Compare commits

..
Author SHA1 Message Date
sysadmin e423dd5870 fix(mcp): remediate B8, B6/B11, and B15 break-glass restart blockers (#664)
- B8: Correct redaction boundary for GITEA_TOKEN= and URI userinfo without destroying adjacent audit evidence or benign sec- text
- B6/B11: Remove false restart execution claims from default executor when GITEA_SANCTIONED_RESTART_HOOK is non-empty
- B15: Document deployable production grant set (runtime.break_glass_restart and gitea.issue.create) for prgs-controller
- Preserve B13, B1, B14 and previously accepted corrections
2026-07-29 01:23:10 -04:00
sysadminandClaude Opus 4.8 c67f39b40e fix(#664): remediate PR #908 review #641 blockers B13, B1, B14, B6/B11, and B8
Register runtime.break_glass_restart in the multi-service operation normalizer
and enforce it through the real profile gate. Authorize only the exact trusted
prgs-controller profile plus that capability; remove substring controller
authority so declared reconciler roles and cleanup_merged_pr_branch semantics
are preserved. Route non-dry-run apply through a canonical injectable executor
delegate with truthful execution flags. Correct under/over-redaction for
credentials while preserving benign sec- text.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
2026-07-28 23:33:25 -04:00
sysadmin 4463a300ba fix(#664): remediate break-glass restart workflow blockers B1, B6/B11, B9, B10, B8, and B12 2026-07-28 22:44:43 -04:00
sysadmin 75794609d1 fix(mcp): remediate break-glass restart authorization and audit findings (#664) 2026-07-28 21:45:31 -04:00
jcwalker3 da3294fbe5 Merge branch 'master' into feat/issue-664-break-glass-restart 2026-07-28 08:39:10 -05:00
sysadmin c1ecadce8e feat(mcp): implement emergency break-glass MCP restart workflow (#664) 2026-07-25 17:06:38 -04:00
29 changed files with 2000 additions and 6464 deletions
-111
View File
@@ -1913,88 +1913,6 @@ class ControlPlaneDB:
),
)
def retire_lease_worktree_path(
self,
lease_id: str,
*,
expected_path: str | None = None,
reason: str = "missing_worktree_path_retired",
) -> dict[str, Any]:
"""Retire a missing worktree_path binding from a control-plane lease (#970).
Clears worktree_path on the lease row, updates provenance_json with
durable retirement audit proof, and writes a worktree_binding_retired event.
Fail closed: if expected_path is provided and does not match current
worktree_path, the update is refused to prevent racing mutations.
"""
now_s = _ts()
with self._tx() as conn:
cols = self._lease_columns(conn)
row = conn.execute(
"SELECT * FROM leases WHERE lease_id = ?",
(lease_id,),
).fetchone()
if not row:
raise ControlPlaneError(f"unknown lease_id {lease_id}")
current_wt = (dict(row).get("worktree_path") or "").strip()
if expected_path and current_wt and os.path.realpath(current_wt) != os.path.realpath(expected_path):
raise ControlPlaneError(
f"cannot retire lease {lease_id} worktree_path: expected '{expected_path}' "
f"does not match current '{current_wt}' (fail closed)"
)
# Parse and update provenance_json
raw_prov = dict(row).get("provenance_json") or "{}"
try:
prov = json.loads(raw_prov) if isinstance(raw_prov, str) else dict(raw_prov)
except Exception:
prov = {}
if not isinstance(prov, dict):
prov = {}
prior_path = current_wt or prov.get("worktree_path")
prov.update({
"worktree_path_retired": True,
"retired_worktree_path": prior_path,
"retired_at": now_s,
"retirement_reason": reason,
"worktree_path": "",
})
prov_json = json.dumps(prov)
if "worktree_path" in cols:
conn.execute(
"UPDATE leases SET worktree_path = '', provenance_json = ? WHERE lease_id = ?",
(prov_json, lease_id),
)
else:
conn.execute(
"UPDATE leases SET provenance_json = ? WHERE lease_id = ?",
(prov_json, lease_id),
)
conn.execute(
"""
INSERT INTO events(work_item_id, event_type, message, created_at)
VALUES (?, 'worktree_binding_retired', ?, ?)
""",
(
row["work_item_id"],
f"lease {lease_id} worktree_path '{prior_path}' retired: {reason}",
now_s,
),
)
return {
"lease_id": lease_id,
"retired": True,
"prior_worktree_path": prior_path,
"retired_at": now_s,
"reason": reason,
}
def abandon_lease(
self,
*,
@@ -3133,32 +3051,3 @@ class ControlPlaneDB:
"live_lease_id": None if live_lease_id is None else str(live_lease_id),
"reconcile_action": "reconcile_required" if stale else "safe_to_resume",
}
def retire_session_checkpoint_worktree_path(
self,
session_id: str,
*,
checkpoint_id: str | None = None,
expected_path: str | None = None,
reason: str = "missing_worktree_path_retired",
) -> dict[str, Any]:
"""Retire a missing worktree_path from session_checkpoints (#970)."""
now_s = _ts()
with self._tx() as conn:
if checkpoint_id:
conn.execute(
"UPDATE session_checkpoints SET worktree_path = '', updated_at = ? WHERE checkpoint_id = ?",
(now_s, checkpoint_id),
)
elif session_id:
conn.execute(
"UPDATE session_checkpoints SET worktree_path = '', updated_at = ? WHERE session_id = ?",
(now_s, session_id),
)
return {
"session_id": session_id,
"checkpoint_id": checkpoint_id,
"retired": True,
"retired_at": now_s,
"reason": reason,
}
-71
View File
@@ -86,74 +86,3 @@ When a namespace returns EOF, follow
When blocked, repair the IDE namespace and re-record a healthy
`client_namespace` assessment before retrying the mutation.
## Connected vs Attached Tool Surface (#708)
MCP servers can report **Connected** at the CLI / host inventory layer while the **active LLM session exposes none of their tool namespaces**.
### Core principle
* **Connected status at host layer ≠ attached tools in active session.**
* Required preflight proof is **live tool visibility + `gitea_whoami` call** through the target namespace, not host `Connected` status alone.
* When servers report Connected but namespaces are absent from attached tools, classify as `mcp_connected_namespaces_missing`.
