re.I,
)
_NOT_VERIFIED_RE = re.compile(r"not verified in this session", re.I)
_LIVE_FETCH_PROOF_RE = re.compile(
r"gitea_view_issue|issue fetched live|live issue fetch|"
r"fetched linked issue.*(?:current|this) session|"
r"linked issue (?:fetch|proof).*(?:current|this) session|"
r"live linked issue proof",
re.I,
)
_ISSUE_ACTION_RECOMMEND_RE = re.compile(
r"(?:recommend(?:ed)?|should|must)\s+(?:close|reopen|comment on)\s+"
r"(?:the\s+)?(?:linked\s+)?issue|"
r"(?:close|reopen|comment on)\s+linked issue\s+#?\d+",
re.I,
)
_CAPABILITY_PROOF_RE = re.compile(
r"capability proof|exact capability|gitea_close_issue|"
r"gitea_mark_issue|gitea_create_issue_comment|gitea_edit_issue",
re.I,
)
_NO_LINKED_ISSUE_VALUES = frozenset({
"",
"none",
"n/a",
"not applicable",
"no linked issue",
"unknown",
})
def _handoff_field_map(report_text: str) -> dict[str, str]:
text = report_text or ""
match = _HANDOFF_SECTION_RE.search(text)
if not match:
return {}
fields: dict[str, str] = {}
for line in text[match.end() :].splitlines():
stripped = line.strip().lstrip("-*").strip()
if ":" not in stripped:
continue
key, value = stripped.split(":", 1)
fields[key.strip().lower()] = value.strip()
return fields
def _extract_linked_issue_number(text: str, fields: dict[str, str]) -> int | None:
linked = fields.get("linked issue", "")
if linked.lower() in _NO_LINKED_ISSUE_VALUES:
return None
match = re.search(r"#?(\d+)", linked)
if match:
return int(match.group(1))
field_match = _LINKED_ISSUE_FIELD_RE.search(text)
if field_match:
num = re.search(r"#?(\d+)", field_match.group(1))
if num:
return int(num.group(1))
return None
def _status_value(text: str, fields: dict[str, str]) -> str:
for key in ("linked issue live status", "linked issue status"):
if fields.get(key):
return fields[key]
match = _LINKED_ISSUE_STATUS_RE.search(text)
return match.group(1).strip() if match else ""
def _live_proof_present(text: str, session: dict, issue_number: int | None) -> bool:
if session.get("live_fetched") or session.get("verified_live"):
return True
if session.get("issue_fetch_proof"):
return True
if _LIVE_FETCH_PROOF_RE.search(text):
return True
if issue_number is not None:
pattern = re.compile(
rf"gitea_view_issue.*#?{issue_number}|"
rf"issue\s+#?{issue_number}.*(?:fetched|viewed).*(?:current|this) session",
re.I,
)
if pattern.search(text):
return True
return False
def assess_reconcile_linked_issue_report(
report_text: str,
*,
reconcile_session: dict | None = None,
) -> dict[str, Any]:
"""Validate linked issue status claims in reconciliation handoffs (#300)."""
text = report_text or ""
session = dict(reconcile_session or {})
fields = _handoff_field_map(text)
reasons: list[str] = []
issue_number = session.get("linked_issue_number")
if issue_number is None:
issue_number = _extract_linked_issue_number(text, fields)
if issue_number is None:
status = _status_value(text, fields)
if status and status.lower() not in _NO_LINKED_ISSUE_VALUES:
if _STATUS_CLAIM_RE.search(status):
reasons.append(
"linked issue status reported without identifying the linked issue (#300)"
)
if not reasons:
return {
"proven": True,
"block": False,
"reasons": [],
"linked_issue_number": None,
"safe_next_action": "proceed",
}
status = _status_value(text, fields)
status_lower = status.lower()
if session.get("issue_missing"):
if status_lower and "not verified" not in status_lower:
reasons.append(
"missing linked issue must be reported as "
"'not verified in this session' (#300)"
)
elif status:
if _NOT_VERIFIED_RE.search(status):
pass
elif _STATUS_CLAIM_RE.search(status):
if not _live_proof_present(text, session, issue_number):
reasons.append(
"linked issue open/closed/resolved status requires live "
"gitea_view_issue proof in the current session (#300)"
)
elif status_lower not in _NO_LINKED_ISSUE_VALUES:
if not _live_proof_present(text, session, issue_number):
reasons.append(
"linked issue status claim requires live fetch proof or "
"'not verified in this session' (#300)"
)
if _ISSUE_ACTION_RECOMMEND_RE.search(text):
if not _CAPABILITY_PROOF_RE.search(text) and not session.get(
"issue_action_capability_proven"
):
reasons.append(
"recommended linked issue close/reopen/comment requires "
"exact capability proof (#300)"
)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": list(dict.fromkeys(reasons)),
"linked_issue_number": issue_number,
"live_proof_present": _live_proof_present(text, session, issue_number),
"safe_next_action": (
"fetch linked issue with gitea_view_issue in this session or "
"report 'Linked issue status: not verified in this session'"
if reasons
else "proceed"
),
}
+233
View File
@@ -0,0 +1,233 @@
"""Explicit worktree and cwd proof for PR review validation (#398)."""
from __future__ import annotations
import re
from typing import Any
_FULL_SHA = re.compile(r"^[0-9a-f]{40}$", re.IGNORECASE)
_PWD_RE = re.compile(
r"(?:^|\n)\s*(?:pwd|working\s+directory|cwd)\s*:\s*(\S+)",
re.IGNORECASE,
)
_HEAD_RE = re.compile(
r"(?:git\s+rev-parse\s+head|observed\s+head\s+sha)\s*:\s*([0-9a-f]{7,40})",
re.IGNORECASE | re.MULTILINE,
)
_EXPECTED_HEAD_RE = re.compile(
r"(?:expected\s+(?:pr\s+)?head\s+sha|candidate\s+head\s+sha|pinned\s+head)\s*:\s*([0-9a-f]{7,40})",
re.IGNORECASE,
)
_STATUS_RE = re.compile(
r"git\s+status\s+(?:--short\s+--branch|--short|-sb)\s*:\s*(.+)",
re.IGNORECASE,
)
_VALIDATION_CMD_RE = re.compile(
r"validation\s+command\s*:\s*(.+)",
re.IGNORECASE,
)
_GIT_C_CMD_RE = re.compile(r"git\s+-C\s+\S+", re.IGNORECASE)
_CD_CMD_RE = re.compile(r"(?:^|&&\s*)cd\s+\S+", re.IGNORECASE)
_BASELINE_CWD_RE = re.compile(
r"baseline\s+(?:worktree|working\s+directory|cwd)\s*:\s*(\S+)",
re.IGNORECASE,
)
_BASELINE_SHA_RE = re.compile(
r"baseline\s+(?:target\s+)?sha\s*:\s*([0-9a-f]{7,40})",
re.IGNORECASE,
)
_BASELINE_CMD_RE = re.compile(
r"baseline\s+validation\s+command\s*:\s*(.+)",
re.IGNORECASE,
)
def _normalize_path(path: str) -> str:
return (path or "").replace("\\", "/").rstrip("/")
def _path_under_branches(path: str, project_root: str | None = None) -> bool:
normalized = _normalize_path(path)
if not normalized:
return False
if "/branches/" in f"{normalized}/":
return True
if normalized.endswith("/branches"):
return True
if project_root:
root = _normalize_path(project_root)
if normalized.startswith(f"{root}/"):
rel = normalized[len(root) + 1 :]
return rel == "branches" or rel.startswith("branches/")
return False
def _expand_sha(sha: str) -> str:
return (sha or "").strip().lower()
def _sha_matches(expected: str, observed: str) -> bool:
exp = _expand_sha(expected)
obs = _expand_sha(observed)
if not exp or not obs:
return False
if len(exp) == 40 and len(obs) == 40:
return exp == obs
return obs.startswith(exp) or exp.startswith(obs)
def _command_has_explicit_cwd(command: str, cwd: str) -> bool:
text = (command or "").strip()
if not text:
return False
if _GIT_C_CMD_RE.search(text):
return True
if _CD_CMD_RE.search(text):
return True
if cwd and cwd in text:
return True
return False
def assess_validation_cwd_proof_report(
report_text: str,
*,
validation_session: dict | None = None,
project_root: str | None = None,
) -> dict[str, Any]:
"""Require cwd/HEAD proof before reviewer validation claims (#398)."""
text = report_text or ""
session = dict(validation_session or {})
reasons: list[str] = []
violations: list[str] = []
claims_validation = bool(
session.get("validation_ran")
or _VALIDATION_CMD_RE.search(text)
or session.get("command")
)
if not claims_validation:
return {
"proven": True,
"block": False,
"claims_validation": False,
"reasons": [],
"violations": [],
"safe_next_action": "proceed",
}
expected_head = (
session.get("expected_head_sha")
or session.get("candidate_head_sha")
or ""
).strip()
if not expected_head:
match = _EXPECTED_HEAD_RE.search(text)
expected_head = (match.group(1) if match else "").strip()
observed_head = (session.get("observed_head_sha") or "").strip()
if not observed_head:
match = _HEAD_RE.search(text)
observed_head = (match.group(1) if match else "").strip()
cwd = (
session.get("working_directory")
or session.get("cwd")
or session.get("pwd")
or ""
).strip()
if not cwd:
match = _PWD_RE.search(text)
cwd = (match.group(1) if match else "").strip().rstrip(",.;")
command = (session.get("command") or "").strip()
if not command:
match = _VALIDATION_CMD_RE.search(text)
command = (match.group(1) if match else "").strip().rstrip(".;")
if not cwd:
reasons.append(
"validation claimed without pwd/working-directory proof (#398)"
)
elif not _path_under_branches(cwd, project_root):
violations.append(
f"validation cwd {cwd!r} is not under branches/ (#398)"
)
reasons.append(
"reviewer validation must run from a branches/ worktree, "
"not the main checkout (#398)"
)
if not observed_head:
reasons.append(
"validation claimed without git rev-parse HEAD / observed HEAD SHA "
"proof (#398)"
)
elif expected_head and not _sha_matches(expected_head, observed_head):
violations.append(
f"observed HEAD {observed_head} does not match expected "
f"PR head {expected_head} (#398)"
)
reasons.append("validation HEAD SHA must match pinned PR head (#398)")
if not _STATUS_RE.search(text) and session.get("git_status") is None:
reasons.append(
"validation claimed without git status --short --branch proof (#398)"
)
if command and cwd and not _command_has_explicit_cwd(command, cwd):
if session.get("tool_working_directory") is not True:
reasons.append(
"validation command must use git -C <worktree>, "
"cd <worktree> && ..., or tool-provided cwd metadata (#398)"
)
baseline_ran = bool(
session.get("baseline_validation_ran")
or _BASELINE_CMD_RE.search(text)
)
if baseline_ran:
baseline_cwd = (session.get("baseline_worktree_path") or "").strip()
if not baseline_cwd:
match = _BASELINE_CWD_RE.search(text)
baseline_cwd = (match.group(1) if match else "").strip().rstrip(",.;")
if not baseline_cwd or not _path_under_branches(baseline_cwd, project_root):
reasons.append(
"baseline validation claimed without baseline worktree cwd "
"under branches/ (#398)"
)
baseline_sha = (session.get("baseline_target_sha") or "").strip()
if not baseline_sha:
match = _BASELINE_SHA_RE.search(text)
baseline_sha = (match.group(1) if match else "").strip()
if not baseline_sha:
reasons.append(
"baseline validation claimed without baseline target SHA (#398)"
)
baseline_cmd = (session.get("baseline_command") or "").strip()
if not baseline_cmd:
match = _BASELINE_CMD_RE.search(text)
baseline_cmd = (match.group(1) if match else "").strip()
if not baseline_cmd:
reasons.append(
"baseline validation claimed without exact baseline command (#398)"
)
proven = not reasons and not violations
return {
"proven": proven,
"block": bool(violations) or not proven,
"claims_validation": True,
"expected_head_sha": expected_head or None,
"observed_head_sha": observed_head or None,
"working_directory": cwd or None,
"reasons": reasons,
"violations": violations,
"safe_next_action": (
"before validation record pwd, git rev-parse HEAD, git status, "
"expected PR head SHA; run commands with git -C or cd in the same line"
if not proven
else "proceed"
),
}
+198
View File
@@ -0,0 +1,198 @@
"""Transient validation failure history verifier (#396).
Reviewer sessions may observe validation failures that later pass on rerun.
Final reports must document every failure observed during the session, not
only the last passing result.
"""
from __future__ import annotations
import re
from typing import Any
_SECTION_RE = re.compile(
r"validation failure history",
re.IGNORECASE,
)
_FAILURE_ENTRY_RE = re.compile(
r"(?:failure\s*(?:#|entry)?\s*\d+|validation failure)\s*:",
re.IGNORECASE,
)
_COMMAND_RE = re.compile(
r"(?:command|validation command)\s*:\s*(.+)",
re.IGNORECASE,
)
_FAILING_TEST_RE = re.compile(
r"(?:failing test|failure|error)\s*:\s*(.+)",
re.IGNORECASE,
)
_CAUSE_RE = re.compile(
r"(?:suspected cause|cause)\s*:\s*(.+)",
re.IGNORECASE,
)
_REPRODUCED_RE = re.compile(
r"(?:reproduced|reproduces)\s*:\s*(yes|no|unknown)",
re.IGNORECASE,
)
_BASELINE_RE = re.compile(
r"(?:on baseline master|baseline master|exists on baseline)\s*:\s*(yes|no|unknown|not checked)",
re.IGNORECASE,
)
_PR_CAUSED_RE = re.compile(
r"(?:pr[- ]caused|pr caused)\s*:\s*(yes|no|unknown|not proven)",
re.IGNORECASE,
)
_TRANSIENT_STATUS_RE = re.compile(
r"(?:passed after transient failure investigation|transient failure investigation)",
re.IGNORECASE,
)
_PLAIN_PASS_RE = re.compile(
r"validation\s*:\s*(?:pass|passed|strong|ok|green)\b",
re.IGNORECASE,
)
_ENV_CLEANUP_RE = re.compile(
r"(?:environmental cleanup|state cleaned|what changed between runs)\s*:",
re.IGNORECASE,
)
_UNKNOWN_CAUSE_RE = re.compile(
r"(?:suspected cause|cause)\s*:\s*(?:unknown|unexplained|not determined)",
re.IGNORECASE,
)
def _failure_documented_in_text(text: str, failure: dict[str, Any]) -> bool:
"""Return True when *failure* appears documented in free-form report text."""
command = (failure.get("command") or "").strip()
failing = (failure.get("failing_test") or failure.get("error") or "").strip()
if command and command not in text:
return False
if failing and failing not in text:
return False
return bool(command or failing)
def _section_has_structured_fields(text: str) -> bool:
if not _SECTION_RE.search(text):
return False
section_start = _SECTION_RE.search(text).start()
section = text[section_start:]
has_command = bool(_COMMAND_RE.search(section))
has_failure = bool(_FAILING_TEST_RE.search(section))
return has_command and has_failure
def assess_validation_failure_history_report(
report_text: str,
*,
validation_session: dict | None = None,
) -> dict[str, Any]:
"""Require final reports to account for transient validation failures (#396)."""
text = report_text or ""
session = dict(validation_session or {})
reasons: list[str] = []
violations: list[str] = []
observed = list(session.get("observed_failures") or [])
final_status = (session.get("final_validation_status") or "").strip().lower()
cause_unknown = any(
(f.get("suspected_cause") or "").strip().lower() in {
"unknown", "unexplained", "not determined", ""
}
and not (f.get("pr_caused") or "").strip().lower() in {"yes", "no"}
for f in observed
) or bool(session.get("cause_unknown"))
if not observed:
return {
"proven": True,
"block": False,
"reasons": [],
"violations": [],
"observed_failure_count": 0,
"safe_next_action": "proceed",
}
documented = _section_has_structured_fields(text)
if not documented:
for failure in observed:
if _failure_documented_in_text(text, failure):
documented = True
break
if not documented:
violations.append(
"session observed validation failure(s) but final report omits "
"Validation failure history"
)
reasons.append(
"every validation failure observed during the session must appear "
"in a Validation failure history section"
)
if documented and not _SECTION_RE.search(text):
reasons.append(
"failure details must appear under an explicit "
"'Validation failure history' heading"
)
for idx, failure in enumerate(observed, start=1):
prefix = f"failure #{idx}"
if not _failure_documented_in_text(text, failure) and documented:
reasons.append(
f"{prefix}: command and failing test/error not documented in report"
)
command = (failure.get("command") or "").strip()
failing = (failure.get("failing_test") or failure.get("error") or "").strip()
if not command:
reasons.append(f"{prefix}: missing command in session failure record")
if not failing:
reasons.append(
f"{prefix}: missing failing test or error in session failure record"
)
plain_pass = bool(_PLAIN_PASS_RE.search(text))
transient_wording = bool(_TRANSIENT_STATUS_RE.search(text))
final_passed = final_status in {
"passed",
"pass",
"passed_after_transient_failure_investigation",
}
if (final_passed or plain_pass) and observed:
if cause_unknown and not transient_wording:
violations.append(
"unknown transient failure cause cannot be erased as plain 'passed'"
)
reasons.append(
"when cause is unknown, use status "
"'passed after transient failure investigation'"
)
elif not transient_wording and final_status != "passed_after_transient_failure_investigation":
if not _ENV_CLEANUP_RE.search(text):
reasons.append(
"later passing rerun must document what changed between runs "
"or environmental cleanup performed"
)
if session.get("environmental_contamination") and not _ENV_CLEANUP_RE.search(text):
reasons.append(
"environmental /tmp state contamination must document cleanup or "
"what changed before the passing rerun"
)
proven = not reasons and not violations
return {
"proven": proven,
"block": bool(violations) or not proven,
"reasons": reasons,
"violations": violations,
"observed_failure_count": len(observed),
"documented": documented,
"safe_next_action": (
"add Validation failure history with command, failing test, cause, "
"reproduction, baseline comparison, and PR-caused evidence; use "
"'passed after transient failure investigation' when appropriate"
if not proven
else "proceed"
),
}
+148
View File
@@ -0,0 +1,148 @@
"""PR-head vs diagnostic validation integrity verifier (#316)."""
from __future__ import annotations
import re
from typing import Any
_DIAGNOSTIC_LABEL_RE = re.compile(
r"diagnostic local experiment|not pr-head validation|diagnostic-only validation",
re.IGNORECASE,
)
_OFFICIAL_VALIDATION_RE = re.compile(
r"(?:official validation|pr-head validation|validation integrity)",
re.IGNORECASE,
)
_OFFICIAL_COMMAND_RE = re.compile(
r"(?:official validation command|pr-head validation command|validation command)",
re.IGNORECASE,
)
_OFFICIAL_RESULT_RE = re.compile(
r"(?:official validation result|pr-head validation result|validation result|validation integrity status)",
re.IGNORECASE,
)
_DIRTY_AFTER_RE = re.compile(
r"(?:worktree dirty after (?:official )?validation|dirty (?:state )?after (?:official )?validation|"
r"validation worktree dirty after)",
re.IGNORECASE,
)
_DIAGNOSTIC_EDIT_RE = re.compile(
r"(?:diagnostic edit(?:ed)? path|file edits by reviewer|diagnostic local edit)",
re.IGNORECASE,
)
_DIAGNOSTIC_COMMAND_RE = re.compile(
r"(?:diagnostic (?:validation )?command|diagnostic test run|diagnostic result)",
re.IGNORECASE,
)
_SUGGESTED_FIX_RE = re.compile(
r"(?:diagnostic results? (?:were )?used only for suggested fix|suggested fix only|"
r"not used as official validation)",
re.IGNORECASE,
)
_INTEGRITY_STATUS_RE = re.compile(
r"validation integrity status\s*:\s*(passed|failed|not run|contaminated)",
re.IGNORECASE,
)
_POST_EDIT_AS_OFFICIAL_RE = re.compile(
r"(?:official validation(?:\s+result)?\s*:\s*pass|validation passed after (?:local )?edit|"
r"full suite passed after diagnostic edit)",
re.IGNORECASE,
)
def _performed_edits(action_log: list[dict] | None) -> list[str]:
paths: list[str] = []
for entry in action_log or []:
if entry.get("gated_rejected") or entry.get("performed") is False:
continue
path = entry.get("path")
if path and entry.get("kind", "file_edit") in {"file_edit", "edit", "write"}:
paths.append(str(path))
return paths
def assess_validation_integrity_report(
report_text: str,
*,
validation_session: dict | None = None,
action_log: list[dict] | None = None,
) -> dict[str, Any]:
"""Separate official PR-head validation from diagnostic edited-worktree tests (#316)."""
text = report_text or ""
session = dict(validation_session or {})
reasons: list[str] = []
diagnostic_edits = list(session.get("diagnostic_edits") or [])
if not diagnostic_edits:
diagnostic_edits = _performed_edits(action_log)
diagnostic_ran = bool(
session.get("diagnostic_validation_ran")
or session.get("diagnostic_runs")
)
official_ran = bool(session.get("official_validation_ran", True))
official_result = (session.get("official_result") or "").strip().lower()
diagnostic_result = (session.get("diagnostic_result") or "").strip().lower()
dirty_after = session.get("worktree_dirty_after_official")
contaminated = bool(session.get("contaminated"))
if official_ran:
if not _OFFICIAL_VALIDATION_RE.search(text) and not _OFFICIAL_COMMAND_RE.search(text):
reasons.append("report missing official PR-head validation section")
if not _OFFICIAL_COMMAND_RE.search(text) and "validation:" not in text.lower():
reasons.append("report missing official validation command")
if not _OFFICIAL_RESULT_RE.search(text):
reasons.append("report missing official validation result or integrity status")
if dirty_after is not None and not _DIRTY_AFTER_RE.search(text):
reasons.append(
"report missing worktree dirty state after official validation"
)
elif dirty_after is None and official_ran and not _DIRTY_AFTER_RE.search(text):
reasons.append(
"report missing worktree dirty state after official validation"
)
if diagnostic_edits or diagnostic_ran:
if not _DIAGNOSTIC_LABEL_RE.search(text):
reasons.append(
"post-edit diagnostic runs must be labeled "
"'Diagnostic local experiment — not PR-head validation'"
)
if diagnostic_edits and not _DIAGNOSTIC_EDIT_RE.search(text):
reasons.append("report missing diagnostic edit path")
if diagnostic_ran and not _DIAGNOSTIC_COMMAND_RE.search(text):
reasons.append("report missing diagnostic command/result")
if not _SUGGESTED_FIX_RE.search(text):
reasons.append(
"report must state diagnostic results were used only for suggested fix"
)
if _POST_EDIT_AS_OFFICIAL_RE.search(text):
reasons.append(
"post-edit diagnostic results cannot be reported as official PR-head validation"
)
if diagnostic_result == "pass" and official_result == "fail":
if "request_changes" not in text.lower() and "failed" not in text.lower():
reasons.append(
"request-changes must cite unmodified PR-head failure, not diagnostic pass"
)
if contaminated:
status = _INTEGRITY_STATUS_RE.search(text)
if not status or "contaminated" not in status.group(1).lower():
reasons.append(
"contaminated validation must report integrity status "
"'contaminated — recovery required'"
)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"diagnostic_edits": diagnostic_edits,
"safe_next_action": (
"separate official PR-head validation from diagnostic experiments; "
"report dirty-after-validation state"
if reasons
else "proceed"
),
}
+184
View File
@@ -0,0 +1,184 @@
"""PR validation worktree read-only edit verifier (#315)."""