### Forbidden unsafe fallbacks
When `mcp_connected_namespaces_missing` is detected, workflows must **fail closed** and must **never** encourage or perform:
* direct imports of MCP server Python modules
* CLI or raw Gitea API mutations as a substitute for native tools
* profile hopping to another MCP profile/namespace to bypass the empty session
* session-state overrides or hand-edited session/ledger files
* process kills (`pkill`), config mtime touches, or `.env` edits
Only sanctioned recovery: **client reconnect path**, followed by full preflight (`whoami` → capability resolve → task).
### Native detection tool
`gitea_assess_mcp_namespace_attachment` classifies the condition and records it in
the session. Pass the namespaces the host reports Connected and the namespaces
actually attached to the active session tool surface:
| Argument | Meaning |
|---|---|
| `connected_servers` | Namespaces the host/CLI reports Connected |
| `attached_session_namespaces` | Namespaces exposed in the active session tool surface |
| `required_namespaces` | Namespaces this workflow needs (defaults to the role namespaces) |
| `discovery_cache_hit` / `discovery_cache_age_seconds` | Client tool-discovery cache state |
| `auto_attach_attempted` / `auto_attach_succeeded` | Whether the runtime auto-attached |
| `session_tool_snapshot_at` / `namespace_connected_at` | Epoch seconds, to detect startup ordering races |
It returns `discovery_status`
(`namespaces_attached` | `connected_but_namespaces_missing` | `disconnected`),
`missing_namespaces`, `proof_of_connected_vs_attached` (per namespace
`{connected, attached}`), `error_type`, `reconnect_required`, `auto_recovered`,
`startup_ordering_race`, `late_attaching_namespaces`, `sanctioned_recovery_tool`,
and a reconnect-only `exact_next_action`.
### Startup ordering
`session_tool_snapshot_at` earlier than a namespace's `namespace_connected_at`
means that namespace could not have been in the session snapshot, however healthy
it looks now. That is reported as `startup_ordering_race` with the affected
namespaces listed — the multi-role parallel-connect case where Connected flips
true after the session tool list was already captured.
### Fail-closed gate
The recorded verdict gates mutations, mirroring the #543 health gate: a namespace
that has not been assessed does not gate, but one recorded Connected-but-unattached
blocks the mutation whose role namespace it is
(`review_pr` / `submit_review``gitea-reviewer`, `merge_pr``gitea-merger`,
`work_issue` / `create_pr``gitea-author`). `gitea_submit_pr_review` and
`gitea_merge_pr` return the block reason plus the hard-stop policy string, and the
only offered recovery is `gitea_request_mcp_reconnect`.
### Telemetry
The `telemetry` block carries `connected_count`, `attached_count`,
`required_count`, `missing_count`, `discovery_status`, `discovery_cache_hit`,
`discovery_cache_age_seconds`, `reconnect_required`, `auto_attach_attempted`,
`auto_recovered`, `startup_ordering_race`, and `error_type`. It contains namespace
names and counts only — never tokens, endpoints, env values, or filesystem paths.
+14
View File
@@ -144,6 +144,19 @@ tool argument expresses caller intent and cannot be self-asserted by a worker
session. `break_glass_requested` and `break_glass_authorized` are both reported,
so a bypass is never silent.
### Break-glass Restart Workflow (`gitea_break_glass_restart`, #664)
The dedicated MCP tool `gitea_break_glass_restart` provides the privileged emergency break-glass restart workflow when graceful drain cannot complete:
- **Authorization (#664 AC1 / B1 / B13 / B15)**: Requires the exact trusted profile `prgs-controller` **and** explicit `runtime.break_glass_restart` and `gitea.issue.create` grants enforced by the real production operation gate (no `gitea.read` fallback). Incident creation is mandatory prior to execution (`gitea.issue.create`), so the deployable production policy for `prgs-controller` includes `allowed_operations`: `["gitea.read", "gitea.pr.close", "gitea.pr.comment", "gitea.issue.comment", "gitea.issue.create", "runtime.break_glass_restart", "gitea.branch.delete", "gitea.decision_lock.irrecoverable_recovery"]`. Ordinary roles, non-controller reconcilers, lookalike profile names (`fake-controller`, …), and env vars cannot authorize. A narrow break-glass capability does **not** redefine the profile's declared global role. Updating a live running `prgs-controller` profile in production requires an operator configuration update and daemon reload post-merge.
- **Required Parameters (#664 AC2)**:
- `reason`: Mandatory non-empty string (min 10 characters).
- `confirmation`: Must equal exactly `"I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION"`.
- `impact_ack`: Must be `True`.
- **Automatic Incident Creation (#664 AC3)**: Creates a Gitea incident issue (`[INCIDENT] [REQUESTED] Break-glass MCP restart invoked by ...`) detailing the reason, timestamp, disrupted sessions, and linking `#652 #653 #655 #630 #658 #662 #664`.
- **Immutable Append-Only Audit Entry**: Records immutable pre-execution (REQUESTED) and post-execution (SUCCEEDED/FAILED) audit log entries with correlation identifiers.
- **Mandatory Reconciliation (#664 AC4)**: Sets `reconciliation_required=True` requiring post-restart reconciliation via `gitea_reconcile_after_restart` (#662).
### Fail closed on apply
A missing, malformed, expired, unclean, tampered, or fingerprint-stale drain
@@ -160,3 +173,4 @@ profiles are operational metadata only.
A representative dry-run report is in
[`mcp-restart-impact-sample.json`](./mcp-restart-impact-sample.json).
-6
View File
@@ -56,7 +56,6 @@ that gates each call, not which tools exist.
- `gitea_assess_conflict_fix_push`
- `gitea_assess_gitea_operation_path`
- `gitea_assess_master_parity`
- `gitea_assess_mcp_namespace_attachment`
- `gitea_assess_mcp_namespace_health`
- `gitea_assess_pr_sync_status`
- `gitea_assess_review_merge_state_machine`
@@ -65,7 +64,6 @@ that gates each call, not which tools exist.
- `gitea_assess_work_issue_duplicate`
- `gitea_assess_worktree_cleanup_integrity`
- `gitea_audit_config`
- `gitea_audit_missing_worktree_bindings`
- `gitea_audit_runtime_recovery_contamination`
- `gitea_audit_stable_branch_contamination`
- `gitea_audit_worktree_cleanup`
@@ -127,20 +125,16 @@ that gates each call, not which tools exist.
- `gitea_post_heartbeat`
- `gitea_publish_unpublished_issue_branch`
- `gitea_quarantine_contaminated_review`
- `gitea_rebind_dirty_same_claimant_author_session`
- `gitea_reclaim_expired_workflow_lease`
- `gitea_reconcile_after_restart`
- `gitea_reconcile_already_landed_pr`
- `gitea_reconcile_issue_claims`
- `gitea_reconcile_merged_cleanups`
- `gitea_reconcile_missing_worktree_bindings`
- `gitea_reconcile_superseded_by_merged_pr`
- `gitea_record_daemon_process_kill_attempt`
- `gitea_record_irrecoverable_decision_lock_provenance`
- `gitea_record_pre_review_command`
- `gitea_record_shell_spawn_outcome`
- `gitea_record_stable_branch_push_attempt`
- `gitea_recover_dirty_orphaned_issue_worktree`
- `gitea_recover_incomplete_bootstrap_lock`
- `gitea_release_merger_pr_lease`
- `gitea_release_reviewer_pr_lease`
+67 -233
View File
@@ -8,239 +8,73 @@
"document. #930's inventory had no such guard and its gitea_mcp_server.py",
"anchors drifted between 7bf4f125 and aad5c8b4."
],
"generated_against_commit": "1dd30ecb1508b559868c2d5d94367bc055d5138e",
"generated_against_commit": "aad5c8b42361d380a8eeb07b94b90815e594c2c5",
"anchors": [
{
"anchor": "gitea_mcp_server.py:25087",
"expect": "mcp_daemon_guard.bind_native_mcp_transport()"
},
{
"anchor": "mcp_daemon_guard.py:49",
"expect": "_PRODUCTION_TRANSPORTS = mcp_transport_config.SUPPORTED_TRANSPORTS"
},
{
"anchor": "mcp_daemon_guard.py:195",
"expect": "def bind_native_mcp_transport"
},
{
"anchor": "irrecoverable_provenance.py:497",
"expect": "def assess_transport_for_auth_mint"
},
{
"anchor": "gitea_mcp_server.py:9192",
"expect": "assess_transport_for_auth_mint()"
},
{
"anchor": "gitea_mcp_server.py:9441",
"expect": "assess_transport_for_auth_mint()"
},
{
"anchor": "mcp_server.py:4",
"expect": "The transport is selected by deployment configuration"
},
{
"anchor": "gitea_mcp_server.py:15630",
"expect": "def _is_client_managed_process"
},
{
"anchor": "gitea_mcp_server.py:15644",
"expect": "def _provenance_mutation_block"
},
{
"anchor": "gitea_mcp_server.py:15652",
"expect": "unsupported_manual_launch"
},
{
"anchor": "gitea_mcp_server.py:19217",
"expect": "server_provenance"
},
{
"anchor": "gitea_mcp_server.py:21741",
"expect": "def _check_mcp_runtimes_diagnostics"
},
{
"anchor": "gitea_mcp_server.py:21761",
"expect": "\"ps\", \"-o\", \"pid,lstart,command\""
},
{
"anchor": "gitea_mcp_server.py:21805",
"expect": "\"ps\", \"eww\""
},
{
"anchor": "gitea_config.py:1172",
"expect": "RECOGNIZED_GITEA_ENV_KEYS"
},
{
"anchor": "gitea_config.py:1233",
"expect": "GITEA_CLIENT_MANAGED"
},
{
"anchor": "gitea_config.py:54",
"expect": "ENV_PROFILE = \"GITEA_MCP_PROFILE\""
},
{
"anchor": "gitea_config.py:97",
"expect": "_REVIEW_MERGE_OPS"
},
{
"anchor": "gitea_config.py:499",
"expect": "repository authorization scope"
},
{
"anchor": "gitea_config.py:956",
"expect": "def _keychain_token"
},
{
"anchor": "gitea_config.py:974",
"expect": "def resolve_token"
},
{
"anchor": "gitea_config.py:1015",
"expect": "def keychain_auth"
},
{
"anchor": "gitea_config.py:294",
"expect": "def _validate_identity_auth"
},
{
"anchor": "mcp_daemon_guard.py:583",
"expect": "def assert_keychain_access_allowed"
},
{
"anchor": "gitea_mcp_server.py:19487",
"expect": "def gitea_list_profiles"
},
{
"anchor": "gitea_mcp_server.py:19538",
"expect": "gitea_config.resolve_token(p)"
},
{
"anchor": "gitea_mcp_server.py:19851",
"expect": "def gitea_audit_config"
},
{
"anchor": "gitea_mcp_server.py:19873",
"expect": "service_summaries(config)"
},
{
"anchor": "gitea_config.py:704",
"expect": "def resolve_service"
},
{
"anchor": "gitea_config.py:837",
"expect": "def service_summaries"
},
{
"anchor": "gitea_config.py:851",
"expect": "_keychain_token(auth.get(\"id\"))"
},
{
"anchor": "gitea_mcp_server.py:17909",
"expect": "\"jenkins-mcp\""
},
{
"anchor": "gitea_mcp_server.py:17915",
"expect": "external-mcp"
},
{
"anchor": "gitea_mcp_server.py:17936",
"expect": "\"glitchtip-mcp\""
},
{
"anchor": "gitea_mcp_server.py:17941",
"expect": "external-mcp"
},
{
"anchor": "mcp_discoverability.py:9",
"expect": "EXPECTED_JENKINS_TOOLS"
},
{
"anchor": "mcp_discoverability.py:17",
"expect": "EXPECTED_GLITCHTIP_TOOLS"
},
{
"anchor": "sentry_incident_bridge.py:36",
"expect": "SENTRY_AUTH_TOKEN"
},
{
"anchor": "sentry_incident_bridge.py:190",
"expect": "def resolve_token"
},
{
"anchor": "sentry_incident_bridge.py:289",
"expect": "Authorization"
},
{
"anchor": "sentry_observability.py:55",
"expect": "SENTRY_DSN"
},
{
"anchor": "master_parity_gate.py:168",
"expect": "def capture_startup_parity"
},
{
"anchor": "master_parity_gate.py:255",
"expect": "mutation_safe"
},
{
"anchor": "gitea_mcp_server.py:19331",
"expect": "def gitea_assess_master_parity"
},
{
"anchor": "gitea_mcp_server.py:193",
"expect": "ACTIVE_WORKTREE_ENV"
},
{
"anchor": "gitea_mcp_server.py:194",
"expect": "AUTHOR_WORKTREE_ENV"
},
{
"anchor": "gitea_mcp_server.py:2352",
"expect": "/tmp/gitea_issue_lock.json"
},
{
"anchor": "gitea_mcp_server.py:10957",
"expect": "def gitea_bootstrap_author_issue_worktree"
},
{
"anchor": "mcp_server.py:13",
"expect": "/tmp/mcp_server_stderr.log"
},
{
"anchor": "issue_lock_store.py:26",
"expect": "DEFAULT_LOCK_DIR"
},
{