from __future__ import annotations
import re
from typing import Any
_VALIDATION_WORKTREE_RE = re.compile(
r"branches/(?:review-pr\d+[\w/-]*|review-[\w-]+)",
re.IGNORECASE,
)
_DIAGNOSTIC_WORKTREE_RE = re.compile(
r"branches/(?:diagnostic[\w/-]*|scratch[\w/-]*)",
re.IGNORECASE,
)
_FILE_EDITS_NONE_RE = re.compile(
r"file edits by reviewer\s*:\s*none\b",
re.IGNORECASE,
)
_FILE_EDITS_FIELD_RE = re.compile(
r"file edits by reviewer\s*:\s*(.+)",
re.IGNORECASE,
)
_VALIDATION_WORKTREE_PATH_RE = re.compile(
r"(?:validation worktree path|review worktree path)\s*:\s*(\S+)",
re.IGNORECASE,
)
_DIAGNOSTIC_WORKTREE_PATH_RE = re.compile(
r"(?:diagnostic (?:scratch )?worktree path|diagnostic worktree)\s*:\s*(\S+)",
re.IGNORECASE,
)
_DIAGNOSTIC_LABEL_RE = re.compile(
r"diagnostic local experiment|not pr-head validation|diagnostic-only",
re.IGNORECASE,
)
_OFFICIAL_VALIDATION_RE = re.compile(
r"(?:official (?:pr-head )?validation|pr-head validation result)",
re.IGNORECASE,
)
_POST_EDIT_OFFICIAL_RE = re.compile(
r"(?:official validation(?:\s+result)?\s*:\s*pass|validation passed after (?:local )?edit)",
re.IGNORECASE,
)
_PERFORMED_EDIT_KINDS = frozenset({"file_edit", "edit", "write", "edited", "created", "wrote"})
def _normalize_path(path: str) -> str:
return (path or "").replace("\\", "/").strip()
def _is_validation_worktree(path: str) -> bool:
return bool(_VALIDATION_WORKTREE_RE.search(_normalize_path(path)))
def _is_diagnostic_worktree(path: str) -> bool:
normalized = _normalize_path(path)
return bool(
_DIAGNOSTIC_WORKTREE_RE.search(normalized)
or "diagnostic" in normalized.lower()
)
def _performed_edits(action_log: list[dict] | None) -> list[dict[str, str]]:
edits: list[dict[str, str]] = []
for entry in action_log or []:
if entry.get("gated_rejected") or entry.get("performed") is False:
continue
kind = (entry.get("kind") or entry.get("action") or "file_edit").strip().lower()
if kind not in _PERFORMED_EDIT_KINDS:
continue
path = (entry.get("path") or "").strip()
if not path:
continue
worktree = _normalize_path(str(entry.get("worktree_path") or entry.get("cwd") or ""))
edits.append({"path": path, "worktree_path": worktree})
return edits
def _extract_validation_path(text: str, session: dict) -> str:
path = _normalize_path(str(session.get("validation_worktree_path") or ""))
if path:
return path
match = _VALIDATION_WORKTREE_PATH_RE.search(text or "")
return _normalize_path(match.group(1)) if match else ""
def _extract_diagnostic_path(text: str, session: dict) -> str:
path = _normalize_path(str(session.get("diagnostic_worktree_path") or ""))
if path:
return path
match = _DIAGNOSTIC_WORKTREE_PATH_RE.search(text or "")
return _normalize_path(match.group(1)) if match else ""
def assess_validation_worktree_edit_report(
report_text: str,
*,
validation_session: dict | None = None,
action_log: list[dict] | None = None,
) -> dict[str, Any]:
"""Block edits in PR validation worktrees; require diagnostic separation (#315)."""
text = report_text or ""
session = dict(validation_session or {})
reasons: list[str] = []
validation_path = _extract_validation_path(text, session)
diagnostic_path = _extract_diagnostic_path(text, session)
validation_edits = list(session.get("validation_worktree_edits") or [])
diagnostic_edits = list(session.get("diagnostic_edits") or [])
edits = _performed_edits(action_log)
if not validation_edits:
for entry in edits:
wt = entry.get("worktree_path") or validation_path
if wt and _is_validation_worktree(wt) and not _is_diagnostic_worktree(wt):
validation_edits.append(entry["path"])
if not diagnostic_edits:
for entry in edits:
wt = entry.get("worktree_path") or ""
if wt and _is_diagnostic_worktree(wt):
diagnostic_edits.append(entry["path"])
elif entry["path"] and diagnostic_path and not validation_edits:
if _is_diagnostic_worktree(diagnostic_path):
diagnostic_edits.append(entry["path"])
claimed_none = bool(_FILE_EDITS_NONE_RE.search(text))
any_edits = bool(validation_edits or diagnostic_edits or edits)
if validation_edits:
reasons.append(
"reviewer must not edit files in the PR validation worktree; "
f"observed edits: {', '.join(validation_edits)}"
)
if diagnostic_edits or session.get("diagnostic_experiment"):
if not diagnostic_path and not _DIAGNOSTIC_WORKTREE_PATH_RE.search(text):
reasons.append(
"diagnostic experiments require a separate diagnostic scratch worktree path"
)
if not _DIAGNOSTIC_LABEL_RE.search(text):
reasons.append(
"diagnostic experiments must be labeled "
"'Diagnostic local experiment — not PR-head validation'"
)
if not _FILE_EDITS_FIELD_RE.search(text) or claimed_none:
reasons.append(
"diagnostic edits must be reported under 'File edits by reviewer'"
)
if any_edits and claimed_none:
reasons.append(
"report claims 'File edits by reviewer: none' but reviewer file edits occurred"
)
if session.get("official_validation_after_edit") or _POST_EDIT_OFFICIAL_RE.search(text):
if validation_edits or (validation_path and diagnostic_edits):
reasons.append(
"post-edit test results cannot be reported as official PR-head validation"
)
if validation_edits and _OFFICIAL_VALIDATION_RE.search(text):
if not _DIAGNOSTIC_LABEL_RE.search(text):
reasons.append(
"validation worktree was edited; official PR-head validation is no longer valid"
)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"validation_worktree_path": validation_path or None,
"diagnostic_worktree_path": diagnostic_path or None,
"validation_worktree_edits": validation_edits,
"diagnostic_edits": diagnostic_edits,
"safe_next_action": (
"keep PR validation worktrees read-only; use a separate diagnostic "
"scratch worktree and report all reviewer file edits"
if reasons
else "proceed"
),
}
+210
View File
@@ -0,0 +1,210 @@
"""Fail-closed reviewer worktree and local-git safety proofs (#233).
Reviewer sessions must never stash, reset, or otherwise manipulate unrelated
local changes from another session. When the active worktree has dirty tracked
files outside the PR scope, the workflow must stop or switch to a disposable
scratch worktree (``scripts/worktree-review``).
Git command policy (#243): reviewers use an allowlist, not a blocklist.
Any ``git`` invocation that does not match ``_READONLY_REVIEWER_GIT`` is
forbidden including ``checkout HEAD --``, ``checkout .``, ``switch``,
and uncommon ``stash`` subcommands that older blocklists missed.
"""
from __future__ import annotations
import re
import shlex
# Read-only git operations reviewers may use for validation (#243 allowlist).
_READONLY_REVIEWER_GIT = re.compile(
r"\bgit\b(?:\s+(?:-C\s+\S+\s+)?)?"
r"(?:fetch|status|diff|log|show|rev-parse|branch(?:\s+--show-current)?|"
r"worktree\s+list|worktree\s+add)",
re.IGNORECASE,
)
_GIT_INVOCATION = re.compile(r"\bgit\b", re.IGNORECASE)
def parse_dirty_tracked_files(porcelain: str) -> list[str]:
"""Return tracked paths with local modifications from ``git status --porcelain``.
Untracked entries (``??``) are ignored they do not block reviewer work
when a scratch worktree is used, and authors may have unrelated untracked
files without implying reviewer interference.
"""
paths: list[str] = []
for line in (porcelain or "").splitlines():
if not line or len(line) < 4:
continue
if line.startswith("??"):
continue
path = line[3:].strip()
if " -> " in path:
path = path.split(" -> ", 1)[1].strip()
if path:
paths.append(path)
return paths
def files_outside_pr_scope(
dirty_files: list[str] | None,
pr_scope_files: list[str] | None,
) -> list[str]:
"""Dirty tracked files not explained by the PR diff file set."""
dirty = [p for p in (dirty_files or []) if p]
scope = {p for p in (pr_scope_files or []) if p}
if not dirty:
return []
if not scope:
return list(dirty)
return [path for path in dirty if path not in scope]
def _is_git_command(command: str) -> bool:
return bool(_GIT_INVOCATION.search((command or "").strip()))
def is_readonly_reviewer_git_command(command: str) -> bool:
"""True when the command is an explicitly allowed read-only git operation."""
text = (command or "").strip()
if not text or not _is_git_command(text):
return False
return bool(_READONLY_REVIEWER_GIT.search(text))
def is_forbidden_reviewer_git_command(command: str) -> bool:
"""True when a git command is not on the reviewer readonly allowlist."""
text = (command or "").strip()
if not text or not _is_git_command(text):
return False
return not is_readonly_reviewer_git_command(text)
def assess_reviewer_git_command_log(commands: list[str] | None) -> dict:
"""Fail closed when reviewer shell history includes forbidden git mutations."""
forbidden = [
cmd for cmd in (commands or []) if is_forbidden_reviewer_git_command(cmd)
]
if forbidden:
return {
"proven": False,
"block": True,
"forbidden_commands": forbidden,
"reasons": [
"reviewer workflow executed forbidden local git mutation: "
f"{cmd!r}"
for cmd in forbidden
],
"safe_next_action": (
"stop; report worktree interference; do not stash/reset/checkout "
"unrelated files — use scripts/worktree-review instead"
),
}
return {
"proven": True,
"block": False,
"forbidden_commands": [],
"reasons": [],
"safe_next_action": "proceed",
}
def assess_reviewer_worktree_proof(proof: dict | None) -> dict:
"""Evaluate reviewer worktree safety before checkout/diff/validation/review.
*proof* keys:
- ``worktree_path`` (required)
- ``porcelain_status`` or ``dirty_files``
- ``pr_scope_files`` (paths in the PR diff)
- ``scratch_used`` (bool)
- ``scratch_path`` (when scratch_used)
- ``git_commands`` (shell commands executed this session)
- ``unrelated_mutations_claimed`` (bool) stash/reset/drop reported
"""
proof = dict(proof or {})
reasons: list[str] = []
worktree_path = (proof.get("worktree_path") or "").strip()
if not worktree_path:
reasons.append("reviewer worktree path not reported; fail closed")
if proof.get("dirty_files") is not None:
dirty_files = list(proof.get("dirty_files") or [])
else:
dirty_files = parse_dirty_tracked_files(proof.get("porcelain_status") or "")
pr_scope = list(proof.get("pr_scope_files") or [])
unrelated = files_outside_pr_scope(dirty_files, pr_scope)
scratch_used = bool(proof.get("scratch_used"))
scratch_path = (proof.get("scratch_path") or "").strip()
is_dirty = bool(dirty_files)
unrelated_dirty = bool(unrelated)
if unrelated_dirty and not scratch_used:
reasons.append(
"worktree has dirty tracked files outside PR scope "
f"({', '.join(unrelated)}); stop or use a scratch worktree"
)
if scratch_used and not scratch_path:
reasons.append(
"scratch worktree was used but scratch_path was not reported"
)
if proof.get("unrelated_mutations_claimed"):
reasons.append(
"reviewer reported stash/reset/checkout cleanup of unrelated "
"local changes; this is forbidden"
)
command_assessment = assess_reviewer_git_command_log(
list(proof.get("git_commands") or [])
)
if command_assessment["block"]:
reasons.extend(command_assessment["reasons"])
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"worktree_path": worktree_path or None,
"is_dirty": is_dirty,
"dirty_files": dirty_files,
"unrelated_dirty_files": unrelated,
"scratch_used": scratch_used,
"scratch_path": scratch_path or None,
"unrelated_mutations_avoided": not bool(
proof.get("unrelated_mutations_claimed")
or command_assessment.get("forbidden_commands")
),
"safe_next_action": (
"proceed"
if proven
else command_assessment.get("safe_next_action")
or "stop; use scripts/worktree-review or report dirty worktree"
),
"forbidden_commands": command_assessment.get("forbidden_commands", []),
}
def assess_author_worktree_continuity(proof: dict | None) -> dict:
"""Authors may keep dirty feature worktrees; reviewers may not manipulate them.
This helper only proves the task role is author when dirty unrelated files
exist it does not grant reviewers an exception.
"""
proof = dict(proof or {})
role = (proof.get("task_role") or "").strip().lower()
dirty_files = list(proof.get("dirty_files") or [])
if role == "author" and dirty_files:
return {
"allowed": True,
"reasons": [
"author task may continue with dirty tracked files in its own "
"worktree; reviewer interference rules do not apply"
],
}
if role == "reviewer" and dirty_files:
return assess_reviewer_worktree_proof(proof)
return {"allowed": True, "reasons": []}
+207
View File
@@ -0,0 +1,207 @@
"""Reviewer worktree ownership and safe-reuse proof verifier (#312)."""
from __future__ import annotations
import re
from typing import Any
_SESSION_OWNED_RE = re.compile(
r"branches/(?:review-pr\d+[\w/-]*|review-[\w-]+)",
re.IGNORECASE,
)
_WORKTREE_PATH_RE = re.compile(
r"(?:review worktree path|worktree path|session-owned worktree)\s*:\s*(\S+)",
re.IGNORECASE,
)
_BRANCHES_PATH_RE = re.compile(r"/branches/|\bbranches/", re.IGNORECASE)
_MAIN_CHECKOUT_RE = re.compile(
r"main checkout|not (?:the )?main checkout|outside branches/",
re.IGNORECASE,
)
_SAFE_REUSE_RE = re.compile(r"safe[- ]reuse proof", re.IGNORECASE)
_REUSE_POLICY_RE = re.compile(
r"(?:project policy|policy) (?:allows|allowing) (?:reuse|reset)",
re.IGNORECASE,
)
_CLEAN_TRACKED_RE = re.compile(
r"(?:clean tracked state|tracked state\s*:\s*clean|no uncommitted tracked)",
re.IGNORECASE,
)
_CLEAN_UNTRACKED_RE = re.compile(
r"(?:clean untracked state|untracked state\s*:\s*clean|no untracked files)",
re.IGNORECASE,
)
_NOT_OTHER_SESSION_RE = re.compile(
r"not owned by (?:another|other) (?:active )?(?:task|session)",
re.IGNORECASE,
)
_BRANCH_BEFORE_RESET_RE = re.compile(
r"(?:branch/head before reset|head before reset|branch before reset)\s*:\s*(\S+)",
re.IGNORECASE,
)
_RESET_TARGET_RE = re.compile(
r"(?:reset target sha|reset target)\s*:\s*([0-9a-f]{7,40})",
re.IGNORECASE,
)
_DESTRUCTIVE_CMD_RE = re.compile(
r"\bgit\b(?:\s+(?:-C\s+\S+\s+)?)?"
r"(?:reset\s+--hard|clean(?:\s+-[A-Za-z]+)*|checkout\s+(?!HEAD\s+--)|switch\s+)",
re.IGNORECASE,
)
_WORKSPACE_NONE_RE = re.compile(r"workspace mutations\s*:\s*none", re.IGNORECASE)
_WORKTREE_MUTATION_RE = re.compile(
r"(?:worktree(?:/index)? mutations|destructive reset)\s*:\s*(?!none\b).+",
re.IGNORECASE,
)
_RESET_IN_REPORT_RE = re.compile(r"reset\s+--hard", re.IGNORECASE)
def _command_text(entry) -> str:
if isinstance(entry, dict):
return str(entry.get("command") or "").strip()
return str(entry or "").strip()
def _destructive_commands(command_log: list | None) -> list[str]:
return [
cmd
for cmd in (_command_text(entry) for entry in (command_log or []))
if cmd and _DESTRUCTIVE_CMD_RE.search(cmd)
]
def _extract_worktree_path(text: str, session: dict) -> str:
path = (session.get("worktree_path") or "").strip()
if path:
return path
match = _WORKTREE_PATH_RE.search(text or "")
return match.group(1).strip() if match else ""
def _is_session_owned_path(path: str) -> bool:
normalized = (path or "").replace("\\", "/")
return bool(_SESSION_OWNED_RE.search(normalized))
def _safe_reuse_fields_present(text: str) -> list[str]:
missing: list[str] = []
if not _WORKTREE_PATH_RE.search(text):
missing.append("exact worktree path")
if not _BRANCHES_PATH_RE.search(text):
missing.append("worktree inside branches/")
if not _MAIN_CHECKOUT_RE.search(text):
missing.append("worktree is not the main checkout")
if not _NOT_OTHER_SESSION_RE.search(text):
missing.append("worktree not owned by another active session")
if not _CLEAN_TRACKED_RE.search(text):
missing.append("clean tracked state")
if not _CLEAN_UNTRACKED_RE.search(text):
missing.append("clean untracked state")
if not _BRANCH_BEFORE_RESET_RE.search(text):
missing.append("branch/head before reset")
if not _RESET_TARGET_RE.search(text):
missing.append("reset target SHA")
if not _REUSE_POLICY_RE.search(text):
missing.append("explicit project policy allowing reuse/reset")
return missing
def assess_worktree_ownership_report(
report_text: str,
*,
ownership_session: dict | None = None,
command_log: list | None = None,
) -> dict[str, Any]:
"""Prove reviewer worktree ownership before reset or validation (#312)."""
text = report_text or ""
session = dict(ownership_session or {})
reasons: list[str] = []
worktree_path = _extract_worktree_path(text, session)
destructive = _destructive_commands(command_log or session.get("command_log"))
if not destructive and session.get("destructive_reset"):
destructive = ["git reset --hard (session)"]
session_owned = bool(
session.get("session_owned")
or (worktree_path and _is_session_owned_path(worktree_path))
)
safe_reuse = bool(session.get("safe_reuse") or _SAFE_REUSE_RE.search(text))
main_checkout = bool(session.get("main_checkout"))
dirty_tracked = session.get("dirty_tracked")
dirty_untracked = session.get("dirty_untracked")
other_session_owned = bool(session.get("other_session_owned"))
if worktree_path or destructive or session.get("validation_ran"):
if not worktree_path:
reasons.append("report missing exact review worktree path")
elif main_checkout or "branches/" not in worktree_path.replace("\\", "/"):
reasons.append("review worktree must be inside branches/, not the main checkout")
elif not _BRANCHES_PATH_RE.search(worktree_path.replace("\\", "/")):
reasons.append("review worktree path must be under branches/")
if other_session_owned and not safe_reuse:
reasons.append(
"reused worktree appears owned by another active task/session; "
"safe-reuse proof required"
)
if dirty_tracked:
reasons.append(
"worktree has uncommitted tracked changes before reset or validation"
)
if dirty_untracked and safe_reuse:
reasons.append("safe-reuse proof requires clean untracked state")
if safe_reuse and not session_owned:
reasons.extend(
f"safe-reuse proof missing {field}"
for field in _safe_reuse_fields_present(text)
)
if destructive:
if not session_owned and not safe_reuse:
reasons.append(
"destructive worktree commands forbidden without session-owned "
"worktree or safe-reuse proof"
)
if _WORKSPACE_NONE_RE.search(text) and (
_RESET_IN_REPORT_RE.search(text) or any("reset" in c.lower() for c in destructive)
):
reasons.append(
"final report must not claim 'Workspace mutations: none' when "
"git reset --hard occurred"
)
if not _WORKTREE_MUTATION_RE.search(text) and not _RESET_IN_REPORT_RE.search(text):
reasons.append(
"destructive reset must be reported under Worktree/index mutations "
"or destructive reset operations"
)
if (
worktree_path
and not session_owned
and not safe_reuse
and (destructive or session.get("validation_ran"))
and not _is_session_owned_path(worktree_path)
):
reasons.append(
"reused branch-named worktree requires safe-reuse proof before reset or validation"
)
proven = not reasons
return {
"proven": proven,
"block": not proven,
"reasons": reasons,
"worktree_path": worktree_path or None,
"session_owned": session_owned,
"safe_reuse": safe_reuse,
"destructive_commands": destructive,
"safe_next_action": (
"use a fresh session-owned review worktree under branches/review-pr<N>-* "
"or document full safe-reuse proof before destructive reset"
if reasons
else "proceed"
),
}
+89
View File
@@ -0,0 +1,89 @@
"""Block author mutations from reviewer-bound MCP namespaces (#209).
Every mutation must prove that the active profile role, inferred MCP
namespace, and declared task role align. Reviewer sessions cannot silently
perform author-side work (especially PR creation).
"""
from __future__ import annotations
import gitea_config
import role_session_router
def infer_mcp_namespace(profile_name: str | None) -> str:
"""Map a runtime profile name to its MCP namespace label."""
lower = (profile_name or "").strip().lower()
if "reviewer" in lower and "author" not in lower:
return "gitea-reviewer"
if "author" in lower:
return "gitea-author"
return profile_name or "gitea-default"
def derive_role_kind(allowed, forbidden=()) -> str:
"""Classify the active profile the same way as ``mcp_server._role_kind``."""
def can(op):
return gitea_config.check_operation(op, allowed, forbidden)[0]
review = can("gitea.pr.approve") or can("gitea.pr.merge")
author = can("gitea.pr.create") or can("gitea.branch.push")
if review and author:
return "mixed"
if review:
return "reviewer"
if author:
return "author"
return "limited"
def check_author_mutation_namespace(
mutation_task: str,
profile: dict,
) -> tuple[bool, list[str]]:
"""Return (allowed, reasons). Fail closed on reviewer/author mismatch."""
required_role = role_session_router.required_role_for_task(mutation_task)
if required_role != "author":
return True, []
allowed = profile.get("allowed_operations") or []
forbidden = profile.get("forbidden_operations") or []
active_role = derive_role_kind(allowed, forbidden)
profile_name = profile.get("profile_name") or ""
namespace = infer_mcp_namespace(profile_name)
if mutation_task == "create_pr":
if active_role == "reviewer" or namespace == "gitea-reviewer":
return False, [
"author mutation 'create_pr' blocked in reviewer MCP namespace "
f"({namespace}); launch gitea-author",
]
if active_role == "reviewer" or namespace == "gitea-reviewer":
if mutation_task == "create_issue":
ok, _ = gitea_config.check_operation(
"gitea.issue.create", allowed, forbidden)
if ok:
return True, []
return False, [
f"author mutation '{mutation_task}' blocked: active session is "
f"reviewer-bound ({profile_name} / {namespace})",
]
return True, []
def mutation_audit_context(mutation_task: str, profile: dict, *,
remote=None, repository=None) -> dict:
"""Structured mutation metadata for audit records (#209)."""
allowed = profile.get("allowed_operations") or []
forbidden = profile.get("forbidden_operations") or []
return {
"mcp_namespace": infer_mcp_namespace(profile.get("profile_name")),
"profile_name": profile.get("profile_name"),
"task_role": role_session_router.required_role_for_task(mutation_task),
"operation": mutation_task,
"remote": remote,
"repository": repository,
}
+218 -4
View File
@@ -5,17 +5,60 @@ returns a route result before any downstream mutation tools run.
"""
from __future__ import annotations
import os
ROUTE_ALLOWED = "allowed_current_session"
ROUTE_WRONG_ROLE = "wrong_role_stop"
ROUTE_TO_AUTHOR = "route_to_author_session"
ROUTE_TO_REVIEWER = "route_to_reviewer_session"
ROUTE_AMBIGUOUS = "ambiguous_task_stop"
ROUTE_INFRA_STOP = "infra_stop"
_CONFLICT_HEAD = b"<" * 7 + b" "
_CONFLICT_TAIL = b">" * 7 + b" "
_CONFLICT_SEPARATOR = b"=" * 7
def python_bytes_have_conflict_markers(content: bytes) -> bool:
"""Return True when *content* contains git merge-conflict marker lines."""
for line in content.splitlines():
stripped = line.rstrip(b"\r\n")
if stripped.startswith(_CONFLICT_HEAD):
return True
if stripped.startswith(_CONFLICT_TAIL):
return True
if stripped == _CONFLICT_SEPARATOR:
return True
return False
def skip_python_scan_walk_root(project_root: str, walk_root: str) -> bool:
"""Skip venv/git/cache and sibling worktrees under orchestration checkout.