"anchor": "issue_lock_store.py:83",
"expect": "def session_pointer_path"
},
{
"anchor": "issue_lock_store.py:98",
"expect": "def is_process_alive"
},
{
"anchor": "mcp_session_state.py:27",
"expect": "DEFAULT_STATE_DIR"
},
{
"anchor": "control_plane_db.py:47",
"expect": "DEFAULT_DB_PATH"
},
{
"anchor": "control_plane_db.py:380",
"expect": "mode=0o700"
},
{
"anchor": "control_plane_db.py:386",
"expect": "sqlite3.connect"
},
{
"anchor": "control_plane_db.py:1145",
"expect": "os.getpid()"
},
{
"anchor": "gitea_mcp_server.py:12871",
"expect": "owner_pid_alive"
}
{"anchor": "gitea_mcp_server.py:24721", "expect": "bind_native_mcp_transport(transport=\"stdio\")"},
{"anchor": "mcp_daemon_guard.py:45", "expect": "_PRODUCTION_TRANSPORTS = frozenset({\"stdio\"})"},
{"anchor": "mcp_daemon_guard.py:174", "expect": "def bind_native_mcp_transport"},
{"anchor": "irrecoverable_provenance.py:497", "expect": "def assess_transport_for_auth_mint"},
{"anchor": "gitea_mcp_server.py:9129", "expect": "assess_transport_for_auth_mint()"},
{"anchor": "gitea_mcp_server.py:9378", "expect": "assess_transport_for_auth_mint()"},
{"anchor": "mcp_server.py:4", "expect": "Runs over stdio."},
{"anchor": "gitea_mcp_server.py:15412", "expect": "def _is_client_managed_process"},
{"anchor": "gitea_mcp_server.py:15442", "expect": "def _provenance_mutation_block"},
{"anchor": "gitea_mcp_server.py:15450", "expect": "unsupported_manual_launch"},
{"anchor": "gitea_mcp_server.py:19001", "expect": "server_provenance"},
{"anchor": "gitea_mcp_server.py:21442", "expect": "def _check_mcp_runtimes_diagnostics"},
{"anchor": "gitea_mcp_server.py:21462", "expect": "\"ps\", \"-o\", \"pid,lstart,command\""},
{"anchor": "gitea_mcp_server.py:21506", "expect": "\"ps\", \"eww\""},
{"anchor": "gitea_config.py:1172", "expect": "RECOGNIZED_GITEA_ENV_KEYS"},
{"anchor": "gitea_config.py:1233", "expect": "GITEA_CLIENT_MANAGED"},
{"anchor": "gitea_config.py:54", "expect": "ENV_PROFILE = \"GITEA_MCP_PROFILE\""},
{"anchor": "gitea_config.py:97", "expect": "_REVIEW_MERGE_OPS"},
{"anchor": "gitea_config.py:499", "expect": "repository authorization scope"},
{"anchor": "gitea_config.py:956", "expect": "def _keychain_token"},
{"anchor": "gitea_config.py:974", "expect": "def resolve_token"},
{"anchor": "gitea_config.py:1015", "expect": "def keychain_auth"},
{"anchor": "gitea_config.py:294", "expect": "def _validate_identity_auth"},
{"anchor": "mcp_daemon_guard.py:440", "expect": "def assert_keychain_access_allowed"},
{"anchor": "gitea_mcp_server.py:19258", "expect": "def gitea_list_profiles"},
{"anchor": "gitea_mcp_server.py:19309", "expect": "gitea_config.resolve_token(p)"},
{"anchor": "gitea_mcp_server.py:19552", "expect": "def gitea_audit_config"},
{"anchor": "gitea_mcp_server.py:19574", "expect": "service_summaries(config)"},
{"anchor": "gitea_config.py:704", "expect": "def resolve_service"},
{"anchor": "gitea_config.py:837", "expect": "def service_summaries"},
{"anchor": "gitea_config.py:851", "expect": "_keychain_token(auth.get(\"id\"))"},
{"anchor": "gitea_mcp_server.py:17707", "expect": "\"jenkins-mcp\""},
{"anchor": "gitea_mcp_server.py:17713", "expect": "external-mcp"},
{"anchor": "gitea_mcp_server.py:17734", "expect": "\"glitchtip-mcp\""},
{"anchor": "gitea_mcp_server.py:17739", "expect": "external-mcp"},
{"anchor": "mcp_discoverability.py:9", "expect": "EXPECTED_JENKINS_TOOLS"},
{"anchor": "mcp_discoverability.py:17", "expect": "EXPECTED_GLITCHTIP_TOOLS"},
{"anchor": "sentry_incident_bridge.py:36", "expect": "SENTRY_AUTH_TOKEN"},
{"anchor": "sentry_incident_bridge.py:190", "expect": "def resolve_token"},
{"anchor": "sentry_incident_bridge.py:289", "expect": "Authorization"},
{"anchor": "sentry_observability.py:55", "expect": "SENTRY_DSN"},
{"anchor": "master_parity_gate.py:168", "expect": "def capture_startup_parity"},
{"anchor": "master_parity_gate.py:255", "expect": "mutation_safe"},
{"anchor": "gitea_mcp_server.py:19102", "expect": "def gitea_assess_master_parity"},
{"anchor": "gitea_mcp_server.py:190", "expect": "ACTIVE_WORKTREE_ENV"},
{"anchor": "gitea_mcp_server.py:191", "expect": "AUTHOR_WORKTREE_ENV"},
{"anchor": "gitea_mcp_server.py:2348", "expect": "/tmp/gitea_issue_lock.json"},
{"anchor": "gitea_mcp_server.py:10894", "expect": "def gitea_bootstrap_author_issue_worktree"},
{"anchor": "mcp_server.py:10", "expect": "/tmp/mcp_server_stderr.log"},
{"anchor": "issue_lock_store.py:26", "expect": "DEFAULT_LOCK_DIR"},
{"anchor": "issue_lock_store.py:83", "expect": "def session_pointer_path"},
{"anchor": "issue_lock_store.py:98", "expect": "def is_process_alive"},
{"anchor": "mcp_session_state.py:27", "expect": "DEFAULT_STATE_DIR"},
{"anchor": "control_plane_db.py:47", "expect": "DEFAULT_DB_PATH"},
{"anchor": "control_plane_db.py:380", "expect": "mode=0o700"},
{"anchor": "control_plane_db.py:386", "expect": "sqlite3.connect"},
{"anchor": "control_plane_db.py:1145", "expect": "os.getpid()"},
{"anchor": "gitea_mcp_server.py:12801", "expect": "owner_pid_alive"}
]
}
+29 -35
View File
@@ -5,11 +5,7 @@ What the adversary is, what each boundary protects, and which services may share
- **Issue:** #956 (Remote-MCP threat model), child of epic #929, cross-linked to #955.