When *project_root* is itself a worktree inside ``branches/``, still scan
that tree do not treat the ``branches`` path segment as a skip signal.
"""
rel = os.path.relpath(walk_root, project_root)
if rel == ".":
return False
head = rel.split(os.sep, 1)[0]
if head in ("venv", ".git", ".pytest_cache"):
return True
if head == "branches":
nested = os.path.join(project_root, "branches")
if os.path.isdir(nested) and (
walk_root == nested or walk_root.startswith(nested + os.sep)
):
return True
return False
REVIEWER_TASKS = frozenset({
"review_pr",
"merge_pr",
"blind_pr_queue_review",
"pr_queue_cleanup",
"pr-queue-cleanup",
"request_changes_pr",
"approve_pr",
})
@@ -31,6 +74,15 @@ AUTHOR_TASKS = frozenset({
"comment_pr",
"address_pr_change_requests",
"delete_branch",
"work_issue",
"work-issue",
"reconcile_landed_pr",
})
RECONCILER_TASKS = frozenset({
"reconcile_already_landed_pr",
"reconcile_already_landed",
"reconcile-landed-pr",
})
TASK_REQUIRED_ROLE = {
@@ -47,14 +99,30 @@ TASK_REQUIRED_ROLE = {
"review_pr": "reviewer",
"merge_pr": "reviewer",
"blind_pr_queue_review": "reviewer",
"pr_queue_cleanup": "reviewer",
"pr-queue-cleanup": "reviewer",
"request_changes_pr": "reviewer",
"approve_pr": "reviewer",
"work_issue": "author",
"work-issue": "author",
"reconcile_landed_pr": "author",
"reconcile_already_landed_pr": "reconciler",
"reconcile_already_landed": "reconciler",
"reconcile-landed-pr": "reconciler",
# #309: reconciler tasks close already-landed PRs/issues only.
"reconcile_close_landed_pr": "reconciler",
"reconcile_close_landed_issue": "reconciler",
}
WRONG_ROLE_REVIEWER_MSG = (
"Wrong role/session for reviewer task. Launch reviewer MCP namespace."
)
WRONG_ROLE_RECONCILER_MSG = (
"Wrong role/session for reconciler task. Launch a reconciler-capable "
"MCP namespace/profile with exact close capability."
)
_session_last_route: dict | None = None
@@ -93,6 +161,29 @@ def route_task_session(
_record_route(result)
return result
if required_role == "reviewer":
infra = assess_infra_stop()
if infra["infra_stop"]:
detail = "; ".join(infra.get("infra_stop_reasons") or [])
message = (
"infra_stop: Unresolved merge conflict or mid-merge state detected "
f"in MCP runtime source ({detail}). Please resolve all conflicts "
"manually, finish/abort the merge, and retry."
)
result = {
"task_type": task_type,
"required_role": required_role,
"active_role": active_role_kind,
"active_profile": active_profile,
"route_result": ROUTE_INFRA_STOP,
"downstream_allowed": False,
"reasons": [message],
"message": message,
"infra_stop_assessment": infra,
}
_record_route(result)
return result
if allowed_in_current_session:
result = {
"task_type": task_type,
@@ -127,6 +218,26 @@ def route_task_session(
_record_route(result)
return result
if required_role == "reconciler":
result = {
"task_type": task_type,
"required_role": required_role,
"active_role": active_role_kind,
"active_profile": active_profile,
"route_result": ROUTE_WRONG_ROLE,
"downstream_allowed": False,
"reasons": [
WRONG_ROLE_RECONCILER_MSG,
"Reconciler tasks cannot run in author or reviewer "
"sessions without exact close capability.",
],
"message": WRONG_ROLE_RECONCILER_MSG,
"runtime_switching_supported": runtime_switching_supported,
"profile_switch_blocked": not runtime_switching_supported,
}
_record_route(result)
return result
if required_role == "author":
route = ROUTE_TO_AUTHOR
message = (
@@ -175,11 +286,56 @@ def _record_route(result: dict):
_session_last_route = dict(result)
def sync_route_from_capability(capability: dict) -> None:
"""Align sticky route state with operation-scoped capability resolution (#228)."""
capability = capability or {}
task = (capability.get("requested_task") or "").strip()
required_role = capability.get("required_role_kind")
if not task or not required_role:
return
if capability.get("allowed_in_current_session"):
_record_route({
"task_type": task,
"required_role": required_role,
"active_role": required_role,
"active_profile": capability.get("active_profile"),
"route_result": ROUTE_ALLOWED,
"downstream_allowed": True,
"reasons": [],
"message": (
f"Operation-scoped task '{task}' resolved for current session; "
"proceed."
),
})
return
if required_role == "reviewer" and capability.get("stop_required"):
_record_route({
"task_type": task,
"required_role": required_role,
"active_role": capability.get("required_role_kind"),
"active_profile": capability.get("active_profile"),
"route_result": ROUTE_WRONG_ROLE,
"downstream_allowed": False,
"reasons": [WRONG_ROLE_REVIEWER_MSG],
"message": WRONG_ROLE_REVIEWER_MSG,
})
def check_author_mutation_after_reviewer_stop(mutation_task: str) -> tuple[bool, list[str]]:
"""Block author-side fallback after a reviewer wrong_role_stop (#206)."""
"""Block author-side fallback after a reviewer wrong_role_stop (#206).
An explicit operation-scoped author capability resolution for the same
*mutation_task* clears the sticky reviewer denial (#228).
"""
last = _session_last_route
if not last:
return True, []
if (
last.get("route_result") == ROUTE_ALLOWED
and last.get("task_type") == mutation_task
and last.get("required_role") == "author"
):
return True, []
if last.get("route_result") != ROUTE_WRONG_ROLE:
return True, []
if last.get("required_role") != "reviewer":
@@ -188,8 +344,66 @@ def check_author_mutation_after_reviewer_stop(mutation_task: str) -> tuple[bool,
return False, [
WRONG_ROLE_REVIEWER_MSG,
"Author-side mutations are blocked after a reviewer-task "
"wrong_role_stop unless the operator explicitly changes the "
"task and relaunches an author MCP session.",
"wrong_role_stop unless the operator explicitly resolves the "
"author task via gitea_resolve_task_capability.",
f"Attempted fallback mutation: {mutation_task}",
]
return True, []
return True, []
def first_conflict_marker_path(project_root: str | None = None) -> str | None:
"""Return the first .py path containing a git conflict marker, or None."""
root_dir = project_root or os.path.dirname(os.path.abspath(__file__))
for root, dirs, files in os.walk(root_dir):
if skip_python_scan_walk_root(root_dir, root):
continue
for file in files:
if not file.endswith(".py"):
continue
file_path = os.path.join(root, file)
try:
with open(file_path, "rb") as f:
if python_bytes_have_conflict_markers(f.read()):
return file_path
except OSError:
pass
return None
def _default_project_root() -> str:
override = (os.environ.get("GITEA_MCP_PROJECT_ROOT") or "").strip()
if override:
return os.path.realpath(override)
return os.path.dirname(os.path.abspath(__file__))
def assess_infra_stop(project_root: str | None = None) -> dict:
"""Recompute infra_stop from live git and source scan state (#285)."""
root_dir = os.path.realpath(project_root or _default_project_root())
reasons: list[str] = []
mid_merge = False
git_dir = os.path.join(root_dir, ".git")
if os.path.isdir(git_dir):
for marker in ("MERGE_HEAD", "rebase-merge", "rebase-apply"):
marker_path = os.path.join(git_dir, marker)
if os.path.exists(marker_path):
mid_merge = True
reasons.append(f"git state: {marker} present under {git_dir}")
conflict_file = first_conflict_marker_path(root_dir)
if conflict_file:
reasons.append(
f"conflict markers detected in {conflict_file} (project_root={root_dir})"
)
infra_stop = mid_merge or bool(conflict_file)
return {
"infra_stop": infra_stop,
"project_root": root_dir,
"conflict_file": conflict_file,
"mid_merge": mid_merge,
"infra_stop_reasons": reasons,
}
def check_mid_merge(project_root: str | None = None) -> bool:
"""Return True if the repository is mid-merge, mid-rebase, or has conflict markers."""
return assess_infra_stop(project_root)["infra_stop"]
+6
View File
@@ -0,0 +1,6 @@
#!/usr/bin/env bash
set -euo pipefail
script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
repo_root="$(cd "$script_dir/.." && pwd)"
cd "$repo_root"
exec python3 -m webui "$@"
+86
View File
@@ -0,0 +1,86 @@
#!/usr/bin/env bash
set -euo pipefail
# sync-gitea-wiki.sh — mirror the repo-tracked docs/wiki/ pages into the
# Gitea native wiki. docs/wiki/ remains the source of truth; the Gitea Wiki
# is a read convenience mirrored FROM it, never edited directly.
#
# scripts/sync-gitea-wiki.sh dry-run (default): print the plan
# scripts/sync-gitea-wiki.sh --push actually sync, ONLY with the exact
# confirmation below
#
# Safety contract:
# - Dry-run is the default and performs no network or repository-mutating
# operation (only a local read of the origin remote URL).
# - A push requires GITEA_WIKI_SYNC_CONFIRM to equal exactly
# "SYNC WIKI <repo-name>" (repo name derived from the origin remote).
# Anything else refuses before any clone happens.
# - The wiki remote is derived from the local origin remote; nothing is
# hardcoded here and no credential material is read, printed, or stored
# by this script — git's own configured auth is used as-is.
# - Only *.md pages from docs/wiki/ are mirrored; nothing else is touched
# and no page is deleted from the wiki by this script.
here="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
repo_root="$(cd "$here/.." && pwd)"
wiki_src="$repo_root/docs/wiki"
[ -d "$wiki_src" ] || { echo "error: $wiki_src not found" >&2; exit 1; }
if git -C "$repo_root" remote get-url origin >/dev/null 2>&1; then
origin_url="$(git -C "$repo_root" remote get-url origin)"
elif git -C "$repo_root" remote get-url prgs >/dev/null 2>&1; then
origin_url="$(git -C "$repo_root" remote get-url prgs)"
else
echo "error: configure an origin or prgs git remote" >&2
exit 1
fi
repo_name="$(basename "$origin_url" .git)"
wiki_remote="${origin_url%.git}.wiki.git"
expected_confirm="SYNC WIKI $repo_name"
pages=()
while IFS= read -r -d '' f; do
pages+=("$(basename "$f")")
done < <(find "$wiki_src" -maxdepth 1 -name '*.md' -print0 | sort -z)
mode="${1:-}"
if [ "$mode" != "--push" ]; then
echo "dry-run: would sync ${#pages[@]} pages from docs/wiki/ to the '$repo_name' Gitea Wiki:"
for p in "${pages[@]}"; do
echo " $p"
done
echo "dry-run: no git or network operation performed."
echo "To sync for real: GITEA_WIKI_SYNC_CONFIRM=\"$expected_confirm\" $0 --push"
exit 0
fi
if [ "${GITEA_WIKI_SYNC_CONFIRM:-}" != "$expected_confirm" ]; then
echo "refused: --push requires GITEA_WIKI_SYNC_CONFIRM to equal exactly:" >&2
echo " $expected_confirm" >&2
exit 1
fi
workdir="$(mktemp -d)"
trap 'rm -rf "$workdir"' EXIT
echo "cloning wiki repository for '$repo_name'"
if ! git clone --quiet -- "$wiki_remote" "$workdir/wiki" 2>/dev/null; then
echo "error: could not clone the wiki repository. If this repo's wiki has" >&2
echo "never been initialized, create its first page once in the Gitea UI" >&2
echo "(Wiki tab -> New Page), then re-run this script." >&2
exit 1
fi
cp "$wiki_src"/*.md "$workdir/wiki/"
if git -C "$workdir/wiki" status --porcelain | grep -q .; then
git -C "$workdir/wiki" add -A
git -C "$workdir/wiki" commit --quiet -m "docs: sync from repo docs/wiki (source of truth)"
echo "pushing wiki update for '$repo_name'"
git -C "$workdir/wiki" push --quiet
echo "done: ${#pages[@]} pages synced."
else
echo "done: wiki already up to date; nothing pushed."
fi
+18 -2
View File
@@ -38,8 +38,26 @@ fi
branch="$1"
start_ref="${2:-prgs/master}"
script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
repo_root="$(cd "$script_dir/.." && pwd)"
# Enforce issue-linked, traceable branch names (issue → branch → worktree → PR).
if [[ "$allow_unlinked" -eq 0 ]]; then
locked_branch=$(python3 -c "
import sys
sys.path.insert(0, '$repo_root')
import issue_lock_store
print(issue_lock_store.resolve_locked_branch_for_session('$branch'))
")
if [[ -z "$locked_branch" ]]; then
echo "Error: No session issue lock is bound. Call gitea_lock_issue before branch creation (fail closed)." >&2
exit 2
fi
if [[ "$branch" != "$locked_branch" ]]; then
echo "Error: Requested branch '$branch' does not match locked branch '$locked_branch' (fail closed)." >&2
exit 2
fi
if [[ "$branch" =~ ^(fix|feat|docs|chore)/issue-[0-9]+-.+ ]] \
|| [[ "$branch" =~ ^review/pr-[0-9]+-.+ ]]; then
:
@@ -58,8 +76,6 @@ EOF
fi
fi
script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
repo_root="$(cd "$script_dir/.." && pwd)"
worktree_name="${branch//\//-}"
worktree_path="$repo_root/branches/$worktree_name"
+135 -526
View File
@@ -1,576 +1,185 @@
---
name: llm-project-workflow
description: >-
Portable, safe operating workflow for LLMs working on any Git/forge project:
issue-first, isolated branch worktrees, no self-review/self-merge, distinct
author/reviewer profiles, cleanup after merge, and fail-closed behavior.
Use at the start of any implementation, review, or merge task on a repo.
Router skill for safe LLM project work: identify task mode, load the matching
canonical workflow file, enforce mode isolation, and emit the correct final
report schema. Use at the start of any implementation, review, merge,
reconciliation, or issue-filing task.
---
# LLM Project Workflow
# LLM Project Workflow Skill
A reusable workflow any LLM can follow to work on any repository safely. Copy
this `skills/llm-project-workflow/` directory into another project unchanged;
adapt only the forge-specific names in [Adapting to a project](#adapting-to-a-project).
This skill is a **router**. Do not perform project work from this file alone.
The core promise: **an LLM never does unsafe or untracked work.** Every change
is tracked by an issue, isolated in its own worktree, reviewed by a different
identity, and cleaned up only after a real merge.
Before any project mutation, identify the task mode and load the matching
workflow file.
## Workflow modes
| Task mode | Workflow | Final report schema |
|-----------|----------|---------------------|
| PR review / approval / merge | [`workflows/review-merge-pr.md`](workflows/review-merge-pr.md) | [`schemas/review-merge-final-report.md`](schemas/review-merge-final-report.md) |
| Reconcile already-landed open PRs | [`workflows/reconcile-landed-pr.md`](workflows/reconcile-landed-pr.md) | [`schemas/reconcile-landed-final-report.md`](schemas/reconcile-landed-final-report.md) |
| Create or update Gitea issues | [`workflows/create-issue.md`](workflows/create-issue.md) | [`schemas/create-issue-final-report.md`](schemas/create-issue-final-report.md) |
| Work on an assigned issue / author code | [`workflows/work-issue.md`](workflows/work-issue.md) | [`schemas/work-issue-final-report.md`](schemas/work-issue-final-report.md) |
| PR-only queue cleanup (one canonical review per PR) | [`workflows/pr-queue-cleanup.md`](workflows/pr-queue-cleanup.md) | [`schemas/pr-queue-cleanup-final-report.md`](schemas/pr-queue-cleanup-final-report.md) |
## Universal rules
- Prove identity, active profile, runtime context, and **exact** capability before
mutation.
- A nearby capability does not count.
- Do not self-review or self-merge.
- Do not mix modes in one run.
- If the required workflow cannot be loaded, stop and produce a recovery handoff
only.
- Final report must use the schema for the loaded workflow.
- If a task requires a different mode, stop and produce a handoff for the
correct workflow.
## Mode isolation
A run that starts in `review-merge-pr` mode may not create process issues,
implement fixes, or edit source files.
A run that starts in `reconcile-landed-pr` mode may not approve, request
changes, merge, implement fixes, or create normal issues.
A run that starts in `create-issue` mode may not review, approve, request
changes, merge, implement fixes, create branches, commit, push, or create PRs.
A run that starts in `work-issue` mode may not review, approve, request changes,
merge, close PRs, or act as reviewer.
A run that starts in `pr-queue-cleanup` mode may not claim issues, create
branches, edit implementation files, file new issues, or review a second PR
after any terminal review mutation.
If the task requires a different mode, stop and produce a handoff for the
correct workflow.
---
## Definitions
- **Merged**: Gitea PR metadata says `merged=true`.
- **Landed**: Equivalent content is present on remote `master`, but PR metadata may not say merged.
- **Landed**: Equivalent content is present on remote `master`, but PR metadata
may not say merged.
- **Closed-not-merged**: PR state is closed and `merged=false`.
- **Reconciled**: A human/LLM verified whether closed-not-merged content landed, partially landed, or was lost, and repaired issue/label/tracker state.
- **Reconciled**: Verified whether closed-not-merged or already-landed content
is present on the target branch; issue/label/tracker state repaired.
## A. Issue-first rule
## Work Selection Rule for LLMs
**No repository change without a tracking issue.** This includes creating,
editing, deleting, or `chmod`-ing files; docs; scripts; commits; pushes; and PRs.
Before starting any issue or PR work, acquire or verify a work lease. Do not
begin coding, reviewing, fixing, branching, committing, pushing, commenting, or
creating a PR until you prove the target is not already being worked.
1. Before any change, confirm a tracking issue exists.
2. If none exists, create one first (title + problem + scope + acceptance).
3. Claim it (assign yourself or apply the `status:in-progress` label) and comment
that work is starting, including the planned branch name.
4. **If the issue cannot be created or claimed, stop.** Do not touch files.
Required checks:
Reading the repo, running read-only status/`git log`, and creating/claiming the
issue itself are allowed from the orchestration checkout without a prior issue.
1. List open PRs.
2. Search for PRs linked to the target issue.
3. Search local and remote branches for the issue number.
4. Search registered worktrees for the issue branch.
5. Check dirty worktrees.
6. Check active leases or recent handoffs.
7. Check whether the issue was already completed by a merged PR.
Additional issue-first rules:
If another active session owns the lease, stop with "work already claimed" or
produce a handoff.
- Do not implement code without an issue unless explicitly authorized.
- **Design-only work uses a discussion/RFC issue** — create one or comment on
the existing one. Design debates belong on the issue, where other LLMs
comment directly. Discussion-only tasks must **not** create branches or PRs;
their comments should include recommendations, risks, open questions, and a
Controller Handoff (§K; compact format unless high-risk).
- **If the repo/tracker home for the work is unclear, stop and ask for an
owner decision.** Do not create a new repository or a new tracker unless
explicitly approved by the owner.
For Gitea-Tools: `gitea_lock_issue` is the fail-closed lease gate before author
mutations; `status:in-progress` and claim comments are supporting lease signals.
## B. Isolated worktree rule
## Global LLM Worktree Rule
**Never implement or review in the main checkout.** The main checkout is for
orchestration and status only (issue creation, `git status`, creating worktrees).
The main project checkout is a stable control checkout on `master`, `main`, or
`dev`. All LLM task work must happen inside the project's `branches/` directory.
- Each issue gets its own branch worktree under an ignored `branches/` directory.
- Review work uses a **separate** review worktree, never the author's folder.
- Dirty work in one branch folder must not block starting another issue.
- No LLM may edit another issue's worktree unless explicitly assigned to it.
- Branch folders are removed only after the PR is merged/closed **and** cleanup
is explicitly part of the task.
If `cwd` is not inside `branches/`, stop before any file edit, test write,
commit, merge, rebase, or cleanup. The main checkout is orchestration-only.
Every implementation branch **must include its issue number** so it is
traceable end to end: **issue → branch → worktree folder → PR → cleanup.**
## Shell Spawn Hard-Stop Rule
Allowed implementation patterns:
`exit_code: -1` with empty stdout/stderr means the shell failed to spawn — not a
command failure. After two consecutive spawn failures, hard-stop shell use for
the session and emit a recovery report (#258).
- `fix/issue-123-short-description`
- `feat/issue-123-short-description`
- `docs/issue-123-short-description`
- `chore/issue-123-short-description`
## Isolated worktree naming
Review-only branches:
Implementation: `(fix|feat|docs|chore)/issue-<number>-<short-description>`
- `review/pr-456-short-description`
Review: `review/pr-<number>-<short-description>`
Use a filesystem-safe folder under `branches/` by replacing slashes with
hyphens, for example `branches/fix-issue-123-short-description`.
## Subagent Tool-Budget Guardrails
`scripts/worktree-start` **enforces** this: it rejects an implementation branch
that does not match `(fix|feat|docs|chore)/issue-<number>-…` (or a
`review/pr-<number>-…` branch), unless `--allow-unlinked` is passed. Traceability
is maintained by:
General-purpose subagents on **single-step MCP tasks** (for example
`gitea_commit_files`) must not expand into 100+ tool-call retry spirals with
WebFetch/Playwright/manual-encoding fallbacks (issue #259).
- the branch name (contains the issue number),
- a claim comment on the issue, e.g.
`Claimed. Branch: fix/issue-123-short-description. Worktree: branches/fix-issue-123-short-description.`,
- the PR body — `Closes #123` or `Fixes #123` when the PR should close the issue
(do NOT use `Implements #123` or `Refs #123` to close, as Gitea will not auto-close),
- cleanup after merge — remove the remote branch, local branch, and the issue
worktree folder, and drop `status:in-progress`.
Default budgets (stop when exceeded):
For projects using `Gitea-Tools` helpers:
- **Single-step MCP mutation** (`commit_files`, `create_pr`, `lock_issue`):
15 tool calls, 5 minutes wall time.
- **Review / merge queue inspection**: 40 tool calls, 15 minutes.
- **Non-mutating exploration**: 60 tool calls, 20 minutes.
```bash
scripts/worktree-start fix/issue-123-example # → branches/fix-issue-123-example
scripts/worktree-review fix/issue-123-example # → branches/review-fix-issue-123-example (detached)
scripts/worktree-clean --delete-branch fix/issue-123-example
```
Rules:
Manual equivalent:
1. When the main session has `gitea.repo.commit`, call `gitea_commit_files`
directly — do not delegate commit to a subagent (#260).
2. After shell spawn failure (#258), attempt the native MCP tool once before
any fallback; shell unavailability never authorizes WebFetch/Playwright/
manual base64.
3. Never resume a failed subagent into a larger retry loop or spawn a second
subagent for the same deterministic step — stop and report.
4. When `gitea_commit_files` is available, forbid WebFetch, Playwright,
manual encoding, and ad-hoc `_encode_*` / `_emit_*` helpers in the repo.
```bash
git fetch <remote> --prune
git worktree add -b fix/issue-123-example branches/fix-issue-123-example <remote>/master
cd branches/fix-issue-123-example
```
Worktree folder: branch with `/` replaced by `-` under `branches/`.
`venv/` and similar are not copied into new worktrees — run checks with a known
interpreter path, or create a venv inside the branch folder.
Helpers: `scripts/worktree-start`, `scripts/worktree-review`,
`scripts/worktree-clean`.
## C. Identity and profile safety
## Identity and profile safety
- Use canonical execution profiles where available; the profile is the role, not the LLM. A task selects a profile; a profile is not permanently assigned.
- **Author and reviewer identities must be distinct.**
- Never place raw tokens/passwords in an LLM/MCP client config. Reference secrets by keychain id or environment variable name only. Prefer a single canonical config file selected by two env vars, e.g.:
- `GITEA_MCP_CONFIG` — path to the canonical profiles file
- `GITEA_MCP_PROFILE` — the profile to activate
- **Dual-Profile MCP Launcher Pattern (Recommended):** To avoid relaunch bottlenecks and PR-author deadlocks, register multiple instances of the same MCP server in the client's configuration simultaneously (e.g., `gitea-author` and `gitea-reviewer`), each pointing to its respective `GITEA_MCP_PROFILE`.
- Tool calls become namespace-scoped: `mcp__gitea-author__*` and `mcp__gitea-reviewer__*`.
- **Trust Model:** Separate tokens remain separate. Profile gates enforce allowed operations, `whoami` is still checked, and self-review/self-merge prevention remains mandatory. This pattern is for convenience and does not bypass security gates.
- **Deadlock Warning:** Reviewer/merge identities must not be used to create PRs, as this makes the reviewer the PR author in Gitea and blocks independent review. PRs should normally be created by the author/work identity, keeping the reviewer identity available for reviews.
- **Fallback:** If a dual-server launcher is not available in the client, relaunch or restart the client with the correct profile environment variable before claiming work.
- **If the authenticated user equals the PR author, stop** — no self-review, no self-merge.
- Author and reviewer identities must be distinct.
- Never place raw tokens in LLM/MCP config.
- Use `gitea_whoami` and `gitea_resolve_task_capability` before mutating.
## D. Branch naming
```text
fix/issue-123-short-description
feat/issue-123-short-description
docs/issue-123-short-description
review/pr-456-scope-check
```
Worktree folder = branch with `/` replaced by `-`
(`branches/fix-issue-123-short-description`).
## E. Start-work workflow
1. Verify the orchestration checkout (right repo, clean tree).
2. Fetch/prune: `git fetch <remote> --prune`.
3. Confirm local `master` equals remote `master` (`git rev-list --left-right --count <remote>/master...master``0 0`).
4. Create/claim the issue (§A).
5. Create the isolated worktree (§B) from latest remote `master`.
6. Implement the narrow scope only — no unrelated refactors or formatting churn.
7. Add/update focused tests when behavior changes.
8. Run the checks (tests, compile/lint, `git diff --check`, secret scan).
Record the branch name and `HEAD` SHA at validation time — the drift
check in step 9 compares against exactly this state.
9. **Branch proof before commit (#177):** prove and state, immediately
before staging/committing (`author_proofs.verify_branch_for_commit`,
`author_proofs.detect_branch_drift`):
- current branch (`git branch --show-current`) equals the intended
feature branch from the issue claim
- current branch is not `master`, `main`, `develop`, `development`, or
`dev`
- branch and `HEAD` have not changed since validation (step 8) — in a
shared checkout another session may switch branches mid-session;
treat that as expected and **stop before committing** when detected
If any check fails, stop and reconcile; do not commit.
10. Commit with an issue-linked message.
11. **Branch proof before push (#177):** prove that the local branch, the
push target branch, and the intended issue branch all match, and that
none of them is a protected branch
(`author_proofs.verify_push_target`). If a commit accidentally landed
on a protected branch, do **not** push: report the accident and the
exact repair steps (`author_proofs.assess_protected_branch_commit`) —
never silently continue after a repair.
12. Push the branch.
13. Open a PR to `master`. The final report must include the branch proofs
from steps 9 and 11 (`author_proofs.build_commit_push_report`).
14. **If you are the author, stop before review/merge.**
15. **Normal issue work must not directly push to `master`.** PR content should be merged through the forge PR merge mechanism.
16. Direct push to `master` is allowed only as a documented recovery exception. If used, the final report must include:
- why the PR merge path could not be used
- exact commits pushed
- PR metadata state
- issue labels/state repaired
- whether the PR is closed-not-merged
## F. Review workflow
1. Use a separate review worktree (`scripts/worktree-review <branch>`), detached.
2. Verify your authenticated identity.
3. Verify the PR author — **you must not be the author.** Self-review
contamination must be *evidence-backed* (#173): state the authenticated
reviewer identity, the PR author identity, whether this session
authored/touched the PR branch, and the evidence source for any
"same-session author" claim. If the evidence is missing, report the
status as **unknown** — never declare contamination by assumption — and
choose another PR or stop (`review_proofs.assess_self_review_contamination`).
4. Verify the worktree is clean.
5. **Checkout proof (#173):** before reviewing or validating, prove and
state: the selected PR head SHA from Gitea (pinned), the local checkout
SHA (`git rev-parse HEAD`), that `HEAD ==` the pinned PR head SHA, and
that the diff base is the PR base branch. If `HEAD` does not match the
pinned head, **stop before review/merge**
(`review_proofs.verify_pinned_head_checkout`).
6. **Inventory proof (#173 + repo disambiguation hardening):** a blind queue
review must prove listing completeness before claiming "only PRs found".
Use repo-name disambiguation:
- "Gitea-Tools" / "gitea tool" / "MCP Gitea tool" / "gitea MCP tool" /
"gitea-tools repo" resolve **only** to `Scaled-Tech-Consulting/Gitea-Tools`.
- "mcp-control-plane" resolves only to `Scaled-Tech-Consulting/mcp-control-plane`.
- Ambiguous ("open PRs", no explicit repo, "MCP Gitea tooling") → inventory
**both** configured repos.
Report must state exactly which repo(s) were checked. If only one checked:
"Only <repo> was checked. Other configured repos were not checked. This is
not a complete queue inventory." Never let a single-repo zero hide PRs in
the other.
Both configured repos must be reported with state filter, pagination proof,
and open-PR count (`review_proofs.assess_inventory_completeness` and
`resolve_repos_from_user_reference`).
7. **Role-boundary proof (#175):** a reviewer queue task must not silently
become author implementation. If no eligible PR exists, stop with the
queue report. Do not claim issues, create branches, commit, push, or open
PRs unless the operator explicitly retasks the run as author work. Mixed
reviewer+author namespace use must be reported with a justification, and
scratch-only notes are not durable evidence unless posted or committed
intentionally (`review_proofs.assess_role_boundary`).
8. Inspect the full diff; confirm scope matches the linked issue; flag unrelated files.
9. Run the tests. Validation reporting must include the exact command and
exact results: pass/fail, counts of tests passed/skipped/failed, any
ignored paths and why they are safe to ignore, and whether the command
differs from the repository's canonical validation command. Only claim a
validation result after the command has completed and its output has
been read (`review_proofs.assess_validation_report`).
10. **Do not merge if checks fail. Do not merge if the reviewer is the author.**
11. **#179 A-bar proofs** (all fail closed when missing —
`review_proofs.assess_capability_evidence`, `assess_sweep_evidence`,
`assess_live_state_recheck`, `assess_role_boundary`):
- Capability claims must cite exact `gitea_resolve_task_capability`
output (or runtime context); a bare "capability checks passed" is
downgraded.
- The secret/provenance sweep must state the exact command/script/
pattern/named method and the scope scanned.
- Immediately before submitting a review verdict (and again before any
merge), re-read live PR state and prove: still open, live head ==
pinned head, base unchanged, no unresolved blocking review state.
- Reviewer runs stay in the reviewer namespace; any author-namespace
call requires an explicit justification in the report.
12. The final report must distinguish (`review_proofs.build_final_report`):
identity eligible; PR author different from reviewer; session
contamination absent (with evidence); validation performed on the pinned
head; capability evidence; sweep verdict; live-state recheck; role
boundary; merge performed; issue status verified. If any proof is
missing, stop or downgrade the result instead of merging confidently.
## G. Merge / cleanup workflow
Only an eligible (non-author) reviewer merges. Before merging: always verify
the authenticated identity **and** the PR author; cite exact capability
evidence for merge_pr (#179); respect runtime profile gates; run independent
validation (do not trust the author's reported results); perform the **final
live-state recheck** (#179 — PR still open, live head == pinned head, base
unchanged, no unresolved blocking review state) immediately before the merge
mutation; and merge with a **pinned head SHA** and, where supported, the
**expected changed-file set**, so a moved head or widened diff refuses the
merge. After a real merge:
1. Confirm remote `master` actually contains the merge commit or expected squashed changes via post-merge file-presence verification (A PR is not done just because `master` moved or is marked "closed". Verify that expected files added/modified in the PR are actually present on `master` using `git pull`, `git log --oneline -- <file>`, or `git merge-base --is-ancestor`; linked issues are closed; `status:in-progress` is removed).
2. Close/release the issue.
3. Whenever an issue is closed, check for `status:in-progress`: remove it, or report why it could not be removed.
4. Do not delete the remote source branch until: PR `merged=true`, or reconciliation confirms content is safely landed, or the issue owner explicitly abandons the work.
5. Remove the local branch.
6. Remove the branch worktree folder (`scripts/worktree-clean --delete-branch <branch>`). Branches/worktrees are cleaned only after the above is verified.
7. Fetch/prune.
8. Confirm the main checkout is clean and current (`0 0` vs remote).
9. Final merge/reconciliation reports must include: PR metadata (state, merged flag, merge commit/hash), Git content (remote master hash, expected content present or not), and the exact post-merge verification method used & results.
Never run cleanup before the merge is confirmed on remote `master`.
## H. Fail-closed cases
**Stop and report — take no mutating action — if:**
- No issue exists and one cannot be created.
- Worktree state is unclear or unexpected.
- Branch/PR state conflicts with the prompt (e.g. prompt says "merged" but it is not).
- A PR is closed but not merged (closed with `merged=false`). In this case:
- stop normal review/merge
- do not delete branches/worktrees
- do not start dependent work
- run reconciliation
- Local `master` is ahead of remote unexpectedly.
- The authenticated user is the PR author (for review/merge).
- Secrets/tokens appear in the diff.
- Tests fail.
- A cleanup step would delete unmerged work.
When in doubt, stop and surface the discrepancy; do not guess or work around a gate.
## I. Recovery patterns
- **Dirty worktree from another issue:** do not touch it. Start your issue in its
own new worktree; unrelated dirty work must not block you.
- **Local `master` ahead of remote unexpectedly:** do not push `master`. Confirm
the commits are preserved on a feature branch (local + remote) first, then
`git reset --hard <remote>/master` to realign. Never discard commits that are
not safely pushed elsewhere.
- **PR closed but not merged (`merged=false`):** do not merge. Run reconciliation: compare PR content to remote `master` and decide:
- **fully landed:** comment that content is present on `master`, remove `status:in-progress`, keep/close issue as appropriate, clean up only after content equivalence is confirmed.
- **partially landed:** do not clean up, reopen issue if needed, create corrective issue/PR for missing pieces.
- **not landed:** reopen issue if needed, reopen PR or create replacement PR, do not clean up source branch/worktree.
- **Branch deleted before merge:** if the commits still exist locally (a branch or
reflog), re-push them and reopen the PR; otherwise recover via
`git fsck --lost-found`. Preserve first, then proceed.
- **Unauthorized/untracked file created:** do not commit it. Leave pre-existing
untracked artifacts (e.g. editor/agent dirs, reports) alone; stage only the
files your issue names (`git add <files>`, never blind `git add -A`).
- **Preserve commits before a reset:** confirm the target commits are reachable
from a branch that is pushed to the remote, then reset. Verify with
`git branch --contains <sha>` and `git log <remote>/<branch>`.
## J. Prompt snippets
Ready-to-copy templates live in [`templates/`](templates/):
- [`start-issue.md`](templates/start-issue.md) — start a new issue.
- [`review-pr.md`](templates/review-pr.md) — review a PR.
- [`merge-pr.md`](templates/merge-pr.md) — merge a PR (eligible reviewer only).
- [`recover-bad-state.md`](templates/recover-bad-state.md) — recover from bad state.
- [`reconcile-closed-not-merged-pr.md`](templates/reconcile-closed-not-merged-pr.md) — reconcile a closed-not-merged PR.
- [`worktree-cleanup.md`](templates/worktree-cleanup.md) — clean up after merge.
- [`release-tag.md`](templates/release-tag.md) — create a release tag.
## K. Controller Handoff (required, every task)
Every LLM task **must end with a `Controller Handoff`** (exact title) — whether the
task was implementation, review, merge, issue triage, documentation,
discussion-only, or blocked planning. It lets a controller LLM understand the
current state immediately, without rereading the conversation.
The section title must be exactly "Controller Handoff" (or "Controller Handoff Summary" for long form). Reports without it are downgraded (see review_proofs.assess_controller_handoff).
**The compact format is the default.** It is written for controller-LLM
readability, not as a full human status report. PR bodies still carry the
full review detail — the handoff never replaces PR documentation.
Compact format (default, canonical field set per issue #182):
```md
## Controller Handoff
- Task:
- Repo:
- Role:
- Identity:
- Issue/PR:
- Branch/SHA:
- Files changed:
- Validation:
- Mutations:
- Current status:
- Blockers:
- Next:
- Safety:
```
Every task must end with a section titled exactly `Controller Handoff`. Compact
format canonical field set per issue #182; mode-specific schemas in
`schemas/*-final-report.md` define required fields. Use the final report schema
for the loaded workflow mode — not the legacy compact block alone.
`review_proofs.assess_controller_handoff()` validates presence.
Role-specific fields (append to the compact block):
## Prompt templates
- review/merge tasks: `Selected PR:`, `Reviewer eligibility:`,
`Pinned reviewed head:`, `Review decision:`, `Merge result:`,
`Linked issue status:`, `Cleanup status:`
- author tasks: `Selected issue:`, `Claim/comment status:`,
`PR number opened:`, `No review/merge:` (explicit confirmation)
- queue/inventory tasks: `Repositories checked:`, `Open PR counts:`,
`Selected PR or reason none selected:`, `Inventory completeness:`
Ready-to-copy task prompts live in [`templates/`](templates/):
The section title must be exactly `Controller Handoff`.
`review_proofs.assess_controller_handoff()` validates this section; reports
missing it (or missing required fields) are downgraded. The handoff never
replaces the full report — it is the compact continuation summary at the end,
and the full report must still carry exact validation results and mutation
confirmation.
The `Safety:` line is never omitted; it is usually:
```text
no self-review; no self-merge; no tags; no secrets; no prod
```
Rules (both formats):
- Never omit the handoff, and never omit the safety confirmations.
- Never bury blockers in earlier text only — they must appear here.
- If you opened a PR, state clearly that review is needed.
- If you reviewed but could not merge, name the exact gate that blocked it.
- If you only commented on a discussion issue, say no code review is needed
but owner/design feedback may be needed.
- If release state was touched, state exactly which tag/commit changed and why.
- If blocked (permissions, missing repo, missing second reviewer identity,
stale dependency, unclear tracker home): stop and report clearly; **never
bypass classifiers, profile gates, missing permissions, or live-consent
requirements**; give the owner concrete options.
**Use the long format below instead of the compact one only when the task was
high-risk or complex** — i.e. when any of these happened:
- a merge, tag, or release
- failed validation
- permissions/profile gates blocked work
- secrets or production access were involved
- a complicated owner decision
- multiple repos or cross-issue state
- the owner explicitly asks for the full format
Long format (high-risk/complex tasks only):
```md
## Controller Handoff Summary
### Work performed
Briefly state what was done.
### Current state
Include:
- current repo
- current branch or master commit
- issue number(s)
- PR number(s), if any
- whether work is complete, blocked, ready for review, or discussion-only
### Files changed
List files changed, or say `None`.
### Validation
List commands run and results, or say `Not applicable — discussion only`.
### Issues encountered
List errors, confusing state, permission/profile problems, stale branches,
failing tests, missing labels, or blocked decisions.
### Review needed?
Say one of:
- `No review needed — discussion/comment only`
- `Review needed — PR is open`
- `Independent non-author review needed`
- `Owner decision needed`
- `Blocked`
### Next recommended action
State exactly what should happen next.
### Safety confirmations
Confirm:
- no self-review
- no self-merge
- no release/tag changes unless explicitly requested
- no secrets committed
- no production access used unless explicitly authorized
```
### Example blocked handoff
```md
## Example blocked handoff
### Work performed
Audited phase-2 MCP Control Plane planning. Found target repo
`mcp-control-plane` does not exist. Prepared issue pack but did not file it.
### Current state
- Repo: `Scaled-Tech-Consulting/Gitea-Tools`, unmodified
- Target repo: `mcp-control-plane`, missing
- Issues: none open in Gitea-Tools
- PRs: none open
- Status: blocked pending owner decision
### Files changed
None.
### Validation
Tracker/repo audit only. No code validation required.
### Issues encountered
Repo creation was denied by permission/classifier because it would be scope
escalation without live consent.
### Review needed?
Owner decision needed.
### Next recommended action
Owner must choose:
1. create `Scaled-Tech-Consulting/mcp-control-plane`
2. authorize repo creation while present
3. file phase-2 issues in Gitea-Tools instead
### Safety confirmations
- no self-review
- no self-merge
- no release/tag changes
- no secrets committed
- no production access used
```
- [`start-issue.md`](templates/start-issue.md) — author work (loads `work-issue.md`)
- [`review-pr.md`](templates/review-pr.md) — review (loads `review-merge-pr.md`)
- [`pr-queue-cleanup.md`](templates/pr-queue-cleanup.md) — one PR per cleanup run
- [`merge-pr.md`](templates/merge-pr.md) — merge (loads `review-merge-pr.md`)
- [`recover-bad-state.md`](templates/recover-bad-state.md)
- [`reconcile-closed-not-merged-pr.md`](templates/reconcile-closed-not-merged-pr.md)
- [`worktree-cleanup.md`](templates/worktree-cleanup.md)
- [`release-tag.md`](templates/release-tag.md)
## Adapting to a project
Replace these project-specific names when copying the skill elsewhere:
| Placeholder | Meaning | Example here |
|-------------|---------|--------------|
| `<remote>` | Git remote for the forge | `prgs` |
| default branch | Integration branch | `master` |
| profile env vars | Canonical config + profile selectors | `GITEA_MCP_CONFIG`, `GITEA_MCP_PROFILE` |
| `branches/` | Ignored worktree directory | `branches/` |
| helper scripts | Worktree helpers | `scripts/worktree-start` / `-review` / `-clean` |
The rules in §A–§K are project-agnostic and should not change.
| Placeholder | Example here |
|-------------|--------------|
| `<remote>` | `prgs` |
| default branch | `master` |
| profile env vars | `GITEA_MCP_CONFIG`, `GITEA_MCP_PROFILE` |
| `branches/` | `branches/` |
| helpers | `scripts/worktree-start` / `-review` / `-clean` |
## Versioning And Tagging
Releases follow SemVer: **`vMAJOR.MINOR.PATCH`** (use **`v0.x.y`** while
unstable). Choose the bump by the largest change since the last tag:
- **PATCH** — bug fixes, docs, tests, wrappers, non-breaking workflow polish.
- **MINOR** — new tools/helpers/config features; backward-compatible behavior.
- **MAJOR** — breaking config/schema/API behavior or a changed MCP contract.
Tags must:
- be created **only from `master`** (the exact commit on remote `master`),
- be created **only after the full test suite passes**,
- be **annotated** tags (`git tag -a`), never lightweight,
- include release notes / a changelog summary referencing the merged PRs/issues.
**Never tag** feature branches, dirty worktrees, unreviewed or self-authored
work, or commits not present on remote `master`.
Additional tag rules:
- Do **not** create, move, delete, or push tags unless explicitly instructed.
- Tag only **after** the intended PR is merged, and tag only the **verified
final master merge commit** (never the PR branch head unless the merge
commit is exactly that commit).
- Always **report the tag target commit** in the final report / handoff.
Release process (see [`templates/release-tag.md`](templates/release-tag.md)):
1. `git fetch <remote> --prune`.
2. Verify local `master` equals remote `master` (`0 0`) and the tree is clean.
3. Run the full test suite; stop on any failure.
4. Inspect merged issues/PRs since the last tag
(`git log --oneline <last-tag>..<remote>/master`).
5. Choose the version bump.
6. Create the annotated tag on remote `master` with release notes.
7. Push the tag.
8. Create/update release notes if the forge supports it.
Where present, `scripts/release-tag` automates this with all gates built in
(SemVer, fetch/prune, on-master, clean tree, local==remote master, HEAD on
remote master, no duplicate tag, tests, annotated-only). Safe by default: no
push without `--push`; `--dry-run` changes nothing; `--skip-tests` must be
explicit and warns.
Releases follow SemVer from remote `master` only, after full test suite passes.
See [`templates/release-tag.md`](templates/release-tag.md) and
`scripts/release-tag`.
@@ -0,0 +1,46 @@
# Create-issue controller handoff schema
**Task mode:** `create-issue`
End every create-issue run with a section titled exactly `Controller Handoff`.
Use this canonical field set. Do not omit fields — use `none` or
`not verified in this session` where appropriate.
Do not use legacy fields: `Workspace mutations`, `Mutations: None` (when
mutations occurred).
```md
## Controller Handoff
- Task:
- Repo:
- Role:
- Identity:
- Active profile:
- Runtime context:
- Requested issue task:
- Workflow source:
- Capability proof:
- Duplicate search terms:
- Duplicate search pagination proof:
- Duplicates found:
- Issues created:
- Issues commented:
- Issues edited:
- Issues skipped as duplicates:
- Labels/assignees/milestones changed:
- File edits by issue creator:
- Worktree/index mutations:
- Git ref mutations:
- MCP/Gitea mutations:
- Issue mutations:
- Label/assignment/milestone mutations:
- External-state mutations:
- Read-only diagnostics:
- Blockers:
- Current status:
- Safe next action:
- Safety statement:
```
Identity format: `username / profile` (not personal email unless required — #305).
@@ -0,0 +1,31 @@
# PR-queue-cleanup final report schema
One report per cleanup run (one run = one PR). Fields must all be present;
use `none` where nothing occurred. Validated by
`pr_queue_cleanup.assess_pr_queue_cleanup_report` (fail closed).
* Task: pr-queue-cleanup
* Workflow source: workflows/pr-queue-cleanup.md (+ version/commit/hash)
* Repo:
* Role/profile:
* Identity:
* PR inventory pagination proof: (inventory_complete / final page / total_count)
* Queue ordering proof:
* Earlier PRs skipped: (with live per-PR proof)
* Selected PR: (exactly one)
* Pinned head SHA:
* Review decision: (single terminal decision)
* Merge authorized for PR: true/false (explicit per-PR operator authorization)
* Merge gates result:
* Merge result: (none / not attempted / merged SHA / blocker)
* Run stop point: (which §4 chain rule ended the run)
* Next suggested PR: (named, not continued to)
* File edits by reviewer:
* Worktree/index mutations:
* Git ref mutations:
* MCP/Gitea mutations:
* Issue mutations: none (required — forbidden in cleanup mode)
* Branch mutations: none (required — forbidden in cleanup mode)
* Read-only diagnostics:
* Blockers:
* Safe next action: (fresh run for the next PR)
@@ -0,0 +1,53 @@
# Reconcile-landed controller handoff schema
**Task mode:** `reconcile-landed-pr`
End every reconciliation run with a section titled exactly `Controller Handoff`.
Use this canonical field set. Do not omit fields — use `none` or
`not verified in this session` where appropriate.
Reject stale author/reviewer fields: `PR number opened`, `Pinned reviewed head`,
`Scratch worktree used`, `Workspace mutations`, `Mutations: None` (when mutations
occurred).
```md
## Controller Handoff
- Task:
- Repo:
- Role:
- Identity:
- Active profile:
- Runtime context:
- Selected PR:
- PR live state:
- Candidate head SHA:
- Target branch:
- Target branch SHA:
- Ancestor proof:
- Linked issue:
- Linked issue live status:
- Eligibility class:
- Capabilities proven:
- Missing capabilities:
- PR comments posted:
- Issue comments posted:
- PRs closed:
- Issues closed:
- File edits by reconciler:
- Worktree/index mutations:
- Git ref mutations:
- MCP/Gitea mutations:
- Reconciliation mutations:
- External-state mutations:
- Read-only diagnostics:
- Blockers:
- Current status:
- Safe next action:
- Safety statement:
- No review/merge confirmation:
```
Identity format: `username / profile` (not personal email unless required — #305).
`git fetch` belongs under `Git ref mutations`, not read-only diagnostics (#297).
@@ -0,0 +1,113 @@
# Review-merge controller handoff schema
**Task mode:** `review-merge-pr`
End every review/merge run with a section titled exactly `Controller Handoff`.