- **Depends on:** #930 (closed) — `docs/remote-mcp/coupling-inventory.md`.
- **Blocks:** #932, #933, #934, #938.
- **Generated against commit:** `1dd30ecb1508b559868c2d5d94367bc055d5138e` (#708's
namespace-attachment gate). Originally generated against
`aad5c8b42361d380a8eeb07b94b90815e594c2c5` (`master`), re-anchored at
`a143cd065ba06e1a2bdc5143a19ec156e53650ef` when #931's transport bind seam shifted the
cited lines, and re-anchored again when #708 shifted them further.
- **Generated against commit:** `aad5c8b42361d380a8eeb07b94b90815e594c2c5` (`master`).
- **Scope:** documentation only. This child changes no server behavior. It adds one
document, one anchor fixture, and the test that enforces them.
@@ -31,7 +27,7 @@ document cites an anchor the fixture does not cover.
This guard exists because #930 did not have one. Its inventory was generated at
`7bf4f125`; by `aad5c8b4` its `gitea_mcp_server.py` anchors had drifted — the transport
bind it cited at line 23750 now lives at `gitea_mcp_server.py:25087`, and its
bind it cited at line 23750 now lives at `gitea_mcp_server.py:24721`, and its
client-managed provenance anchor at 14588 now lands in an unrelated function. Nothing
failed, because nothing checked. Anchors into a ~24,700-line module rot silently, and a
security document that cannot prove its own citations is worse than none, because it is
@@ -75,19 +71,19 @@ authenticate the *caller*, not the *intent*.
## 3. Trust boundaries
"Crossing requires today" is what the code actually enforces at
`1dd30ecb1508b559868c2d5d94367bc055d5138e`, not what the design intends.
`aad5c8b42361d380a8eeb07b94b90815e594c2c5`, not what the design intends.
| ID | Boundary | Protects | Crossing requires today | Crossing must require remotely |
| -- | -------- | -------- | ----------------------- | ------------------------------ |
| B1 | LLM client ↔ MCP server session | A1, A3, A10 — that a mutating session was established through the sanctioned client path | A single configured bind (`gitea_mcp_server.py:25087`) validated against one closed allowlist (`mcp_daemon_guard.py:49`, `mcp_daemon_guard.py:195`) — since #931 the identifier comes from deployment configuration and defaults to the local transport, so the boundary no longer rests on a literal, but it still rests on the *bind* rather than on an authenticated caller; client-managed provenance (`gitea_mcp_server.py:15630`) or a refusal (`gitea_mcp_server.py:15652`); production transport before recovery-authorization mint (`irrecoverable_provenance.py:497`, consumed at `gitea_mcp_server.py:9192` and `gitea_mcp_server.py:9441`) | An authenticated handshake issuing a server-side session identity bound to a principal, with the transport recorded in provenance. The physical proof (a pipe) must become a cryptographic one. |
| B1 | LLM client ↔ MCP server session | A1, A3, A10 — that a mutating session was established through the sanctioned client path | A literal `stdio` bind (`gitea_mcp_server.py:24721`) inside a closed allowlist (`mcp_daemon_guard.py:45`, `mcp_daemon_guard.py:174`); client-managed provenance (`gitea_mcp_server.py:15412`) or a refusal (`gitea_mcp_server.py:15450`); production transport before recovery-authorization mint (`irrecoverable_provenance.py:497`, consumed at `gitea_mcp_server.py:9129` and `gitea_mcp_server.py:9378`) | An authenticated handshake issuing a server-side session identity bound to a principal, with the transport recorded in provenance. The physical proof (a pipe) must become a cryptographic one. |
| B2 | Role ↔ role | A9 — that author, reviewer, merger, and reconciler are distinct authorities | **The process boundary only.** The role is a property of the process, read once from `GITEA_MCP_PROFILE` (`gitea_config.py:54`). A caller gets author permissions by connecting to the author process. Review and merge are the operations singled out for extra care (`gitea_config.py:97`) | A per-request principal, so the role follows from the credential presented and cannot be selected by reaching a different endpoint. |
| B3 | MCP server ↔ credential store | A3, A8 — that only sanctioned code turns a profile into a token | `_keychain_token` shelling out to the login keychain (`gitea_config.py:956`), dispatched by `resolve_token` (`gitea_config.py:974`) with the reference type built at `gitea_config.py:1015`, gated by `assert_keychain_access_allowed` (`mcp_daemon_guard.py:583`). Inline secrets are rejected at config load (`gitea_config.py:294`) | A credential provider keyed by the *request* principal, returning only that principal's credential, with the source recorded and the value never returned. |
| B3 | MCP server ↔ credential store | A3, A8 — that only sanctioned code turns a profile into a token | `_keychain_token` shelling out to the login keychain (`gitea_config.py:956`), dispatched by `resolve_token` (`gitea_config.py:974`) with the reference type built at `gitea_config.py:1015`, gated by `assert_keychain_access_allowed` (`mcp_daemon_guard.py:440`). Inline secrets are rejected at config load (`gitea_config.py:294`) | A credential provider keyed by the *request* principal, returning only that principal's credential, with the source recorded and the value never returned. |
| B4 | MCP server ↔ Gitea | A1, A2 — that only authorized calls reach the forge | A bearer token over TLS. Server-side, nothing distinguishes one role's token from another beyond the account it belongs to | Unchanged at the forge; the endpoint in front of it must refuse unauthenticated and plaintext connections before tool dispatch. |
| B5 | MCP server ↔ caller's filesystem | A7 — that a tool acts on the *caller's* disk or refuses | Nothing. The server's disk *is* the caller's disk. Worktree bootstrap writes directly (`gitea_mcp_server.py:10957`); the active workspace is process-global (`gitea_mcp_server.py:193`, `gitea_mcp_server.py:194`) | An explicit per-tool classification, enforced at dispatch, refusing filesystem tools over a transport that cannot reach the caller's disk. A green verdict about the wrong disk is the failure to prevent. |