Use this canonical field set. Do not omit fields — use `none` or
`not verified in this session` where appropriate.
Do not use legacy fields: `Pinned reviewed head`, `Scratch worktree used`,
`Workspace mutations`, `Mutations: None` (when mutations occurred).
```md
## Controller Handoff
- Task:
- Repo:
- Role:
- Identity:
- Active profile:
- Runtime context:
- Selected PR:
- Linked issue:
- Eligibility class:
- Queue ordering policy:
- Inventory pagination proof:
- Earlier PRs skipped:
- Candidate head SHA:
- Reviewed head SHA:
- Target branch:
- Target branch SHA:
- Already-landed gate:
- Author-safety result:
- Prior request-changes state:
- Review worktree used:
- Review worktree path:
- Review worktree inside branches:
- Review worktree HEAD state:
- Review worktree dirty before validation:
- Review worktree dirty after validation:
- Baseline worktree used:
- Baseline worktree path:
- Files reviewed:
- Validation:
- Official validation integrity status:
- Terminal review mutation:
- Review decision:
- Merge preflight:
- Merge result:
- Linked issue status:
- Main checkout branch:
- Main checkout dirty state:
- Main checkout updated:
- File edits by reviewer:
- Worktree/index mutations:
- Git ref mutations:
- MCP/Gitea mutations:
- Review mutations:
- Merge mutations:
- Cleanup mutations:
- External-state mutations:
- Read-only diagnostics:
- Blockers:
- Current status:
- Safe next action:
- Safety statement:
```
### Already-landed handoff overrides
When eligibility class is `ALREADY_LANDED_RECONCILE_REQUIRED`:
- Reviewed head SHA: `none`
- Review worktree used: `false`
- Review worktree path: `none`
- Review decision: `none`
- Merge result: `none`
Identity format: `username / profile` (not personal email unless required — #305).
### Queue-status-only runs (no selected PR)
When the run inventories the queue but selects no PR for review:
- Selected PR: `none`
- Already-landed gate, Author-safety result, Merge preflight: `not applicable` or `not run` — never `passed`
- Review worktree detail fields: `not applicable` or `none` when Review worktree used is `false`
- Blockers must not be `none` if the narrative says all open PRs are conflicted, blocked, or unverified
- Inventory pagination proof must cite final-page metadata, not default page-size assumptions
Verifier: `review_proofs.assess_queue_status_report()`.
Narrative final report and controller handoff must agree on eligibility class,
candidate/reviewed head SHA, mutation state, worktree usage, review decision,
terminal review mutation, merge result, and linked issue status.
### Proof-backed claims (#395)
Proof-sensitive claims must cite explicit command/tool evidence in the report
or structured MCP metadata — not narrative alone:
- **Inventory complete:** `has_more=false`, `is_final_page=true`,
`inventory_complete=true`, and/or `total_count` from `gitea_list_prs`.
- **Skipped earlier PRs:** merge-simulation command output, conflict file list,
or live `gitea_get_pr_review_feedback` / mergeability fields.
- **Baseline validation:** baseline worktree path, baseline target SHA,
dirty-before/after, exact command, exact result.
- **Master integration:** exact `git merge` / `git rebase` command and exit status.
- **Cleanup:** final `git worktree list` (or remove commands) proving session
worktrees were removed.
When a claim relies on prior-session blocker state or MCP metadata only, label
the proof source explicitly (`command`, `MCP metadata`, `prior blocker`,
`not checked`). Do not use `live proof` without that classification.
@@ -0,0 +1,73 @@
# Work-issue controller handoff schema
**Task mode:** `work-issue`
End every work-issue run with a section titled exactly `Controller Handoff`.
Use this canonical field set. Do not omit fields — use `none` or
`not verified in this session` where appropriate.
Do not use legacy fields: `Workspace mutations`, `Mutations: None` (when
mutations occurred).
```md
## Controller Handoff
- Task:
- Repo:
- Role:
- Identity:
- Active profile:
- Runtime context:
- Selected issue:
- Eligibility class:
- Issue ordering policy:
- Issue inventory pagination proof:
- Earlier issues skipped:
- Duplicate active work proof:
- Claim/lock state:
- Stable branch:
- Stable branch SHA:
- Branch name:
- Worktree path:
- Worktree inside branches:
- Worktree branch/HEAD state:
- Worktree dirty before implementation:
- Files changed:
- Validation:
- Baseline comparison:
- Commit SHA:
- Push result:
- PR number:
- PR URL:
- PR verification:
- Main checkout branch:
- Main checkout dirty state:
- Main checkout used for task work:
- File edits by author:
- Worktree/index mutations:
- Git ref mutations:
- MCP/Gitea mutations:
- Issue mutations:
- Branch mutations:
- Commit mutations:
- Push mutations:
- PR mutations:
- Cleanup mutations:
- External-state mutations:
- Read-only diagnostics:
- Blockers:
- Current status:
- Safe next action:
- Safety statement:
```
Identity format: `username / profile` (not personal email unless required — #305).
Narrative final report and controller handoff must agree on eligibility class,
selected issue, and mutation ledger categories (#319, #320).
`git fetch` and ref-updating commands belong under `Git ref mutations`, not
`Read-only diagnostics` (#297).
Forbidden claims without proof (#330): `next eligible issue`, `issue claimed`,
`validation passed`, `PR created`, `worktree clean`, `all gates passed`, etc.
@@ -5,6 +5,10 @@ Copy, fill the `<...>` fields, and paste as the task prompt.
```text
Task: merge PR #<pr> for issue #<n> if it is eligible and checks pass.
Load the canonical workflow first:
`skills/llm-project-workflow/workflows/review-merge-pr.md` (task mode: review-merge-pr).
Final report schema: `schemas/review-merge-final-report.md`.
Rules (llm-project-workflow):
- Only an eligible, NON-author reviewer merges. If authenticated user == PR
author → STOP.
@@ -0,0 +1,25 @@
# Template: PR-only queue cleanup (one PR per run)
Copy, fill the `<...>` fields, and paste as the task prompt.
```text
Task: PR-only queue cleanup for <repo>.
Load workflows/pr-queue-cleanup.md and workflows/review-merge-pr.md before any
mutation. Route pr_queue_cleanup through gitea_route_task_session (reviewer
profile required).
Rules:
- One run = exactly one selected PR = one terminal review decision.
- Build full open-PR inventory with pagination proof before selection.
- Forbidden: issue claiming, branch creation, implementation edits, issue filing,
reviewing a second PR after a terminal mutation in this run.
- After REQUEST_CHANGES: stop. After APPROVED: merge only this PR and only if
operator explicitly authorized merge for this PR in this run.
- Report Next suggested PR without continuing to it.
Operator PR list (optional): <pr numbers or "oldest eligible from inventory">
Merge authorized for selected PR in this run: <true|false>
End with the pr-queue-cleanup final report schema.
```
@@ -17,9 +17,36 @@ Repo name disambiguation (Gitea-Tools blind review hardening):
configured repos were not checked. This is not a complete queue inventory."
- A single-repo "no open PRs" result MUST NOT be reported as global "no open PRs"
if the other configured repo was not inventoried.
- PR inventory trust gate (#196): before reporting "no open PRs" or "queue empty",
the workflow must run `pr_inventory_trust_gate` (via the live inventory path or
`review_proofs.assess_reviewer_queue_inventory`). Only `trusted_empty` allows a
clean empty-queue stop. Report `pr_inventory_trust_gate.status`, reasons, and
corroboration in the final report. A bare `[]` from `gitea_list_prs` is never
sufficient proof.
- Empty-queue report wall (#198): if the final report claims "no open PRs",
"queue empty", or "nothing to review", it must include verbatim:
`pr_inventory_trust_gate.status`, trust-gate reasons, corroboration,
remote/owner/repo/state filter, and the inventory MCP profile. A recent merge
commit is not valid corroboration. Author-bound sessions must not present
reviewer queue inventory as a reviewer decision.
Load the canonical workflow first:
`skills/llm-project-workflow/workflows/review-merge-pr.md` (task mode: review-merge-pr).
Final report schema: `schemas/review-merge-final-report.md`.
Rules (llm-project-workflow):
- Review in a SEPARATE detached review worktree, never the author's folder.
- Worktree safety (#233): before checkout, diff, validation, review, or merge,
report the starting worktree path and whether it was dirty. If unrelated
tracked files exist outside the PR scope, STOP or run
`scripts/worktree-review <pr-head-branch>` and validate in the scratch path.
Scratch-clone validation is the norm; tests must not assume the shared
development worktree or a repo-local ``venv/`` (#245).
NEVER run `git stash`, `git stash pop/drop`, `git checkout --`, `git reset`,
or `git clean` to manage another session's dirty files.
- Final report must state: Worktree path, Worktree dirty (yes/no),
Scratch worktree used (yes/no + path if yes), and confirm no unrelated local
files were modified, stashed, reset, or dropped.
- You must NOT be the PR author. If the authenticated user == PR author, stop.
A different LLM-Agent-SHA does NOT make you a different actor — only a
different authenticated Gitea user does (docs/llm-agent-sha.md).
@@ -5,6 +5,10 @@ Copy, fill the `<...>` fields, and paste as the task prompt.
```text
Task: implement <issue title / one-line goal>.
Load canonical workflow: skills/llm-project-workflow/workflows/work-issue.md
Final report schema: skills/llm-project-workflow/schemas/work-issue-final-report.md
Router: skills/llm-project-workflow/SKILL.md (task mode: work-issue)
Rules (llm-project-workflow):
- No repo changes without a tracking issue. If none exists, create one first;
if it can't be created, stop.
@@ -13,6 +17,21 @@ Rules (llm-project-workflow):
- Do not self-review or self-merge.
Steps:
0. Work Selection Rule — before any claim, branch, or file edits, acquire or
verify a work lease. Required checks: list open PRs; search PRs linked to
the target issue; search local/remote branches for the issue number; search
registered worktrees for the issue branch; check dirty worktrees; check
active leases or recent handoffs; check whether a merged PR already
completed the issue. If another session owns the lease, stop (continue only
as lease owner, review the existing PR, hand off, request takeover after
expiry, or report "work already claimed"). Never open a parallel branch/PR
unless the old branch is proven abandoned and takeover is recorded.
0b. Global LLM Worktree Rule — before any mutation, prove and state: project
root; cwd; current branch; stable branch for the main checkout (master/main/dev);
session-owned worktree path under branches/. If cwd is not inside branches/,
STOP (no exceptions — not for docs, tests, small fixes, review fixes, conflicts,
or cleanup). Main checkout is control-only: read-only inspect, fetch, create
worktrees, stable-branch update after merge, explicit repair.
1. Identity Checklist: Before claiming work, verify and state:
- Required identity/profile for this task: author (allowed to push branches / create PRs)
- Current authenticated identity (from whoami): <username>
@@ -0,0 +1,666 @@
---
task_mode: create-issue
canonical: true
final_report_schema: ../schemas/create-issue-final-report.md
---
# Create issue workflow (canonical)
**Task mode:** `create-issue`
This file is the canonical issue-creation workflow for Gitea-Tools. Load it
before any issue mutation. Final report schema:
[`schemas/create-issue-final-report.md`](../schemas/create-issue-final-report.md).
**Default task prompt:**
> Create or update Gitea issues in this project only if every identity,
> capability, duplicate-search, issue-scope, final-report, mutation-ledger,
> and proof-wording gate passes.
Do not improvise around the gates. Follow project skills, MCP gates, and
workflow rules exactly.
This is an issue-creation workflow. It is not a reviewer workflow and not an
implementation workflow.
---
## 0. Load the canonical workflow first
Before starting issue creation or issue update work, check whether the project provides a canonical create-issue workflow through a project skill, runbook, or MCP helper.
If available, load it first and report:
* workflow source
* workflow version, commit, or hash
* whether this prompt conflicts with the loaded workflow
If the canonical workflow cannot be loaded and the project requires it, stop and produce a recovery handoff only.
## 1. Mode isolation
This run is `create-issue` mode only.
Do not:
* review PRs
* approve PRs
* request changes
* merge PRs
* close PRs
* close issues unless the user explicitly asks and exact close capability is proven
* implement code
* edit repo files
* create branches
* create commits
* push branches
* create PRs
* run tests unless the canonical workflow explicitly requires validation for issue creation
* perform reviewer-only actions
* perform author/coder-only actions
* perform MCP repair
If the task requires review, merge, issue implementation, or MCP repair mode, stop and produce a handoff for the correct workflow.
Do not mix modes in one run.
## 2. Start with live identity, profile, runtime, and capability checks
Prove:
* authenticated identity
* active profile
* repo/project
* runtime context
* exact capability for reading/searching issues
* exact capability for creating issues, if creating issues
* exact capability for commenting on issues, if commenting on existing issues
* exact capability for editing issues, if editing existing issues
* exact capability for applying labels, if applying labels
* exact capability for assigning issues, if assigning issues
* exact capability for closing issues, only if explicitly requested
A nearby capability does not count.
Examples:
* `create_issue` does not authorize `issue_comment`
* `issue_comment` does not authorize `create_issue`
* `create_pr` does not authorize `create_issue`
* `review_pr` does not authorize `create_issue`
* `merge_pr` does not authorize `issue_comment`
* `gitea.read` does not authorize creating, commenting, editing, labeling, assigning, or closing issues
If exact capability cannot be proven, stop and produce a recovery handoff only.
## 3. Stop immediately on blocked infrastructure
If any of the following appears, stop immediately:
* `infra_stop`
* MCP reconnect failure
* stale capability state
* missing capability
* workspace mismatch
* dirty control checkout, if the canonical workflow treats that as blocking
* broken canonical workflow loading
* failed required preflight
* capability resolver warning that says the current state may be unsafe
* stale or inconsistent runtime context
Do not continue duplicate search, issue creation, issue commenting, issue editing, labeling, assignment, or cleanup.
Produce an executable recovery handoff only.
Blocked recovery handoffs must not include direct issue-create or issue-comment replay commands.
Blocked handoffs must say to rerun the full workflow after the blocker clears.
## 4. Main checkout rule
This workflow should not mutate repo files.
Do not edit files in the main checkout.
Do not create branches.
Do not create commits.
Do not push.
Do not run implementation work.
Do not run reviewer validation.
Reading repository files is allowed only when needed to understand the requested issue and only if the canonical workflow permits it.
If the main checkout is dirty and the project treats dirty control checkout as blocking, stop and produce a recovery handoff.
## 5. No raw MCP repair during issue creation
Do not run `pkill`, kill MCP processes, edit MCP config, restart servers, or perform control-checkout repair during issue creation.
If MCP repair is required, stop issue creation and produce a separate `CONTROL-CHECKOUT REPAIR MODE` handoff.
Do not mix MCP repair mode with create-issue mode.
After repair, rerun the full workflow from the beginning.
## 6. No background task tools
Do not use `schedule`, `manage_task`, background jobs, async waits, delayed task tools, or monitoring tasks during issue creation.
Use direct commands and MCP tools only.
If a required action cannot complete synchronously, stop and produce a recovery handoff.
Do not say “I will check later,” “I will monitor,” or “I will continue in the background.”
## 7. No local Gitea fallback during normal issue creation
During normal issue-creation workflows, do not read Gitea profile secret files.
Do not inspect or open files such as:
* `profiles.json`
* local token stores
* credential files
* local Gitea auth/profile config files
* `.env` files containing Gitea credentials
* keychain dumps
* token helper outputs
Do not run local Gitea helper scripts when MCP tools are available.
Use MCP tools for Gitea operations.
Local fallback is allowed only in explicit recovery mode when MCP is unavailable and identity/profile/capability can be independently proven.
If local fallback is used, report:
* why MCP was unavailable
* exact identity proof
* exact profile proof
* exact repo proof
* exact capability proof
* exact local command used
Do not use local fallback to bypass MCP gates.
## 8. Understand the requested issue work
Before searching or creating issues, restate the requested issue-creation task in operational terms.
Identify:
* target repo/project
* issue topic
* issue type, if known
* whether this is a new issue, duplicate check, issue update, or issue-comment task
* whether the user provided exact title/body text
* whether acceptance criteria were provided
* whether multiple issues are requested
* whether labels, assignees, milestones, or links are requested
* whether any requested action requires capability beyond issue creation
Do not invent missing requirements.
If the request is ambiguous but safe to proceed, make a reasonable best-effort issue with clear assumptions.
If ambiguity would cause unsafe or wrong mutation, stop and ask for clarification or produce a recovery handoff according to project policy.
## 9. Duplicate search before mutation
Before creating any issue, search open and closed issues for duplicates.
Duplicate search must happen before each issue creation unless a batch search clearly covers all proposed issues.
Search terms must include:
* exact proposed issue title
* key noun phrase from the problem
* key workflow/tool name
* key failure phrase or error phrase
* likely alternate wording
* linked PR number, issue number, or file name, if relevant
If the user supplied search terms, use those too.
For each proposed issue, report:
* search terms used
* matching issue numbers/titles
* whether each match is open or closed
* whether any match fully covers the requested issue
* whether any match partially covers the requested issue
* whether a new issue is still needed
Do not create a duplicate issue if an existing issue fully covers the problem.
If a duplicate exists and fully covers the problem, stop issue creation for that topic and report the duplicate.
If a duplicate exists but is missing important acceptance criteria, comment on the existing issue only if exact `issue_comment` capability is proven and the user/task authorizes commenting.
If commenting is not authorized, report the existing issue and the missing criteria in the final handoff.
## 10. Issue inventory pagination rule
If listing/searching issues returns paginated results, follow pagination until the tool proves there are no more pages.
Do not assume search results or issue inventory are complete.
Pagination proof must not rely on assumed default API page size.
Search/inventory is complete only if one of the following is proven:
* the MCP response explicitly says there is no next page / `has_more=false` / final page
* the workflow traversed pages until an empty page or explicit final page was returned
* the tool response includes total-count or pagination metadata proving all relevant issues were returned
* the request explicitly set `page` / `limit` / `per_page`, and the response explicitly proves the server honored that page size and did not truncate results
Do not say “duplicate search complete” merely because the result count is less than an assumed default page size.
If pagination metadata is absent and the tool cannot page, report `ISSUE_SEARCH_PAGINATION_UNPROVEN`.
If duplicate search cannot be trusted, do not create the issue unless the canonical workflow explicitly permits best-effort issue creation with that limitation disclosed.
## 11. Issue creation scope rule
Create only issues within the requested scope.
Do not create extra issues just because related problems are noticed.
Do not create process-hardening issues unless the user explicitly requested process-hardening or the current task is explicitly about workflow/tooling gaps.
Do not create implementation issues during reviewer mode.
Do not create reviewer issues during work-on-issue mode.
Do not create issues in a different repository unless the user explicitly asked and exact capability is proven.
If multiple issues are requested, create only the requested issues and only after duplicate search for each one.
If a proposed issue is too broad, split it only if the user requested splitting or the canonical workflow requires issue granularity.
## 12. Issue content quality rule
Every created issue must be actionable.
Include, when applicable:
* title
* problem statement
* observed evidence
* expected behavior
* required behavior
* acceptance criteria
* affected workflow/tool/files
* safety or security considerations
* duplicate search summary
* related issues or PRs
* non-goals, if useful
Acceptance criteria must be concrete and testable.
Avoid vague issues like:
* “make workflow better”
* “fix LLM behavior”
* “improve process”
* “handle this better”
Instead, describe the exact wall, gate, verifier, test, schema, helper, or prompt change required.
## 13. Issue title rule
Use concise, specific titles.
Good title patterns:
* `Enforce <specific gate>`
* `Add verifier for <specific report/proof problem>`
* `Split <large workflow> into <specific components>`
* `Block <unsafe action> during <workflow mode>`
* `Require <proof type> before <claim/action>`
Avoid titles that are too broad or emotional.
The title should be unique enough that duplicate search can find it later.
## 14. Issue body rule
Issue body must include the full acceptance criteria.
Do not create placeholder issues.
Do not create issues with only a title unless the user explicitly requested title-only creation.
If the user supplied exact issue body text, preserve it unless it contains unsafe instructions, stale facts, or contradictions.
If edits are needed, make the smallest correction necessary and report the correction.
Do not silently change requested meaning.
## 15. Labels, assignees, and metadata
Apply labels, assignees, milestones, or project fields only if:
* the user requested them, or
* the canonical workflow requires them, and
* exact capability is proven.
Do not guess labels if project label policy is unknown.
If labels are useful but capability or policy is unclear, mention recommended labels in the final report instead of applying them.
Do not assign issues to people unless explicitly requested or required by project workflow.
## 16. Comment-on-existing issue rule
Comment on an existing issue only if:
* an existing issue partially covers the requested work, or
* the user asked to add information to an existing issue, or
* the canonical workflow requires duplicate consolidation comments, and
* exact `issue_comment` capability is proven.
Comment must be specific and useful.
Include:
* why the existing issue is relevant
* what acceptance criteria or evidence should be added
* whether this avoids creating a duplicate
Do not comment just to say “duplicate found” unless the project workflow requires it.
Do not close duplicate issues unless explicitly requested and exact close capability is proven.
## 17. No hidden mutations
Do not perform unreported mutations.
Every issue creation, issue comment, issue edit, label change, assignment, milestone change, close/reopen action, or external-state change must be reported.
If a tool call is dry-run-only, confirmation-gated, rejected, or no-op, report it separately from performed mutations.
A dry run is not a mutation.
A rejected call is not a performed mutation.
A successful issue creation is an issue mutation.
A successful issue comment is an issue mutation.
A successful label/assignment/milestone update is an issue mutation or external-state mutation.
## 18. Issue creation gate
Before creating each issue, verify:
* identity is still valid
* active profile is still valid
* runtime context is still safe
* exact `create_issue` capability is still valid
* duplicate search was completed or limitation was explicitly allowed
* proposed title is not a duplicate
* proposed body includes actionable acceptance criteria
* target repo is correct
* no mode switch has occurred
If any gate fails, do not create the issue.
Produce a recovery handoff or duplicate report.
## 19. Issue commenting gate
Before commenting on an existing issue, verify:
* identity is still valid
* active profile is still valid
* runtime context is still safe
* exact `issue_comment` capability is still valid
* target issue number is correct
* comment body is specific and useful
* comment will not duplicate an existing comment
* no mode switch has occurred
If any gate fails, do not comment.
Produce a recovery handoff or report the intended comment as a recommendation only.
## 20. Issue edit/update gate
Before editing an existing issue, verify:
* identity is still valid
* active profile is still valid
* runtime context is still safe
* exact edit capability is still valid
* target issue number is correct
* update is explicitly requested or required by canonical workflow
* update does not erase useful existing content
* no mode switch has occurred
If any gate fails, do not edit.
Prefer commenting over editing unless the user explicitly requested an edit or the canonical workflow requires issue body updates.
## 21. Final report must be precise
Include:
* canonical workflow source/version/hash, if available
* authenticated identity/profile
* repo/project
* runtime context summary
* exact capability proof summary
* requested issue-creation task
* duplicate search terms used
* duplicate search result
* pagination/final-page proof for issue search, if applicable
* issues created, with issue numbers and URLs
* existing issues commented, with issue numbers and URLs
* existing issues edited, with issue numbers and URLs
* issues skipped as duplicates, with issue numbers and titles
* labels/assignees/milestones applied, if any
* blockers, if stopped
* confirmation that no PR review, approval, request-changes, merge, branch, checkout, commit, push, or repo-file mutation was performed
If the report and actual tool/command log disagree, fix the report before final output.
## 22. Final report must distinguish mutation types
Do not use the legacy field `Workspace mutations`.
Use only precise categories:
* File edits by issue creator:
* Worktree/index mutations:
* Git ref mutations:
* MCP/Gitea mutations:
* Issue mutations:
* Label/assignment/milestone mutations:
* External-state mutations:
* Read-only diagnostics:
Use precise wording:
* `File edits by issue creator: none`
* `Worktree/index mutations: none`
* `Git ref mutations: none`
* `Issue mutations: ...`
If no repo files were edited, say:
`File edits by issue creator: none`
If no branches/worktrees were touched, say:
`Worktree/index mutations: none`
If no git refs were updated, say:
`Git ref mutations: none`
Do not hide issue mutations inside vague `MCP/Gitea mutations`.
## 23. Local artifact and report consistency rule
Do not create local walkthrough, notes, markdown, JSON, or report artifacts during issue-creation runs unless the canonical workflow or operator explicitly requires it.
If any file is edited, created, generated, or written, report it under `File edits by issue creator`.
For each file write, report:
* exact path
* whether it was inside the repo
* whether it was tracked or untracked
* why it was created
* whether final status was checked after the write
Do not say `File edits by issue creator: none` if any file write occurred.
Do not write files after the final clean-status check unless you rerun and report a new final clean-status check.
Default behavior: do not create local artifacts during issue creation.
## 24. Forbidden final-report claims unless proven
Do not claim:
* `duplicate search complete`
* `no duplicate found`
* `issue created`
* `issue commented`
* `issue updated`
* `label applied`
* `capability proven`
* `runtime safe`
* `all gates passed`
* `no file edits`
* `no unsafe mutation`
* `no PR mutation`
* `no repo mutation`
* `pagination complete`
* `final page`
* `no next page`
unless the corresponding proof is included.
If anything blocks safe issue creation or issue update, stop immediately and produce an executable recovery handoff.
Do not improvise around the gates.
## 25. Proof wording enforcement
The following phrases are forbidden unless directly supported by current-session evidence:
* duplicate search complete
* no duplicate found
* issue created
* issue commented
* issue updated
* labels applied
* capability proven
* runtime safe
* all gates passed
* no file edits
* no unsafe mutation
* no PR mutation
* no repo mutation
* pagination complete
* final page
* no next page
If the proof comes from prior state rather than a command/tool run in the current session, label it as prior proof, not live proof.
If a tool call was rejected, confirmation-gated, dry-run-only, or no-op, report it separately from performed mutations.
## 26. Final self-check before output
Before final output, check the report for contradictions.
Verify:
* if any file was edited, `File edits by issue creator` is not `none`
* if any worktree was added/removed, `Worktree/index mutations` lists it
* if any fetch happened, `Git ref mutations` lists it
* if any issue was created, `Issue mutations` lists it
* if any issue was commented, `Issue mutations` lists it
* if any issue was edited, `Issue mutations` lists it
* if any labels/assignees/milestones were changed, the correct mutation category lists it
* if duplicate search is claimed complete, pagination/final-page proof is present or limitation is disclosed
* if no duplicate is claimed, search terms and results are present
* if issue created is claimed, issue number and URL are present
* if no PR mutation is claimed, no PR tool/action was used
* if no repo mutation is claimed, no branch/worktree/file/commit/push action occurred
* if all gates passed is claimed, every required gate has proof
If any contradiction exists, fix the final report before output.
## 27. Controller handoff schema
End every run with a controller handoff using this schema.
Do not omit fields. Use `none` or `not verified in this session` where appropriate.
Controller Handoff:
* Task:
* Repo:
* Role:
* Identity:
* Active profile:
* Runtime context:
* Requested issue task:
* Workflow source:
* Capability proof:
* Duplicate search terms:
* Duplicate search pagination proof:
* Duplicates found:
* Issues created:
* Issues commented:
* Issues edited:
* Issues skipped as duplicates:
* Labels/assignees/milestones changed:
* File edits by issue creator:
* Worktree/index mutations:
* Git ref mutations:
* MCP/Gitea mutations:
* Issue mutations:
* Label/assignment/milestone mutations:
* External-state mutations:
* Read-only diagnostics:
* Blockers:
* Current status:
* Safe next action:
* Safety statement:
## 28. Stop conditions summary
Stop immediately and produce a recovery handoff if:
* canonical workflow is required but cannot be loaded
* identity/profile/capability cannot be proven
* runtime context is blocked
* infra stop appears
* MCP reconnect fails
* capability state is stale
* duplicate search cannot be performed and best-effort creation is not allowed
* issue search pagination cannot be proven and best-effort creation is not allowed
* duplicate fully covers the requested issue
* requested issue body is unsafe or not actionable
* target repo cannot be proven
* create_issue capability is missing
* issue_comment capability is missing for a required comment
* issue edit capability is missing for a required edit
* mode switch would be required
* any report contradiction cannot be resolved
Blocked handoffs must not include direct issue-create or issue-comment replay commands.
Blocked handoffs must say to rerun the full workflow after the blocker clears.
Do not improvise around the gates.
@@ -0,0 +1,86 @@
---
task_mode: pr-queue-cleanup
canonical: true
final_report_schema: ../schemas/pr-queue-cleanup-final-report.md
---
# PR-only queue cleanup workflow (canonical)
**Task mode:** `pr-queue-cleanup`
Reviewer-role mode for cleanup periods when the queue holds many open PRs.
Each run dispatches **exactly one** canonical review for **exactly one** PR,
then stops after any terminal review mutation. The next PR always requires a
new run with fresh identity, capability, and inventory proof.
This mode composes with the canonical review workflow: for the selected PR,
load and follow [`workflows/review-merge-pr.md`](review-merge-pr.md) in full.
This file adds the cleanup-mode boundaries around that per-PR run; it does
not replace the review workflow.
## 0. Load the canonical workflow first
Load this file and `workflows/review-merge-pr.md` before any mutation and
report source, version/commit/hash, and any conflict with the operator
prompt. If either cannot be loaded, stop and produce a recovery handoff.
## 1. Mode isolation — forbidden actions
PR-only cleanup mode forbids, with no exceptions:
* issue claiming (`claim_issue`, `mark_issue`, `lock_issue`)
* branch creation or push (`create_branch`, `push_branch`)
* implementation edits of any kind
* new issue filing (`create_issue`)
* PR creation (`create_pr`) and commit tools (`commit_files`)
* reviewing a second PR after any terminal review mutation
`pr_queue_cleanup.check_cleanup_task_allowed` fails closed on these tasks.
If author-side work is needed, stop and hand off to `work-issue` mode in a
separate session.
## 2. Identity, capability, and routing
Route `pr_queue_cleanup` through `gitea_route_task_session` — reviewer role
required; author sessions receive `wrong_role_stop`. Prove `gitea.pr.review`
capability before selection. Merge additionally requires `gitea.pr.merge`
plus the explicit per-PR authorization in §4.
## 3. Inventory and selection
Build the complete open-PR inventory with pagination proof
(`inventory_complete`, final page, `total_count`) before any selection
claim. Select exactly one PR according to project queue ordering rules
(oldest-first unless the operator queue says otherwise), skipping earlier
PRs only with live per-PR proof. No multi-PR validation and no batch report
may substitute for per-PR proof.
## 4. Terminal mutation chain
`pr_queue_cleanup.resolve_cleanup_run_state` is the authority:
* After `REQUEST_CHANGES` → the run stops.
* After `COMMENT` or a proof-backed skip → the run stops.
* After `APPROVED` → the run may continue **only** to same-PR merge, and only
when the operator explicitly authorized merge for that specific PR in this
run (`Merge authorized: true`) and every merge gate passes.
* After merge, or on any merge blocker → the run stops.
* Any terminal mutation targeting a PR other than the selected PR is a hard
stop and must be reported as a violation.
## 5. Fresh run per PR
The next PR requires a new run with fresh identity, capability, inventory,
and ordering proof. Do not carry pinned SHAs, eligibility classes, or
validation results across runs.
## 6. Final report
Use the schema in
[`schemas/pr-queue-cleanup-final-report.md`](../schemas/pr-queue-cleanup-final-report.md).
The report must include the **Next suggested PR** (from the proven ordering)
without continuing to it, exactly one **Selected PR**, the single terminal
decision, pagination proof, and `Issue mutations: none` / `Branch mutations:
none`. `pr_queue_cleanup.assess_pr_queue_cleanup_report` validates these
fields and fails closed on batch reviews, missing pagination proof, missing
next-suggested-PR, unauthorized merges, or any issue/branch mutation.
@@ -0,0 +1,409 @@
---
task_mode: reconcile-landed-pr
canonical: true
final_report_schema: ../schemas/reconcile-landed-final-report.md
---
# Reconcile already-landed open PR workflow (canonical)
**Task mode:** `reconcile-landed-pr`
This file is the canonical reconciliation workflow for open PRs whose head SHA
is already an ancestor of the target branch. Load it before any reconciliation
mutation. Final report schema:
[`schemas/reconcile-landed-final-report.md`](../schemas/reconcile-landed-final-report.md).
**Default task prompt:**
> Reconcile already-landed open PRs in this project. Do not review or merge
> normal PRs. Close or comment only when exact capability is proven.
Do not improvise around the gates. Follow project skills, MCP gates, and
workflow rules exactly.
This is a reconciliation workflow. It is not a normal PR review/merge workflow.
---
## 0. Load the canonical workflow first
Before starting reconciliation work, check whether the project provides a
canonical reconcile-landed-PR workflow through a project skill, runbook, or MCP
helper.
If available, load it first and report:
* workflow source
* workflow version, commit, or hash
* whether this prompt conflicts with the loaded workflow
If the canonical workflow cannot be loaded and the project requires it, stop and
produce a recovery handoff only.
## 1. Mode isolation
This run is `reconcile-landed-pr` mode only.
**Do not review or merge normal PRs.**
Do not:
* approve PRs
* request changes on PRs
* merge PRs
* implement code
* edit repo files
* create branches
* create commits
* push branches
* create PRs
* run normal PR validation as review approval input
* perform author/coder implementation work
* perform raw MCP repair
If the task requires review, merge, issue implementation, or MCP repair mode,
stop and produce a handoff for the correct workflow.
Do not mix modes in one run.
## 2. Start with live identity, profile, runtime, and capability checks
Prove:
* authenticated identity
* active profile (reconciler or author with close capabilities, as required)
* repo/project
* runtime context
* exact capability for reading/listing PRs and issues
* exact capability for PR inspect (`gitea.read` / view PR)
* exact capability for issue inspect
* exact capability for PR comment, if commenting
* exact capability for issue comment, if commenting
* exact capability for PR close, if closing PRs
* exact capability for issue close, if closing issues
A nearby capability does not count.
Examples:
* `review_pr` does not authorize PR close
* `merge_pr` does not authorize PR close or issue close
* `create_issue` does not authorize `issue_comment`
* `issue_comment` does not authorize PR close
* `gitea.read` does not authorize close or comment mutations
If exact capability cannot be proven, stop and produce a recovery handoff only.
## 3. Stop immediately on blocked infrastructure
If any of the following appears, stop immediately:
* `infra_stop`
* MCP reconnect failure
* stale capability state
* missing capability
* workspace mismatch
* broken canonical workflow loading
* failed required preflight
* capability resolver warning that says the current state may be unsafe
* stale or inconsistent runtime context
Do not continue inventory, ancestry proof, commenting, closing, or cleanup.
Produce an executable recovery handoff only.
Blocked recovery handoffs must not include direct close or comment replay
commands.
Blocked handoffs must say to rerun the full workflow after the blocker clears.
## 4. Main checkout rule
This workflow should not mutate repo files.
Do not edit files in the main checkout.
Do not create branches, commits, or pushes.
Do not run implementation or reviewer validation worktrees for code edits.
Reading repository files is allowed only when needed to understand
reconciliation scope and only if this workflow permits it.
## 5. No raw MCP repair during reconciliation
Do not run `pkill`, kill MCP processes, edit MCP config, restart servers, or
perform control-checkout repair during reconciliation.
If MCP repair is required, stop and produce a separate `CONTROL-CHECKOUT REPAIR
MODE` handoff.
After repair, rerun the full workflow from the beginning.
## 6. No background task tools
Do not use `schedule`, `manage_task`, background jobs, async waits, delayed task
tools, or monitoring tasks during reconciliation.
Use direct commands and MCP tools only.
If a required action cannot complete synchronously, stop and produce a recovery
handoff.
## 7. No local Gitea fallback during normal reconciliation
During normal reconciliation workflows, do not read Gitea profile secret files.
Do not inspect `profiles.json`, local token stores, credential files, `.env`
Gitea credentials, keychain dumps, or token helper outputs.
Do not run local Gitea helper scripts when MCP tools are available.
Use MCP tools for Gitea operations.
Local fallback is allowed only in explicit recovery mode when MCP is unavailable
and identity/profile/capability can be independently proven.
## 8. Build a complete live open PR inventory
List open PRs for the target repo according to project policy.
Follow pagination until the tool proves there are no more pages.
Do not assume inventory is complete.
Pagination proof must not rely on assumed default API page size.
Inventory is complete only if one of the following is proven:
* the MCP response explicitly says there is no next page / `has_more=false` /
final page
* the workflow traversed pages until an empty page or explicit final page was
returned
* the tool response includes total-count or pagination metadata proving all
relevant PRs were returned
* the request explicitly set `page` / `limit` / `per_page`, and the response
explicitly proves the server honored that page size and did not truncate results
If pagination cannot be proven, report `INVENTORY_PAGINATION_UNPROVEN` and stop
unless project policy allows best-effort reconciliation with that limitation
disclosed.
## 9. Already-landed proof
For each candidate PR, prove whether the PR head SHA is already landed on the
target branch.
**Already-landed proof** must include:
* PR number and title
* candidate head SHA (full 40-hex)
* target branch name
* target branch SHA (full 40-hex) after fetch
* ancestor proof method (`git merge-base --is-ancestor`, equivalent forge API, or
documented project helper)
* ancestor proof result (true/false)
* live PR state (open/closed, merged flag)
Do not classify a PR as already-landed without live ancestor proof.
If ancestry cannot be proven, classify as `ANCESTRY_UNPROVEN` and skip close
mutations.
## 10. Linked issue live verification
If the PR claims to close or link an issue, fetch the linked issue live before
reporting its status.
If the linked issue was not fetched live in the current session, report:
`Linked issue status: not verified in this session`
Do not claim `issue open`, `issue closed`, or `issue resolved` without live
proof.
## 11. Reconciliation selection rules
Select PRs eligible for reconciliation:
* open PR state
* `merged=false` unless project policy says otherwise
* head SHA is ancestor of target branch (already-landed proof passed)
* not selected for normal review/merge in this run
Eligibility class for selected PRs: `ALREADY_LANDED_RECONCILE_REQUIRED`
Do not select PRs that fail already-landed proof for normal review/merge
treatment in this mode.
## 12. Reconciliation comment policy
Post a reconciliation comment only if:
* exact PR-comment or issue-comment capability is proven
* the comment adds durable evidence (ancestor proof summary, recommended close
action, linked issue status)
* the comment will not duplicate an equivalent recent reconciliation comment
If comment capability is missing, record the intended comment in the final
handoff only.
## 12A. Partial reconciliation policy (#302)
The durable policy when `close_pr` capability is missing is
**comment-then-stop** (`resolve_partial_reconciliation_plan` in
`review_proofs.py` enforces it):
* `close_pr` proven → full reconciliation: close the PR and report the
close result (`FULL_RECONCILE_CLOSE_ALLOWED`).
* `close_pr` missing, `comment_pr` proven → post exactly one
reconciliation comment carrying PR head SHA, target branch SHA,
ancestor proof, linked issue status, and the required missing
capability, then stop for a human or authorized close
(`PARTIAL_RECONCILE_COMMENT_THEN_STOP`).
* `comment_pr` also missing → no Gitea mutation; produce a recovery
handoff recording the intended comment and required capabilities
(`RECOVERY_HANDOFF_ONLY`).
* Ancestry not proven → no mutation regardless of capability
(`GATE_NOT_PROVEN`).
Final reports must explain why the comment was or was not posted and
name the missing capability (`assess_partial_reconciliation_report`).
## 13. PR close rules
Close a PR only if:
* already-landed proof passed in this session
* exact PR-close capability is proven
* PR is still open at mutation time (live re-fetch)
* head SHA still matches the proved candidate head SHA
If PR-close capability is missing, produce a recovery handoff with exact PR,
proof, and required capability. Do not loop forever re-blocking the reviewer
queue.
## 14. Issue close rules
Close a linked issue only if:
* exact issue-close capability is proven
* linked issue was fetched live
* issue resolution is justified by landed content and project policy
* issue is still open at mutation time
If issue-close capability is missing, report the gap in the handoff.
## 15. Missing capability behavior
If any required mutation capability is missing:
* do not improvise with review/merge tools
* do not ask the operator to bypass capability gates
* produce a recovery handoff listing exact missing capabilities
* include safe next action (profile switch, human close, or dedicated reconciler
profile)
## 16. Mutation classification
Use precise mutation categories in the final report:
* File edits by reconciler: (expect `none`)
* Worktree/index mutations:
* Git ref mutations: (`git fetch` belongs here, not read-only diagnostics)
* MCP/Gitea mutations:
* Reconciliation mutations: (PR comment, issue comment, PR close, issue close)
* External-state mutations:
* Read-only diagnostics:
Do not use legacy `Workspace mutations`.
## 17. Identity privacy rule
Report identity as `username / profile` (#305).
Do not disclose personal email in final reports unless explicitly required.
## 18. Precise final report
Include:
* canonical workflow source/version/hash
* authenticated identity/profile
* repo/project
* capability proof summary (separate lines for inspect, comment, PR close,
issue close)
* inventory pagination proof
* selected PR(s) with already-landed proof
* linked issue live status
* mutations performed or blocked
* missing capabilities
* confirmation that no normal review, approval, request-changes, or merge was
performed
## 19. Local artifact and report consistency rule
Do not create local walkthrough, notes, markdown, JSON, or report artifacts
during reconciliation unless explicitly required.
If any file is edited, report under `File edits by reconciler`.
Default: no repo file edits.
## 20. Forbidden unsupported claims unless proven
Do not claim:
* `already-landed`
* `PR closed`
* `issue closed`
* `pagination complete`
* `inventory complete`
* `all gates passed`
* `no unsafe mutation`
unless the corresponding proof is included.
## 21. Proof wording enforcement
Forbidden unless supported by current-session evidence:
* pagination complete
* final page
* no next page
* PR closed
* issue closed
* all gates passed
If proof comes from prior state, label as prior proof, not live proof.
## 22. Final self-check before output
Verify:
* no normal review/merge mutations occurred
* `git fetch` is under Git ref mutations if it occurred
* already-landed proof is present for each selected PR
* handoff uses reconciliation schema, not author/reviewer merge schema
* no contradiction between narrative report and controller handoff
## 23. Controller handoff schema
End every run with `Controller Handoff` per
[`schemas/reconcile-landed-final-report.md`](../schemas/reconcile-landed-final-report.md).
## 24. Stop conditions summary
Stop immediately and produce a recovery handoff if:
* canonical workflow cannot be loaded
* identity/profile/capability cannot be proven
* runtime context is blocked
* `infra_stop` appears
* inventory pagination cannot be proven and best-effort is not allowed
* already-landed proof cannot be completed
* required close capability is missing and mutation was attempted
* live PR/issue state contradicts proof
* any report contradiction cannot be resolved
Do not improvise around the gates.
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,969 @@
---
task_mode: work-issue
canonical: true
final_report_schema: ../schemas/work-issue-final-report.md
---
# Work issue workflow (canonical)
**Task mode:** `work-issue`
This file is the canonical author/coder workflow for Gitea-Tools. Load it
before any issue implementation mutation. Final report schema:
[`schemas/work-issue-final-report.md`](../schemas/work-issue-final-report.md).
**Default task prompt:**
> Find the next eligible issue in this project, work on it only if all gates
> pass, and create a PR when complete.
Do not improvise around the gates. Follow project skills, MCP gates, and
workflow rules exactly.
This is an author/coder workflow. It is not a reviewer workflow.
---
## 0. Load the canonical workflow first
Before starting issue work, check whether the project provides a canonical work-on-issue workflow through a project skill, runbook, or MCP helper.
If available, load it first and report:
* workflow source
* workflow version, commit, or hash
* whether this prompt conflicts with the loaded workflow
If the canonical workflow cannot be loaded and the project requires it, stop and produce a recovery handoff only.
## 1. Mode isolation
This run is `work-issue` mode only.
Do not:
* review PRs
* approve PRs
* request changes
* merge PRs
* close PRs unless the PR creation workflow explicitly does so through Gitea automation
* close unrelated issues
* mutate reviewer state
* perform reviewer-only actions
* create process-hardening issues unless explicitly authorized and the workflow switches to issue-creation mode
If the task requires review, merge, issue creation, or MCP repair mode, stop and produce a handoff for the correct workflow.
Do not mix modes in one run.
## 2. Start with live identity, profile, runtime, and capability checks
Prove:
* authenticated identity
* active author/coder profile
* repo/project
* runtime context
* exact capability for reading issues
* exact capability for claiming/locking issues, if available
* exact capability for branch creation, if handled through MCP
* exact capability for pushing branches, if applicable
* exact capability for creating PRs
* exact capability for commenting on issues or PRs, if needed
A nearby capability does not count.
Examples:
* `create_issue` does not authorize `issue_comment`
* `review_pr` does not authorize `merge_pr`
* `create_pr` does not authorize `merge_pr`
* `issue_comment` does not authorize `create_issue`
* `gitea.read` does not authorize issue claim, PR creation, or branch mutation
If capability cannot be proven, stop and produce a recovery handoff only.
### 2A. Pre-task role routing (#139)
Before issue selection or any mutation, resolve the composite author task:
* `gitea_route_task_session(task_type="work-issue", remote=…)` — must return
`route_result: allowed_current_session` and `downstream_allowed: true`
* `gitea_resolve_task_capability(task="work_issue", remote=…)` — must show
`required_role_kind: author` and `allowed_in_current_session: true`
Hyphen (`work-issue`) and underscore (`work_issue`) aliases are equivalent.
If routing returns `ambiguous_task_stop`, `wrong_role_stop`, or
`route_to_author_session`, stop and produce a recovery handoff only — do not
fall back to reviewer tools or guess a different profile.
## 3. Stop immediately on blocked infrastructure
If any of the following appears, stop immediately:
* `infra_stop`
* MCP reconnect failure
* stale capability state
* dirty control checkout
* dirty task worktree
* missing capability
* workspace mismatch
* stale target branch state
* broken canonical workflow loading
* failed required preflight
* capability resolver warning that says the current state may be unsafe
* stale or inconsistent runtime context
Do not continue issue selection, claiming, implementation, validation, commit, push, PR creation, cleanup, or handoff mutation.
Produce an executable recovery handoff only.
Blocked recovery handoffs must not include direct commit, push, or PR replay commands.
Blocked handoffs must say to rerun the full workflow after the blocker clears.
## 4. Main checkout rule
The main project checkout must stay on `master`, `main`, or `dev`.
Do not do task work in the main checkout.
Do not edit files in the main checkout.
Do not run tests in the main checkout.
Do not commit from the main checkout.
Do not create PRs from the main checkout.
All task work must happen under the projects `branches/` directory.
No exceptions for small fixes, docs, tests, cleanup, conflict resolution, emergencies, or “just one file.”
If the main checkout is dirty before selection, stop and produce a recovery handoff.
If the main checkout becomes dirty during the run, stop and produce a recovery handoff unless the change is explicitly allowed by the canonical workflow.
### Stacked PRs (explicit exception, #484)
Normal author work stays base-equivalent to `master`/`main`/`dev`. A **stacked PR** — deliberately based on another unmerged PR's branch — is the only sanctioned non-master base, and only when the operator/controller explicitly chooses it:
- Branch the `branches/` worktree from the dependency's branch, then lock with `gitea_lock_issue(..., stacked_base_branch=<dep-branch>, stacked_base_pr=<open-PR#>)`. The lock fails closed unless that open PR owns the branch; arbitrary or stale branches are rejected.
- Open the PR with `gitea_create_pr(base=<dep-branch>)`. The body must state: `Stacked on PR #<X> / issue #<Y>`, `Base branch: <dep-branch>`, `Head branch: <this-branch>`, `Do not merge before PR #<X>`, and note retarget/rebase to `master` after the dependency lands if required.
- This does not relax the main-checkout rule or bypass the issue lock — work still happens under `branches/`, and the approved base is recorded on the lock.