| B6 | MCP server ↔ coordination state | A6, A9 — mutual exclusion | Local files and a local SQLite database, with liveness judged from the local process table (`issue_lock_store.py:98`), keyed on paths under one user's home (`issue_lock_store.py:26`, `mcp_session_state.py:27`, `control_plane_db.py:47`) and on `os.getpid()` (`control_plane_db.py:1145`, `gitea_mcp_server.py:12871`). A legacy global slot still exists at `gitea_mcp_server.py:2352`, and the session-pointer file is named per PID (`issue_lock_store.py:83`) | One authority per ownership question, with liveness from session identity and expiry, and atomic acquire, renew, and release across hosts. |
| B5 | MCP server ↔ caller's filesystem | A7 — that a tool acts on the *caller's* disk or refuses | Nothing. The server's disk *is* the caller's disk. Worktree bootstrap writes directly (`gitea_mcp_server.py:10894`); the active workspace is process-global (`gitea_mcp_server.py:190`, `gitea_mcp_server.py:191`) | An explicit per-tool classification, enforced at dispatch, refusing filesystem tools over a transport that cannot reach the caller's disk. A green verdict about the wrong disk is the failure to prevent. |
| B6 | MCP server ↔ coordination state | A6, A9 — mutual exclusion | Local files and a local SQLite database, with liveness judged from the local process table (`issue_lock_store.py:98`), keyed on paths under one user's home (`issue_lock_store.py:26`, `mcp_session_state.py:27`, `control_plane_db.py:47`) and on `os.getpid()` (`control_plane_db.py:1145`, `gitea_mcp_server.py:12801`). A legacy global slot still exists at `gitea_mcp_server.py:2348`, and the session-pointer file is named per PID (`issue_lock_store.py:83`) | One authority per ownership question, with liveness from session identity and expiry, and atomic acquire, renew, and release across hosts. |
| B7 | Gitea integration ↔ unrelated integrations | A4, A5 — that a Gitea compromise is not a CI and observability compromise | **Nothing.** See §5. The Gitea server reads Jenkins and GlitchTip secrets (`gitea_config.py:851`, reached from `gitea_config.py:837`) and holds the Sentry token (`sentry_incident_bridge.py:190`) | A hard process boundary. This is the boundary #956 exists to create. |
| B8 | Tenant ↔ tenant (`prgs` / `mdcps` / `local-lab`) | A2 — that one organization's compromise is not another's | Convention. One configuration declares all three contexts; `resolve_service` fails closed on a *disabled* context (`gitea_config.py:704`) but the credentials of enabled ones remain reachable in-process. A per-profile repository scope exists (`gitea_config.py:499`) | Separate deployments, or at minimum per-tenant credential scopes with no process able to resolve both. |
| B9 | Deployed code ↔ merged policy | A1, A10 — that the running server enforces the rules that were actually merged | Comparing this process's startup commit against this disk (`master_parity_gate.py:168`), conjoined into a single verdict (`master_parity_gate.py:255`) published by `gitea_mcp_server.py:19331` | Freshness defined against the deployed build identity, with an explicit fail-closed verdict when undeterminable. |
| B9 | Deployed code ↔ merged policy | A1, A10 — that the running server enforces the rules that were actually merged | Comparing this process's startup commit against this disk (`master_parity_gate.py:168`), conjoined into a single verdict (`master_parity_gate.py:255`) published by `gitea_mcp_server.py:19102` | Freshness defined against the deployed build identity, with an explicit fail-closed verdict when undeterminable. |
### What no boundary constrains
@@ -132,10 +128,10 @@ Two flows deserve attention because neither is obvious from the code:
1. **The keychain flow fans out.** B3 is drawn once but resolves credentials for *every*
configured profile and service, not only the active one. `gitea_list_profiles`
(`gitea_mcp_server.py:19487`) reports each profile's credential status by calling
`resolve_token` on it (`gitea_mcp_server.py:19538`), and `gitea_audit_config`
(`gitea_mcp_server.py:19851`) reports service credential status through
`service_summaries` (`gitea_mcp_server.py:19873`).
(`gitea_mcp_server.py:19258`) reports each profile's credential status by calling
`resolve_token` on it (`gitea_mcp_server.py:19309`), and `gitea_audit_config`
(`gitea_mcp_server.py:19552`) reports service credential status through
`service_summaries` (`gitea_mcp_server.py:19574`).
2. **The return path is a flow too.** Content read from Gitea travels back into the model
and is treated as instruction. This is the ADV2 edge, and it is the only edge in the
diagram with no authentication on it, because it is not a request.
@@ -160,7 +156,7 @@ an attacker holding a token calls the API, not our tools.
| CR10 | MDCPS GlitchTip read credential | macOS keychain, read from the Gitea server process (`gitea_config.py:851`) | B7 | Read error events and their payloads (A5). Enabled today. |
| CR11 | `SENTRY_AUTH_TOKEN` | Process environment, read in-process (`sentry_incident_bridge.py:36`, `sentry_incident_bridge.py:190`), sent as a bearer header (`sentry_incident_bridge.py:289`) | B7 | Read and reconcile Sentry issues (A5). Not a keychain credential — an env var, so it is inherited by anything the process spawns. |
| CR12 | `SENTRY_DSN` | Process environment (`sentry_observability.py:55`) | B7 | Write events into the observability project. Low read value, real forgery value: an attacker can inject fabricated events into the record (A10). |
| CR13 | macOS login keychain access | The operator's login session; gated by `assert_keychain_access_allowed` (`mcp_daemon_guard.py:583`) | B3, ADV5 | **Every other credential in this table except CR11 and CR12.** This is the aggregation point. |
| CR13 | macOS login keychain access | The operator's login session; gated by `assert_keychain_access_allowed` (`mcp_daemon_guard.py:440`) | B3, ADV5 | **Every other credential in this table except CR11 and CR12.** This is the aggregation point. |
| CR14 | Coordination-store access (no secret) | Filesystem permissions — `control_plane_db.py:47`, created `0o700` (`control_plane_db.py:380`), opened with a local file lock (`control_plane_db.py:386`) | B6, ADV5 | Full read/write of locks, leases, and decision records (A6). **There is no credential here at all** — anything running as the operator can rewrite ownership. |
### Findings
@@ -180,16 +176,16 @@ the credential, and an attacker holding the token does not call our tools.