## 5. No raw MCP repair during normal issue work
Do not run `pkill`, kill MCP processes, edit MCP config, restart servers, or perform control-checkout repair during normal issue work.
If MCP repair is required, stop issue work and produce a separate `CONTROL-CHECKOUT REPAIR MODE` handoff.
Do not mix MCP repair mode with work-on-issue mode.
Do not use successful repair as permission to resume the same issue workflow. After repair, rerun the full workflow from the beginning.
## 6. No background task tools
Do not use `schedule`, `manage_task`, background jobs, async waits, delayed task tools, or monitoring tasks during issue work.
Use direct commands and MCP tools only.
If a required action cannot complete synchronously, stop and produce a recovery handoff.
Long synchronous commands, such as a test suite, are allowed only if they are run directly and reported with exact command, working directory, and result.
Do not say “I will check later,” “I will monitor,” or “I will continue in the background.”
## 7. No local Gitea fallback during normal issue work
During normal author/coder workflows, do not read Gitea profile secret files.
Do not inspect or open files such as:
* `profiles.json`
* local token stores
* credential files
* local Gitea auth/profile config files
* `.env` files containing Gitea credentials
* keychain dumps
* token helper outputs
Do not run local Gitea helper scripts when MCP tools are available.
Use MCP tools for Gitea operations.
Local fallback is allowed only in explicit recovery mode when MCP is unavailable and identity/profile/capability can be independently proven.
If local fallback is used, report:
* why MCP was unavailable
* exact identity proof
* exact profile proof
* exact repo proof
* exact capability proof
* exact local command used
Do not use local fallback to bypass MCP gates.
## 8. Build a complete live issue inventory
List open issues according to the projects issue selection policy.
Follow pagination until the tool proves there are no more pages.
Do not assume inventory is complete.
Do not claim `next eligible issue`, `oldest eligible issue`, or complete issue inventory unless pagination is proven.
Pagination proof must not rely on assumed default API page size.
Inventory is complete only if one of the following is proven:
* the MCP response explicitly says there is no next page / `has_more=false` / final page
* the workflow traversed pages until an empty page or explicit final page was returned
* the tool response includes total-count or pagination metadata proving all relevant issues were returned
* the request explicitly set `page` / `limit` / `per_page`, and the response explicitly proves the server honored that page size and did not truncate results
Do not say “inventory complete” merely because the result count is less than an assumed default page size.
For each candidate issue, identify:
* issue number
* title
* labels
* status
* author/requester, if relevant
* assignee/owner, if any
* linked PRs, if any
* dependency/blocker labels, if any
* whether it appears already claimed
* whether it appears already implemented or superseded
* whether it is eligible under project rules
Final report must include pagination/final-page proof.
## 9. Issue selection rules
State the issue ordering policy before selecting an issue.
If the project uses oldest-first, explicitly sort or reason by issue number or created date.
Do not rely on API response order unless the tool proves that order matches the project policy.
Do not pick:
* already-claimed issues
* issues assigned to another active worker
* issues with an open PR already covering the work
* duplicate issues
* blocked issues
* dependency-blocked issues
* already implemented issues
* issues outside the current requested scope
* process-hardening issues unless this run was explicitly started for process-hardening work
* reviewer-only issues if this is author/coder mode
For every earlier issue skipped, report:
* issue number
* current status
* blocking category
* proof used
* whether there is an open PR
* whether there is an active claim
* reason it is not eligible
If eligibility cannot be proven, classify it as:
`ISSUE_ELIGIBILITY_UNVERIFIED`
Then stop or produce a recovery handoff according to project policy.
Do not select an issue based only on memory from a previous session.
## 10. Linked PR / duplicate active work proof
Before claiming or working on an issue, check whether there is already an open PR, branch, or active claim for that issue.
If an open PR already exists for the issue, do not implement duplicate work.
Classify the issue as:
`OPEN_PR_EXISTS`
and skip it only if project policy allows skipping.
If branch naming or PR title convention links issues to branches, search for matching branches or PRs.
Report:
* issue number
* linked/open PRs found
* matching branches found, if checked
* active claims found
* duplicate work status
Do not create another branch/PR for the same issue unless the project explicitly allows taking over or updating existing work and exact capability is proven.
### 10A. Duplicate-work gate phases (#400)
Before any file edits, prove duplicate-work clearance with
`gitea_assess_work_issue_duplicate` or `gitea_lock_issue` (which runs the same
gate). The gate checks live:
* open PRs linked to the issue (head branch or Closes/Fixes reference),
* remote branches matching `issue-<number>`,
* active claim leases from structured heartbeats.
Re-check immediately before:
* `gitea_commit_files` (commit),
* branch push,
* `gitea_create_pr` (PR creation).
If a concurrent open PR appears after work begins:
* before commit/push → stop and preserve local work without pushing,
* after commit but before push → stop without pushing,
* after push but before PR creation → stop and produce a reconciliation
handoff instead of opening a PR.
Final reports must state exactly one duplicate-work outcome:
* `duplicate PR prevented`
* `duplicate branch prevented`
* `duplicate commit prevented`
* `duplicate work not prevented`
## 11. Claim or lock the issue before implementation
Claim/lock the issue before implementation if the project provides a claim/lock mechanism.
If claim/lock requires a Gitea mutation, prove exact capability first.
If claim/lock fails, stop.
Do not implement unclaimed work.
If the claim/lock gates are broken, produce a recovery handoff.
Create a tooling issue only if this run is explicitly authorized to switch to issue-creation mode and exact `create_issue` capability is proven.
Report:
* claim mechanism used
* claim result
* claim timestamp, if available
* issue owner/assignee after claim, if available
## 12. Refresh stable branch before branch/worktree creation
Fetch the stable target branch from the remote before creating a task branch or worktree.
Do not rely on stale local `master`, `main`, or `dev`.
Record the fetched stable branch SHA.
If the stable branch cannot be fetched or verified, stop and produce a recovery handoff.
`git fetch`, `git remote update`, and any command that updates refs must be reported under `Git ref mutations`, not read-only diagnostics.
## 13. Branch and worktree ownership rule
Create a fresh session-owned worktree under `branches/`.
Prefer a branch name that includes the issue number, for example:
`feat/issue-<ISSUE_NUMBER>-short-description`
or:
`fix/issue-<ISSUE_NUMBER>-short-description`
Prefer a worktree path like:
`branches/issue-<ISSUE_NUMBER>-short-description`
Before any file edits, prove:
* project root
* current working directory
* main checkout branch
* stable branch
* stable branch SHA
* task branch name
* session-owned worktree path
* worktree path is inside `branches/`
* worktree is not the main checkout
* clean tracked state
* clean untracked state
* worktree HEAD/branch state
Do not reuse an existing worktree unless safe-reuse proof passes.
Safe-reuse proof must include:
* exact worktree path
* worktree is inside `branches/`
* worktree is not the main checkout
* worktree is not owned by another active task/session
* clean tracked state
* clean untracked state
* current branch/head before reset
* reset target SHA
* explicit project policy allowing reuse/reset
Do not run `git reset --hard`, `git clean`, checkout, or other destructive commands unless the worktree is session-owned or safe-reuse proof passes.
If safe-reuse proof cannot be produced, create a fresh session-owned worktree.
## 14. Implementation scope rule
Implement only what is required for the selected issue.
Do not perform opportunistic refactors.
Do not fix unrelated tests unless they are required for the selected issue and clearly documented.
Do not modify reviewer workflow files unless the selected issue explicitly requires workflow changes.
Do not modify Gitea profiles, MCP authorization, tokens, secrets, deployment config, production config, or credentials unless the selected issue explicitly requires it and exact capability/proof gates pass.
Do not introduce provenance markers, agent signatures, temporary files, debug dumps, or generated artifacts unless required.
If implementation uncovers a separate issue, note it in the final report or create a follow-up issue only if exact capability is proven and project policy allows it.
## 15. File edit rule
All edits must happen only inside the session-owned issue worktree.
Do not edit files in the main checkout.
Do not edit files in reviewer worktrees.
Do not edit unrelated worktrees.
Track every edited, created, deleted, or generated file.
If any file is edited, created, generated, or written, report it under `File edits by author`.
For each file write, report:
* exact path
* whether it was inside the repo
* whether it was tracked or untracked
* why it was created
* whether final `git status` was run after the write
Do not say `File edits by author: none` if any file write occurred.
Do not write files after the final clean-status check unless you rerun and report a new final clean-status check.
## 16. Validation rule
Run appropriate validation for the selected issue.
Validation may include:
* targeted tests
* full test suite
* compile checks
* lint checks
* type checks
* diff checks
* secret/provenance checks
* dangerous artifact checks
* project-specific validation
If validation cannot run, explain why and include the exact failure.
Do not hide failures.
Do not claim success if tests failed.
Do not skip required validation silently.
Do not bypass MCP gates.
Report every validation command with:
* exact command
* working directory
* exit code or pass/fail result
* summary count if available
* whether it was targeted, full-suite, compile, lint, diff, secret/provenance, or diagnostic validation
If using bare `pytest`, also report:
* `which pytest`
* `pytest --version`
* whether it resolves to the project venv
Prefer the project venv executable when available.
## 17. Baseline comparison rule
Do not run tests in the main checkout.
If the full suite fails and you need to prove failures are pre-existing, create a clean baseline worktree under `branches/`, such as:
`branches/baseline-master-issue-<ISSUE_NUMBER>`
Baseline comparison must include:
* baseline worktree path
* baseline target SHA
* task branch SHA
* exact command run on both worktrees
* baseline failures
* task branch failures
* proof the failure signatures match
* proof the baseline worktree was clean before and after validation
* proof the issue worktree was clean before and after validation
Do not claim “same as master” unless the clean baseline worktree proof is included.
Do not claim “full-suite failures are pre-existing” unless baseline proof is complete and the failure signatures match.
If full-suite failures differ or proof is incomplete, do not create a PR unless project policy explicitly allows PR creation with documented validation failures.
## 18. Pre-commit review
Before committing, review the actual diff.
Check:
* correctness
* tests
* scope
* security boundaries
* workflow rule compliance
* whether the implementation really satisfies the selected issue
* unrelated changes
* dangerous generated artifacts
* secrets
* provenance markers
* temporary agent files
* debug output
* formatting-only churn
* docs/tests consistency
Run:
* `git status`
* `git diff --stat`
* `git diff`
* project-required diff checks
Do not commit if unrelated or unsafe changes are present.
## 19. Commit rules
Commit only from the session-owned issue worktree.
Do not commit from the main checkout.
Commit only after implementation and required validation pass, unless project policy explicitly allows draft PRs with failing validation.
Commit message must reference the issue number.
Preferred format:
`fix: short summary (Closes #<ISSUE_NUMBER>)`
or:
`feat: short summary (Closes #<ISSUE_NUMBER>)`
Before commit, prove:
* worktree path
* branch name
* selected issue number
* staged files
* diff summary
* validation status
After commit, record:
* commit SHA
* commit message
* changed files
Do not amend, reset, rebase, squash, or force-push unless the project workflow explicitly allows it and the worktree is session-owned.
## 20. Push rules
Push only the session-owned task branch.
Do not push `master`, `main`, `dev`, tags, or unrelated branches.
Before push, prove:
* current branch
* upstream/remote target
* commit SHA being pushed
* selected issue number
* branch name matches the issue
After push, report:
* remote
* branch
* pushed commit SHA
* push result
If push fails, stop and produce a recovery handoff.
## 20A. Conflict-fix lease and push gate (#399)
When pushing to an existing PR branch to resolve merge conflicts:
1. Call `gitea_acquire_conflict_fix_lease` before any push.
2. Call `gitea_assess_conflict_fix_push` immediately before `git push` with:
* branch head before push
* branch head after push (local)
* session worktree path
* push cwd
* whether the push is fast-forward
3. Do not push when a reviewer holds an active lease on the same PR.
4. Do not force-push.
5. Do not push from the main checkout or wrong cwd.
Conflict-fix final reports must state:
* branch head before push
* branch head after push
* active reviewer lease status
* whether push was fast-forward
* whether any reviewer was active
## 21. PR creation rules
Create a PR only if implementation and validation pass, unless project policy explicitly allows draft PRs with documented validation failures.
Do not create a PR if:
* issue was not claimed/locked
* issue eligibility was unproven
* duplicate open PR exists
* task branch does not reference the issue
* implementation is incomplete
* validation failed without allowed exception
* worktree is dirty
* secrets/provenance/dangerous artifacts are present
* capability for PR creation is missing
* runtime context is blocked
* authenticated identity/profile changed unexpectedly
PR must reference or close the issue.
PR body must include:
* summary
* linked issue
* files changed
* validation commands and results
* risk
* exact worktree path
* branch name
* commit SHA
* known limitations, if any
Do not merge your own PR.
Do not approve your own PR.
Do not request changes on your own PR.
After PR creation, fetch or view the PR to verify:
* PR number
* PR URL
* PR title
* base branch
* head branch
* linked issue
* head SHA
* open status
## 22. Cleanup rules
Clean only session-owned temporary/baseline worktrees if the project workflow explicitly allows cleanup.
Do not delete unrelated branches/worktrees.
Do not delete the task worktree if the project expects it to remain for handoff unless policy says cleanup is allowed after PR creation.
Do not update the main checkout unless the canonical workflow explicitly allows it.
Any cleanup is a mutation and must be reported.
## 23. Recovery handoff rules
If blocked, produce a recovery handoff with:
* exact blocker
* failed tool/function, if any
* repo/project
* selected issue, if one was safely selected
* eligibility class
* claim/lock state
* branch name, if created
* worktree path, if created
* stable branch and stable branch SHA, if known
* files changed, if any
* validation state
* commit SHA, if committed
* PR number/URL, if created
* exact state reached before stopping
* safe next action
* statement that no unsafe mutation was attempted
Blocked handoffs must not include direct commit, push, or PR replay commands.
Blocked handoffs must say to rerun the full workflow after the blocker clears.
## 24. Final report must be precise
Include:
* canonical workflow source/version/hash, if available
* authenticated identity/profile
* repo/project
* capability proof summary
* issue inventory proof, including pagination/final-page proof
* issue ordering policy used
* selected issue number/title
* eligibility class
* skipped earlier issues and proof, if any
* duplicate active work proof
* claim/lock result
* stable branch and stable branch SHA
* branch name
* worktree path
* worktree inside `branches/`: true/false
* worktree branch/HEAD state
* worktree dirty before implementation: true/false
* files changed
* validation commands and results
* baseline comparison result, if used
* pre-commit diff review result
* commit SHA and commit message, if committed
* push result, if pushed
* PR number and URL, if created
* PR verification result, if created
* cleanup result
* blockers, if stopped
* confirmation that the main checkout was not used for task work
If the report and actual tool/command log disagree, fix the report before final output.
## 25. Final report must distinguish mutation types
Do not use the legacy field `Workspace mutations`.
Use only precise categories:
* File edits by author:
* Worktree/index mutations:
* Git ref mutations:
* MCP/Gitea mutations:
* Issue mutations:
* Branch mutations:
* Commit mutations:
* Push mutations:
* PR mutations:
* Cleanup mutations:
* External-state mutations:
* Read-only diagnostics:
Issue-lock file (`/tmp/gitea_issue_lock.json`) read/write/delete is always an
external-state mutation. Never claim `External-state mutations: none` after
seeding, restoring, or removing that file. Manual lock seeding is not a normal
recovery path (#447); use `gitea_lock_issue` or the #442 adoption recovery path
instead. Link broader redesign: #438.
`git fetch`, `git remote update`, and any command that updates refs must be listed under `Git ref mutations`, not read-only diagnostics.
If `git reset --hard`, checkout, clean, worktree add/remove, merge simulation, merge abort, or similar commands occurred, report them under `Worktree/index mutations`.
Use precise wording:
* `File edits by author: none`
* `Worktree/index mutations: ...`
* `Git ref mutations: ...`
* `MCP/Gitea mutations: ...`
Do not collapse issue, branch, commit, push, PR, cleanup, or external-state mutations into vague wording.
## 26. Forbidden final-report claims unless proven
Do not claim:
* `next eligible issue`
* `oldest eligible issue`
* `issue claimed`
* `no duplicate work`
* `no open PR`
* `worktree clean`
* `validation passed`
* `same as master`
* `full-suite failures are pre-existing`
* `committed`
* `pushed`
* `PR created`
* `issue closed`
* `main checkout untouched`
* `no file edits`
* `no unsafe mutation`
* `all gates passed`
* `target branch up to date`
unless the corresponding proof is included.
If anything blocks safe work or PR creation, stop immediately and produce an executable recovery handoff.
Do not improvise around the gates.
## 27. Proof wording enforcement
The following phrases are forbidden unless directly supported by current-session evidence:
* next eligible issue
* oldest eligible issue
* inventory complete
* no duplicate work
* issue claimed
* worktree clean
* validation passed
* same as master
* full-suite failures are pre-existing
* committed
* pushed
* PR created
* issue closed
* target branch up to date
* all gates passed
* no unsafe mutation
* no file edits
If the proof comes from prior state rather than a command/tool run in the current session, label it as prior proof, not live proof.
If a tool call was rejected, confirmation-gated, dry-run-only, or no-op, report it separately from performed mutations.
## 28. Final self-check before output
Before final output, check the report for contradictions.
Verify:
* if any file was edited, `File edits by author` is not `none`
* if any worktree was added/removed, `Worktree/index mutations` lists it
* if any fetch happened, `Git ref mutations` lists it
* if any issue was claimed/commented/updated, `Issue mutations` lists it
* if any branch was created, `Branch mutations` lists it
* if any commit was created, `Commit mutations` lists it
* if any push occurred, `Push mutations` lists it
* if any PR was created, `PR mutations` lists it
* if any cleanup happened, `Cleanup mutations` lists it
* if any issue/PR external state changed, `External-state mutations` lists it
* if pagination is claimed complete, final-page proof is present
* if same-as-master is claimed, baseline proof is complete
* if selected issue is claimed next eligible, every earlier issue has proof-backed skip reasoning
* if PR created is claimed, PR verification proof is present
* if main checkout untouched is claimed, main checkout status proof is present
If any contradiction exists, fix the final report before output.
## 29. Controller handoff schema
End every run with a controller handoff using this schema.
Do not omit fields. Use `none` or `not verified in this session` where appropriate.
Controller Handoff:
* Task:
* Repo:
* Role:
* Identity:
* Active profile:
* Runtime context:
* Selected issue:
* Eligibility class:
* Issue ordering policy:
* Issue inventory pagination proof:
* Earlier issues skipped:
* Duplicate active work proof:
* Claim/lock state:
* Stable branch:
* Stable branch SHA:
* Branch name:
* Worktree path:
* Worktree inside branches:
* Worktree branch/HEAD state:
* Worktree dirty before implementation:
* Files changed:
* Validation:
* Baseline comparison:
* Commit SHA:
* Push result:
* PR number:
* PR URL:
* PR verification:
* Main checkout branch:
* Main checkout dirty state:
* Main checkout used for task work:
* File edits by author:
* Worktree/index mutations:
* Git ref mutations:
* MCP/Gitea mutations:
* Issue mutations:
* Branch mutations:
* Commit mutations:
* Push mutations:
* PR mutations:
* Cleanup mutations:
* External-state mutations:
* Read-only diagnostics:
* Blockers:
* Current status:
* Safe next action:
* Safety statement:
## 30. Stop conditions summary
Stop immediately and produce a recovery handoff if:
* canonical workflow is required but cannot be loaded
* identity/profile/capability cannot be proven
* runtime context is blocked
* infra stop appears
* MCP reconnect fails
* capability state is stale
* issue inventory pagination cannot be proven
* issue ordering cannot be proven
* issue eligibility cannot be proven
* duplicate active work cannot be checked
* selected issue is already claimed by another worker
* selected issue already has an open PR
* claim/lock fails
* stable branch cannot be fetched
* task worktree cannot be created safely
* task worktree is dirty before implementation
* validation cannot run
* validation fails without allowed exception
* baseline comparison is required but incomplete
* diff review finds unrelated or unsafe changes
* commit fails
* push fails
* PR creation fails
* PR verification fails
* any report contradiction cannot be resolved
Blocked handoffs must not include direct commit, push, or PR replay commands.
Blocked handoffs must say to rerun the full workflow after the blocker clears.
Do not improvise around the gates.
+211
View File
@@ -0,0 +1,211 @@
"""Stacked-PR support for author issue locks and PR creation (#484).
Normal author work locks a worktree that is base-equivalent to ``master``/
``main``/``dev`` and opens a PR against one of those base branches. A *stacked*
PR is deliberately based on another unmerged PR's branch, so its worktree is not
master-equivalent and its PR base is not a normal base branch.
This module holds the pure decision logic that lets:
* ``gitea_lock_issue`` approve a non-master base **only** when it is explicitly
declared and proven to correspond to an open pull request, and
* ``gitea_create_pr`` accept that approved base while still rejecting arbitrary,
mismatched, or stale (merged/closed) branches.
The normal master-based path is unchanged: when no stacked base is declared, and
when the PR base is a normal base branch, these helpers are inert. Nothing here
bypasses the issue lock a stacked base is recorded *on* the lock and re-checked
at PR time.
"""
from __future__ import annotations
BASE_BRANCHES = frozenset({"master", "main", "dev"})
# Phrases that satisfy the required merge-ordering statement in a stacked PR body.
MERGE_ORDER_PHRASES = ("do not merge before", "do not merge until")
def is_base_branch(base: str | None, base_branches: frozenset[str] | None = None) -> bool:
"""True when ``base`` is a normal base branch (master/main/dev)."""
bases = base_branches or BASE_BRANCHES
return (base or "").strip() in bases
def _pr_head_ref(pr: dict) -> str:
head = pr.get("head") or {}
if isinstance(head, dict):
return (head.get("ref") or "").strip()
return (str(head) if head else "").strip()
def find_open_pr_for_branch(open_prs: list[dict] | None, branch: str | None) -> dict | None:
"""Return the first OPEN PR whose head ref equals ``branch`` (else ``None``)."""
branch = (branch or "").strip()
if not branch:
return None
for pr in open_prs or []:
if (pr.get("state") or "").strip().lower() != "open":
continue
if _pr_head_ref(pr) == branch:
return pr
return None
def assess_stacked_base_declaration(
*,
stacked_base_branch: str | None,
stacked_base_pr: int | None,
open_prs: list[dict] | None,
) -> dict:
"""Validate an explicit stacked-base declaration at lock time.
Returns a dict with ``block`` (fail closed), ``reasons``, ``declared``
(whether a stacked base was requested), and ``approved`` (the metadata to
persist on the lock when valid, else ``None``).
"""
branch = (stacked_base_branch or "").strip()
if not branch:
# No stacked base requested — normal master-based lock path.
return {"block": False, "reasons": [], "approved": None, "declared": False}
if branch in BASE_BRANCHES:
return {
"block": True,
"declared": True,
"approved": None,
"reasons": [
f"stacked base '{branch}' is already a normal base branch; do not "
"declare a base branch as a stacked base"
],
}
if stacked_base_pr is None:
return {
"block": True,
"declared": True,
"approved": None,
"reasons": [
"stacked base branch declared without stacked_base_pr; a stacked PR "
"must cite the open PR that owns the base branch"
],
}
pr = find_open_pr_for_branch(open_prs, branch)
if pr is None:
return {
"block": True,
"declared": True,
"approved": None,
"reasons": [
f"stacked base branch '{branch}' does not correspond to any OPEN pull "
"request; arbitrary or stale branches are not allowed as stacked bases"
],
}
if int(pr.get("number")) != int(stacked_base_pr):
return {
"block": True,
"declared": True,
"approved": None,
"reasons": [
f"declared stacked_base_pr #{stacked_base_pr} does not match the open "
f"PR #{pr.get('number')} that owns base branch '{branch}'"
],
}
return {
"block": False,
"declared": True,
"reasons": [],
"approved": {
"branch": branch,
"pr_number": int(pr.get("number")),
"verified_open": True,
},
}
def assess_stacked_pr_body(
body: str | None, *, base_branch: str | None, pr_number: int | None
) -> list[str]:
"""Return the list of missing stacked-PR documentation fields (empty = ok)."""
text = body or ""
low = text.lower()
missing: list[str] = []
if base_branch and base_branch not in text:
missing.append(f"base branch '{base_branch}'")
if pr_number is not None and f"#{pr_number}" not in text:
missing.append(f"stacked-on PR reference '#{pr_number}'")
if not any(phrase in low for phrase in MERGE_ORDER_PHRASES):
missing.append("merge-ordering statement (e.g. 'Do not merge before PR #<n>')")
return missing
def assess_create_pr_base(
*,
base: str | None,
approved_stacked_base: dict | None,
body: str | None,
open_prs: list[dict] | None,
base_branches: frozenset[str] | None = None,
) -> dict:
"""Validate the PR base at create time.
Normal base branches pass through unchanged (``stacked`` False). A non-base
branch is allowed only when it matches the lock's approved stacked base, that
base still has an open PR, and the body documents the stack.
"""
bases = base_branches or BASE_BRANCHES
base = (base or "").strip()
if base in bases:
return {"block": False, "reasons": [], "stacked": False}
approved = approved_stacked_base or {}
approved_branch = (approved.get("branch") or "").strip()
if not approved_branch:
return {
"block": True,
"stacked": True,
"reasons": [
f"PR base '{base}' is not one of {'/'.join(sorted(bases))} and the "
"issue lock has no approved stacked base; re-lock with an explicit, "
"proof-backed stacked base to open a stacked PR"
],
}
if base != approved_branch:
return {
"block": True,
"stacked": True,
"reasons": [
f"PR base '{base}' does not match the issue lock's approved stacked "
f"base '{approved_branch}'"
],
}
pr = find_open_pr_for_branch(open_prs, base)
if pr is None:
return {
"block": True,
"stacked": True,
"reasons": [
f"approved stacked base '{base}' no longer corresponds to an OPEN pull "
"request (dependency merged, closed, or stale); retarget/rebase onto "
"master or re-lock against a live base"
],
}
pr_number = approved.get("pr_number") or pr.get("number")
missing = assess_stacked_pr_body(body, base_branch=base, pr_number=pr_number)
if missing:
return {
"block": True,
"stacked": True,
"reasons": [
"stacked PR body must document the stack; missing: "
+ ", ".join(missing)
],
}
return {"block": False, "reasons": [], "stacked": True, "stacked_base_pr": pr_number}
+166
View File
@@ -0,0 +1,166 @@
"""Fail-closed subagent delegation gates (#266).
Subagents can bypass or lose context for worktree, capability, mutation,
retry, and reporting rules. These helpers make delegation an explicit,
provable decision instead of a default: deterministic write tasks stay
inline unless the parent session records why a subagent is needed and
hands the subagent the full gate context it must operate under.
Like ``author_proofs``/``review_proofs``, the helpers are pure (no git,
no API calls): the parent workflow gathers the facts and passes them in,
so the same logic works from prompts, harness assertions, and tests.
Nothing here weakens the review/merge/permission gates a delegated
subagent is subject to the same gates as its parent.
"""
# AC1: deterministic write workflows a subagent must never run by default.
DETERMINISTIC_WRITE_TASKS = frozenset({
"claim_issue",
"create_branch",
"edit_code",
"commit",
"push_branch",
"create_pr",
"review_pr",
"merge_pr",
"cleanup_branch",
"close_issue",
})
# Read-only delegation that needs no explicit authorization.
READ_ONLY_TASKS = frozenset({
"read_files",
"code_search",
"inventory_prs",
"summarize_issue",
"explore_codebase",
})
# AC3: context a subagent must inherit from the parent session before any
# authorized write delegation may proceed.
REQUIRED_INHERITED_CONTEXT = (
"issue_lock",
"branch",
"worktree_path",
"identity_profile",
"allowed_tool_class",
"command_deny_list",
"validation_ledger_requirement",
"final_report_schema",
)
# AC4: proof fields a subagent final report must carry — the same fields a
# parent workflow's final report requires.
REQUIRED_SUBAGENT_REPORT_FIELDS = (
"identity_profile",
"worktree_path",
"branch",
"changed_files",
"validation_results",
"workspace_mutations",
)
def _clean(value):
return (value or "").strip() if isinstance(value, str) else value
def _classify_task(task_type):
task = _clean(task_type)
if not task:
return "", "unknown"
if task in DETERMINISTIC_WRITE_TASKS:
return task, "deterministic_write"
if task in READ_ONLY_TASKS:
return task, "read_only"
return task, "unknown"
def _missing_context_fields(inherited_context):
context = inherited_context or {}
missing = []
for field in REQUIRED_INHERITED_CONTEXT:
value = context.get(field)
if not isinstance(value, str) or not value.strip():
missing.append(field)
return missing
def assess_subagent_delegation(task_type, *, explicitly_allowed=False,
justification=None, inherited_context=None):
"""Decide whether delegating *task_type* to a subagent may proceed.
Fail closed: unknown tasks are blocked; deterministic write tasks are
blocked unless explicitly allowed (AC1) with a recorded justification
(AC2) and the full inherited gate context (AC3). Read-only delegation
is allowed without explicit authorization.
Returns {'block', 'allowed', 'task_type', 'task_class', 'reasons',
'missing_context'}.
"""
task, task_class = _classify_task(task_type)
reasons = []
missing_context = []
if task_class == "unknown":
reasons.append(
f"task type '{task}' is not a recognized delegation class; "
"run it inline in the parent session (fail closed)"
)
elif task_class == "deterministic_write":
if not explicitly_allowed:
reasons.append(
f"deterministic write task '{task}' must run inline unless "
"subagent use is explicitly allowed by the parent session"
)
if not _clean(justification):
reasons.append(
"no recorded justification for why a subagent is needed; "
"the parent session must record one before delegating"
)
missing_context = _missing_context_fields(inherited_context)
if missing_context:
reasons.append(
"subagent would not inherit required gate context: "
+ ", ".join(missing_context)
)
block = bool(reasons)
return {
"block": block,
"allowed": not block,
"task_type": task,
"task_class": task_class,
"reasons": reasons,
"missing_context": missing_context,
}
def validate_subagent_report(report_fields):
"""AC4: accept subagent output only with the parent-grade proof fields.
*report_fields* maps field name -> reported value. Missing or blank
proof fields make the report invalid (fail closed).
Returns {'valid', 'block', 'reasons', 'missing_fields'}.
"""
reasons = []
report = report_fields or {}
missing = []
for field in REQUIRED_SUBAGENT_REPORT_FIELDS:
value = report.get(field)
if not isinstance(value, str) or not value.strip():
missing.append(field)
if missing:
reasons.append(
"subagent final report missing required proof fields: "
+ ", ".join(missing)
)
valid = not reasons
return {
"valid": valid,
"block": not valid,
"reasons": reasons,
"missing_fields": missing,
}
+180
View File
@@ -0,0 +1,180 @@
"""Shared task→permission map for resolver and tool gates (#69).
``gitea_resolve_task_capability`` and issue-mutating MCP tools must agree on
which profile operation each task requires. This module is the single source
of truth; regression tests assert tool gates cannot drift from it.
"""
from __future__ import annotations
TASK_CAPABILITY_MAP: dict[str, dict[str, str]] = {
"create_issue": {
"permission": "gitea.issue.create",
"role": "author",
},
"comment_issue": {
"permission": "gitea.issue.comment",
"role": "author",
},
"close_issue": {
"permission": "gitea.issue.close",
"role": "author",
},
"claim_issue": {
"permission": "gitea.issue.comment",
"role": "author",
},
"mark_issue": {
"permission": "gitea.issue.comment",
"role": "author",
},
"lock_issue": {
"permission": "gitea.issue.comment",
"role": "author",
},
"set_issue_labels": {
"permission": "gitea.issue.comment",
"role": "author",
},
"create_branch": {
"permission": "gitea.branch.create",
"role": "author",
},
"push_branch": {
"permission": "gitea.branch.push",
"role": "author",
},
"create_pr": {
"permission": "gitea.pr.create",
"role": "author",
},
"comment_pr": {
"permission": "gitea.pr.comment",
"role": "author",
},
"close_pr": {
"permission": "gitea.pr.close",
"role": "author",
},
"address_pr_change_requests": {
"permission": "gitea.branch.push",
"role": "author",
},
"review_pr": {
"permission": "gitea.pr.review",
"role": "reviewer",
},
"merge_pr": {
"permission": "gitea.pr.merge",
"role": "reviewer",
},
"blind_pr_queue_review": {
"permission": "gitea.pr.review",
"role": "reviewer",
},
"pr_queue_cleanup": {
"permission": "gitea.pr.review",
"role": "reviewer",
},
"pr-queue-cleanup": {
"permission": "gitea.pr.review",
"role": "reviewer",
},
"request_changes_pr": {
"permission": "gitea.pr.request_changes",
"role": "reviewer",
},
"approve_pr": {
"permission": "gitea.pr.approve",
"role": "reviewer",
},
"delete_branch": {
"permission": "gitea.branch.delete",
"role": "author",
},
"commit_files": {
"permission": "gitea.repo.commit",
"role": "author",
},
"gitea_commit_files": {
"permission": "gitea.repo.commit",
"role": "author",
},
"reconcile_merged_cleanups": {
"permission": "gitea.read",
"role": "author",
},
"work_issue": {
"permission": "gitea.pr.create",
"role": "author",
},
"work-issue": {
"permission": "gitea.pr.create",
"role": "author",
},
"reconcile_landed_pr": {
"permission": "gitea.read",
"role": "author",
},
"reconcile-landed-pr": {
"permission": "gitea.read",
"role": "author",
},
"reconcile_already_landed_pr": {
"permission": "gitea.pr.close",
"role": "reconciler",
},
# #309: dedicated reconciler path for already-landed open PRs. Exact
# close capabilities only — never review/approve/request_changes/merge.
"reconcile_close_landed_pr": {
"permission": "gitea.pr.close",
"role": "reconciler",
},
"reconcile_close_landed_issue": {
"permission": "gitea.issue.close",
"role": "reconciler",
},
"post_heartbeat": {
"permission": "gitea.issue.comment",
"role": "author",
},
"reconcile_issue_claims": {
"permission": "gitea.read",
"role": "author",
},
"cleanup_stale_claims": {
"permission": "gitea.issue.comment",
"role": "author",
},
}
# Issue-mutating MCP tools and their resolver task keys.
ISSUE_MUTATION_TOOL_TASKS: dict[str, str] = {
"gitea_create_issue": "create_issue",
"gitea_close_issue": "close_issue",
"gitea_create_issue_comment": "comment_issue",
"gitea_mark_issue": "mark_issue",
"gitea_set_issue_labels": "set_issue_labels",
"gitea_commit_files": "commit_files",
}
def required_permission(task: str) -> str:
"""Return the canonical operation a *task* requires (fail closed)."""
try:
return TASK_CAPABILITY_MAP[task]["permission"]
except KeyError as exc:
raise KeyError(f"Unknown task/action: {task!r} (fail closed)") from exc
def required_role(task: str) -> str:
"""Return author/reviewer role kind for *task* (fail closed)."""
try:
return TASK_CAPABILITY_MAP[task]["role"]
except KeyError as exc:
raise KeyError(f"Unknown task/action: {task!r} (fail closed)") from exc
def tool_required_permission(tool_name: str) -> str:
"""Return the operation an issue-mutating tool must gate on."""
return required_permission(ISSUE_MUTATION_TOOL_TASKS[tool_name])
+11
View File
@@ -25,4 +25,15 @@ def _reset_mutation_authority(monkeypatch):
return
monkeypatch.setattr(mcp_server, "_MUTATION_AUTHORITY", None)
monkeypatch.setattr(mcp_server, "_IDENTITY_CACHE", {})
monkeypatch.setattr(mcp_server, "_REVIEW_DECISION_LOCK", None)
try:
import capability_stop_terminal
capability_stop_terminal.clear()
except Exception:
pass
yield
try:
import capability_stop_terminal
capability_stop_terminal.clear()
except Exception:
pass
+113
View File
@@ -0,0 +1,113 @@
"""Tests for agent temp artifact detection and preflight warnings (#261)."""
import json
import os
import sys
import tempfile
import unittest
from unittest.mock import patch
sys.path.insert(0, str(__import__("pathlib").Path(__file__).resolve().parent.parent))
import agent_temp_artifacts
import mcp_server
class TestAgentTempArtifactPatterns(unittest.TestCase):
def test_detects_root_level_encode_emit_inline(self):
porcelain = (
"?? _encode_commit_payload.py\n"
"?? _emit_payload.py\n"
"?? _inline_b64.py\n"
)
self.assertEqual(
agent_temp_artifacts.find_agent_temp_artifacts_from_porcelain(porcelain),
["_emit_payload.py", "_encode_commit_payload.py", "_inline_b64.py"],
)
def test_ignores_nested_and_unrelated_untracked(self):
porcelain = (
"?? scripts/_encode_x.py\n"
"?? README-draft.md\n"
" M task_capability_map.py\n"
)
self.assertEqual(
agent_temp_artifacts.find_agent_temp_artifacts_from_porcelain(porcelain),
[],
)
class TestPreflightWarnings(unittest.TestCase):
def setUp(self):
self._snapshot = (
mcp_server._preflight_whoami_called,
mcp_server._preflight_capability_called,
)
mcp_server._preflight_whoami_called = True
mcp_server._preflight_capability_called = True
def tearDown(self):
mcp_server._preflight_whoami_called, mcp_server._preflight_capability_called = (
self._snapshot
)
@patch(
"mcp_server._get_workspace_porcelain",
return_value="?? _encode_commit_payload.py\n",
)
def test_assess_preflight_surfaces_warning_not_block(self, _porcelain):
status = mcp_server.assess_preflight_status()
self.assertTrue(status["preflight_ready"])
self.assertEqual(status["preflight_block_reasons"], [])
self.assertEqual(len(status["preflight_warnings"]), 1)
self.assertIn("_encode_commit_payload.py", status["preflight_warnings"][0])
# Issue-write tools are profile-gated (#69); gitea_lock_issue requires
# gitea.issue.comment (see task_capability_map), so the gate must be
# seeded exactly like tests/test_mcp_server.py::TestIssueLocking (#359).
ISSUE_WRITE_ENV = {
"GITEA_ALLOWED_OPERATIONS": (
"gitea.issue.create,gitea.issue.close,gitea.issue.comment"
),
}
class TestIssueLockArtifactWarning(unittest.TestCase):
def setUp(self):
self._lock_dir = tempfile.TemporaryDirectory()
env = {**ISSUE_WRITE_ENV, "GITEA_ISSUE_LOCK_DIR": self._lock_dir.name}
self._env_patcher = patch.dict(os.environ, env, clear=True)
self._env_patcher.start()
def tearDown(self):
self._env_patcher.stop()
self._lock_dir.cleanup()
@patch(
"mcp_server.issue_duplicate_context_fetcher",
return_value=([], [], {"status": "not_claimed"}),
)
@patch("mcp_server.api_get_all", return_value=[])
@patch("mcp_server._auth", return_value="token x")
@patch("mcp_server._resolve", return_value=("h", "o", "r"))
@patch("issue_lock_worktree.read_worktree_git_state")
def test_lock_success_includes_artifact_warning(self, mock_state, *_mocks):
mock_state.return_value = {
"current_branch": "master",
"porcelain_status": "?? _emit_payload.py\n",
"base_equivalent": True,
"inspected_git_root": "/scratch/wt",
"base_branch": "origin/master",
}
result = mcp_server.gitea_lock_issue(
issue_number=261,
branch_name="docs/issue-261-agent-artifact-cleanup",
remote="prgs",
)
self.assertTrue(result["success"])
self.assertIn("warnings", result)
self.assertIn("_emit_payload.py", result["warnings"][0])
if __name__ == "__main__":
unittest.main()
+131
View File
@@ -0,0 +1,131 @@
"""Tests for already-landed PR reconciliation gates (#310)."""
import os
import sys
import unittest
from unittest.mock import patch
sys.path.insert(0, str(__import__("pathlib").Path(__file__).resolve().parent.parent))
import already_landed_reconcile
class TestAssessAlreadyLandedReconciliation(unittest.TestCase):
def test_open_pr_already_landed_allows_close(self):
with patch(
"already_landed_reconcile.is_head_ancestor_of_ref",
return_value=True,
):
assessment = already_landed_reconcile.assess_already_landed_reconciliation(
pr={
"number": 278,
"state": "open",
"title": "Fix thing (Closes #263)",
"body": "",
"head": {"ref": "feat/x", "sha": "abc123"},
"base": {"ref": "master"},
},
project_root="/tmp/repo",
remote="prgs",
target_branch="master",
target_fetch={
"success": True,
"target_branch": "master",
"target_ref": "prgs/master",
"target_branch_sha": "deadbeef",
"git_fetch_command": "git fetch prgs master",
},
)
self.assertTrue(assessment["close_allowed"])
self.assertEqual(
assessment["eligibility_class"],
already_landed_reconcile.ELIGIBILITY_ALREADY_LANDED,
)
self.assertEqual(assessment["linked_issue"], 263)
self.assertTrue(assessment["ancestor_proof"])
def test_not_landed_pr_denies_close(self):
with patch(
"already_landed_reconcile.is_head_ancestor_of_ref",
return_value=False,
):
assessment = already_landed_reconcile.assess_already_landed_reconciliation(
pr={
"number": 99,
"state": "open",
"title": "WIP",
"body": "",
"head": {"ref": "feat/y", "sha": "fff111"},
"base": {"ref": "master"},
},
project_root="/tmp/repo",
remote="prgs",
target_branch="master",
target_fetch={
"success": True,
"target_branch": "master",
"target_ref": "prgs/master",
"target_branch_sha": "deadbeef",
},
)
self.assertFalse(assessment["close_allowed"])
self.assertEqual(
assessment["eligibility_class"],
already_landed_reconcile.ELIGIBILITY_NOT_LANDED,
)
def test_stale_target_branch_denies_close(self):
assessment = already_landed_reconcile.assess_already_landed_reconciliation(
pr={
"number": 99,
"state": "open",
"title": "WIP",
"body": "",
"head": {"ref": "feat/y", "sha": "fff111"},
"base": {"ref": "master"},
},
project_root="/tmp/repo",
remote="prgs",
target_branch="master",
target_fetch={
"success": False,
"target_branch": "master",
"target_ref": "prgs/master",
"target_branch_sha": None,
"reasons": ["git fetch failed"],
},
)
self.assertFalse(assessment["close_allowed"])
self.assertEqual(
assessment["eligibility_class"],
already_landed_reconcile.ELIGIBILITY_STALE_TARGET,
)
def test_closed_pr_denies_close(self):
assessment = already_landed_reconcile.assess_already_landed_reconciliation(
pr={
"number": 50,
"state": "closed",
"title": "Done",
"body": "",
"head": {"ref": "feat/z", "sha": "aaa"},
"base": {"ref": "master"},
},
project_root="/tmp/repo",
remote="prgs",
target_branch="master",
target_fetch={
"success": True,
"target_branch": "master",
"target_ref": "prgs/master",
"target_branch_sha": "deadbeef",
},
)
self.assertFalse(assessment["close_allowed"])
self.assertEqual(
assessment["eligibility_class"],
already_landed_reconcile.ELIGIBILITY_PR_NOT_OPEN,
)
if __name__ == "__main__":
unittest.main()
+151
View File
@@ -0,0 +1,151 @@
"""Tests for already-landed final-report wording validator (#298).
An already-landed PR is reconciliation-only, never review/merge
eligible, and a head SHA may not be called reviewed unless validation
and diff review actually passed. Final reports and controller handoffs
must use the reconciliation wording, not the legacy eligible/reviewed
fields.
"""
import sys
import unittest
from pathlib import Path
sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
from review_proofs import assess_already_landed_report_wording # noqa: E402
GOOD_ALREADY_LANDED = (
"Controller Handoff\n"
"- Oldest open PR requiring action: PR #278\n"
"- Eligibility class: ALREADY_LANDED_RECONCILE_REQUIRED\n"
"- Candidate head SHA: 2a544b7\n"
"- Reviewed head SHA: none\n"
"- Review worktree used: false\n"
)
class TestAlreadyLandedWording(unittest.TestCase):
def test_correct_reconciliation_wording_passes(self):
result = assess_already_landed_report_wording(
GOOD_ALREADY_LANDED, already_landed=True
)
self.assertTrue(result["complete"])
self.assertFalse(result["downgraded"])
self.assertEqual(result["reasons"], [])
def test_oldest_eligible_wording_fails(self):
report = GOOD_ALREADY_LANDED + "- oldest eligible PR: PR #278\n"
result = assess_already_landed_report_wording(
report, already_landed=True
)
self.assertFalse(result["complete"])
self.assertTrue(
any("oldest eligible pr" in r.lower() for r in result["reasons"])
)
def test_next_eligible_wording_fails(self):
report = GOOD_ALREADY_LANDED + "- next eligible PR: PR #278\n"
result = assess_already_landed_report_wording(
report, already_landed=True
)
self.assertFalse(result["complete"])
def test_pinned_reviewed_head_fails(self):
report = GOOD_ALREADY_LANDED + "- Pinned reviewed head: 2a544b7\n"
result = assess_already_landed_report_wording(
report, already_landed=True
)
self.assertFalse(result["complete"])
self.assertTrue(
any("pinned reviewed head" in r.lower() for r in result["reasons"])
)
def test_scratch_worktree_field_fails(self):
report = GOOD_ALREADY_LANDED + "- Scratch worktree used: False\n"
result = assess_already_landed_report_wording(
report, already_landed=True
)
self.assertFalse(result["complete"])
def test_missing_eligibility_class_fails(self):
report = (
"Controller Handoff\n"
"- Oldest open PR requiring action: PR #278\n"
"- Candidate head SHA: 2a544b7\n"
"- Reviewed head SHA: none\n"
"- Review worktree used: false\n"
)
result = assess_already_landed_report_wording(
report, already_landed=True
)
self.assertFalse(result["complete"])
self.assertTrue(
any("eligibility class" in r.lower() for r in result["reasons"])
)
def test_populated_reviewed_head_sha_fails(self):
report = GOOD_ALREADY_LANDED.replace(
"- Reviewed head SHA: none\n",
"- Reviewed head SHA: 2a544b7\n",
)
result = assess_already_landed_report_wording(
report, already_landed=True
)
self.assertFalse(result["complete"])
self.assertTrue(
any("reviewed head" in r.lower() for r in result["reasons"])
)
class TestNonAlreadyLandedReports(unittest.TestCase):
def test_normal_reviewed_report_passes(self):
report = (
"- Selected PR: 281\n"
"- Pinned reviewed head: abc1234\n"
"- Review decision: approve\n"
)
result = assess_already_landed_report_wording(
report, already_landed=False, review_validated=True
)
self.assertTrue(result["complete"])
def test_blocked_infra_report_with_pinned_head_fails(self):
report = (
"- Selected PR: 281\n"
"- Pinned reviewed head: abc1234\n"
"- Current status: blocked on infra_stop\n"
)
result = assess_already_landed_report_wording(
report, already_landed=False, review_validated=False
)
self.assertFalse(result["complete"])
self.assertTrue(
any("before validation" in r.lower() for r in result["reasons"])
)
def test_validation_failed_report_with_reviewed_sha_fails(self):
report = (
"- Selected PR: 281\n"
"- Reviewed head SHA: abc1234\n"
"- Validation: full suite failed\n"
)
result = assess_already_landed_report_wording(
report, already_landed=False, review_validated=False
)
self.assertFalse(result["complete"])
def test_validation_failed_report_with_reviewed_none_passes(self):
report = (
"- Selected PR: 281\n"
"- Reviewed head SHA: none\n"
"- Validation: full suite failed\n"
)
result = assess_already_landed_report_wording(
report, already_landed=False, review_validated=False
)
self.assertTrue(result["complete"])
if __name__ == "__main__":
unittest.main()

Some files were not shown because too many files have changed in this diff Show More