**Finding 3 — Any one role process can resolve every other role's credential.** This is not
inferred; it is demonstrated by tool output. `gitea_list_profiles`
(`gitea_mcp_server.py:19487`) called from the **author** session reports
(`gitea_mcp_server.py:19258`) called from the **author** session reports
`identity_status: "credentials present"` for `prgs-merger`, `prgs-reviewer`,
`prgs-reconciler`, and every `mdcps` profile, because it calls `resolve_token` on each one
(`gitea_mcp_server.py:19538`). The author process does not merely *have access to* the
(`gitea_mcp_server.py:19309`). The author process does not merely *have access to* the
merger's credential — it reads it to answer a status query. B2 is not a credential boundary
in either direction.
**Finding 4 — The Gitea server reads CI and observability secrets.** `gitea_audit_config`
(`gitea_mcp_server.py:19851`) reports `MDCPS Jenkins: enabled, read-only, authenticated`.
That word `authenticated` is produced by `service_summaries` (`gitea_mcp_server.py:19873`,
(`gitea_mcp_server.py:19552`) reports `MDCPS Jenkins: enabled, read-only, authenticated`.
That word `authenticated` is produced by `service_summaries` (`gitea_mcp_server.py:19574`,
defined at `gitea_config.py:837`), whose default check calls `_keychain_token` on the
service's own keychain reference (`gitea_config.py:851`). Producing that one line requires
the Gitea MCP server to read the Jenkins secret and the GlitchTip secret out of the
@@ -197,8 +193,8 @@ keychain. B7 does not exist.
**Finding 5 — Jenkins and GlitchTip are already decomposed; the reach is residual.** Their
tools live in separately registered servers, marked `external-mcp`
(`gitea_mcp_server.py:17909`, `gitea_mcp_server.py:17915`, `gitea_mcp_server.py:17936`,
`gitea_mcp_server.py:17941`) with their own expected tool sets (`mcp_discoverability.py:9`,
(`gitea_mcp_server.py:17707`, `gitea_mcp_server.py:17713`, `gitea_mcp_server.py:17734`,
`gitea_mcp_server.py:17739`) with their own expected tool sets (`mcp_discoverability.py:9`,
`mcp_discoverability.py:17`). The correct decomposition was already chosen. What remains is
a leak across it: the credential *references* still live in the Gitea configuration and are
still resolved by the Gitea process. #75 bundled these services into one control-plane
@@ -209,8 +205,8 @@ GlitchTip, the Sentry bridge runs *inside* the Gitea server, resolving its token
process environment (`sentry_incident_bridge.py:190`) and sending it as a bearer header
(`sentry_incident_bridge.py:289`). Being an environment variable rather than a keychain item
makes it strictly worse: it needs no keychain prompt and is inherited by every subprocess the
server spawns — including the `ps` invocations at `gitea_mcp_server.py:21761` and
`gitea_mcp_server.py:21805`, reached from `gitea_mcp_server.py:21741`.
server spawns — including the `ps` invocations at `gitea_mcp_server.py:21462` and
`gitea_mcp_server.py:21506`, reached from `gitea_mcp_server.py:21442`.
**Finding 7 — The highest-value coordination asset has the weakest gate.** A6 is protected
by filesystem permissions alone (CR14). Corrupting a lease requires no Gitea credential,
@@ -219,15 +215,13 @@ assumes. Every other asset costs an attacker a credential; this one costs nothin
local access, which is exactly ADV5's position.
**Finding 8 — Provenance authenticates the launch, not the caller.** `server_provenance` is
reported as exactly `client_managed` or `manual_launch` (`gitea_mcp_server.py:19217`),
derived from environment inspection (`gitea_mcp_server.py:15630`) with the recognized-key
reported as exactly `client_managed` or `manual_launch` (`gitea_mcp_server.py:19001`),
derived from environment inspection (`gitea_mcp_server.py:15412`) with the recognized-key
allowlist at `gitea_config.py:1172` and the generator that emits the marker at
`gitea_config.py:1233`. Every one of those facts is fixed at process start. A client that is
trustworthy at launch and compromised a minute later remains `client_managed` for the life
of the process, and the transport contract that underwrites it is stated as a property of
the server itself (`mcp_server.py:4`). Since #931 that contract names the configured
transport rather than asserting stdio, but it is still fixed once, at bind, for the life of
the process.
of the process, and the stdio contract that underwrites it is stated as a property of the
server itself (`mcp_server.py:4`).
## 6. Decomposition ruling
@@ -259,7 +253,7 @@ holds the token and calls the API instead of the tool.
**D3 — Credential resolution is scoped to the request principal.** A session must resolve its
own credential and must have no path to any other principal's. The resolve-every-profile
behavior behind `gitea_mcp_server.py:19538` and `gitea_mcp_server.py:19873` must report
behavior behind `gitea_mcp_server.py:19309` and `gitea_mcp_server.py:19574` must report
configured-or-not from configuration alone, without resolving the secret.
*Rationale.* Finding 3. An audit surface that proves a credential exists by fetching it is a
@@ -336,11 +330,11 @@ The client is attached to the local fleet over stdio.
| Boundary | What ADV1 reaches | Stopped by |
| -------- | ----------------- | ---------- |
| B1 | Everything the fleet serves. The client *is* the sanctioned launcher: it satisfies the client-managed check (`gitea_mcp_server.py:15630`) by construction, and provenance is never re-verified after launch (Finding 8). | Nothing. The guard authenticates the launch, not the caller. |
| B1 | Everything the fleet serves. The client *is* the sanctioned launcher: it satisfies the client-managed check (`gitea_mcp_server.py:15412`) by construction, and provenance is never re-verified after launch (Finding 8). | Nothing. The guard authenticates the launch, not the caller. |
| B2 | All five roles — it is attached to all five namespaces. It can author a PR, approve it from the reviewer namespace, and merge it from the merger namespace. | Only the in-process self-review check, which compares `jcwalker3` (author) against `sysadmin` (reviewer) and **passes**, because Finding 1 made them different accounts while leaving reviewer and merger identical. A9 falls in one sequence of legitimate calls. |
| B3 | Every credential in CR1CR10 via CR13, with no additional prompt — the daemon is already sanctioned, so `assert_keychain_access_allowed` (`mcp_daemon_guard.py:583`) returns immediately. | Nothing. |
| B3 | Every credential in CR1CR10 via CR13, with no additional prompt — the daemon is already sanctioned, so `assert_keychain_access_allowed` (`mcp_daemon_guard.py:440`) returns immediately. | Nothing. |
| B4 | A1 and A2 in full. | Branch protection at the forge, to the extent configured. |
| B5 | The operator's checkout and every worktree, through the author tools (`gitea_mcp_server.py:10957`), plus the shared stderr path at `mcp_server.py:13`. | Nothing; the server's disk is the target disk. |
| B5 | The operator's checkout and every worktree, through the author tools (`gitea_mcp_server.py:10894`), plus the shared stderr path at `mcp_server.py:10`. | Nothing; the server's disk is the target disk. |
| B6 | All coordination state — no credential required (CR14). It can forge lease ownership and clear decision locks. | Filesystem permissions, which it already satisfies. |
| B7 | Jenkins (A4) and GlitchTip (A5) secrets via Finding 4, and CR11/CR12 from its own environment. | Nothing. |
| B8 | Both tenants. | Nothing in-process; only the disabled-context check (`gitea_config.py:704`), which does not apply to enabled contexts. |
+99 -10
View File
@@ -27,13 +27,73 @@ ALLOWED = "allowed"
BLOCKED = "blocked"
FAILED = "failed"
SUCCEEDED = "succeeded"
REQUESTED = "requested"
ACCEPTED = "accepted"
PENDING = "pending"
REDACTED = "[REDACTED]"
# A dict key containing any of these (case-insensitive) has its value redacted.
_SECRET_KEY_HINTS = ("token", "password", "secret", "authorization", "auth")
_SECRET_KEY_HINTS = (
"token",
"password",
"passwd",
"pwd",
"secret",
"authorization",
"auth",
"api_key",
"apikey",
"access_key",
"private_key",
"client_secret",
"credential",
)
# A string value starting with one of these has the following run redacted.
_SECRET_VALUE_PREFIXES = ("token ", "Basic ", "Bearer ")
# Space-terminated scheme prefixes only. Colon forms (``token:`` / ``api_key:``)
# and ``Authorization: Bearer …`` are handled by ``_ASSIGNMENT_SECRET_PATTERN``
# so policy/docs text that merely *names* a scheme is not itself flagged as a
# live secret by console detectors.
_SECRET_VALUE_PREFIXES = (
"token ",
"Basic ",
"Bearer ",
)
# Bare token-shaped values only — never a broad ``sec-`` prefix that erases
# ordinary words (#664 B8 over-redaction).
_BARE_SECRET_PATTERN = re.compile(
r'(?i)\b(?:'
r'ghp_[A-Za-z0-9_]{16,}'
r'|gho_[A-Za-z0-9_]{16,}'
r'|ghu_[A-Za-z0-9_]{16,}'
r'|ghs_[A-Za-z0-9_]{16,}'
r'|ghr_[A-Za-z0-9_]{16,}'
r'|sk-live-[A-Za-z0-9_-]{16,}'
r'|sk-proj-[A-Za-z0-9_-]{16,}'
r'|sk-[A-Za-z0-9_-]{20,}'
r'|glpat-[A-Za-z0-9_-]{16,}'
r')\b'
)
# Key/value credentials embedded in free text (password=..., api_key: ..., GITEA_TOKEN=..., etc.).
# Group 1 captures the key name (e.g. GITEA_TOKEN, password, api_key).
# Group 2 captures delimiter/whitespace (=, : ).
# Group 3 captures the secret value, stopping at whitespace or non-secret delimiters (&, ;, ,, quotes, closing brackets).
_ASSIGNMENT_SECRET_PATTERN = re.compile(
r'(?i)\b([A-Za-z0-9_]*?(?:token|password|passwd|pwd|secret|api[_-]?key|access[_-]?key|'
r'client[_-]?secret|private[_-]?key|authorization|credential))\b(\s*[:=]\s*)('
r'"[^"]*"|\'[^\']*\'|'
r'(?:Bearer|Basic|Token)\s+[^\s;&,"\'\)\}\]\>]+|'
r'[^\s;&,"\'\)\}\]\>]+'
r')'
)
# Connection-string style credentials: Password=...; User ID=...; etc.
_CONN_STRING_SECRET_PATTERN = re.compile(
r'(?i)\b((?:password|pwd|user\s*id|uid|username|account)\s*=\s*)([^\s;\'"]+)'
)
# Known synthetic test-only domains/hostnames to preserve
_SYNTHETIC_HOSTS = {
@@ -59,7 +119,8 @@ def redact_urls(text: str) -> str:
if not isinstance(text, str) or not text:
return text
url_pattern = re.compile(r'(https?://[^\s)>\]}]+)', re.IGNORECASE)
# Match any URI scheme (http, https, postgres, mysql, mongodb, redis, etc.)
url_pattern = re.compile(r'([a-z0-9\+\.\-]+://[^\s)>\]}]+)', re.IGNORECASE)
def replace_url(match):
url_str = match.group(1)
@@ -73,11 +134,11 @@ def redact_urls(text: str) -> str:
is_synthetic = True
break
if is_synthetic:
# Rebuild synthetic URL to redact any credentials or query secrets
if is_synthetic or (parsed.username or parsed.password) or parsed.scheme.lower() not in ("http", "https"):
# Rebuild URL to redact any credentials or query secrets
new_netloc = parsed.netloc
if parsed.username or parsed.password:
netloc_clean = parsed.hostname
netloc_clean = parsed.hostname or ""
if parsed.port:
netloc_clean = f"{netloc_clean}:{parsed.port}"
new_netloc = f"[REDACTED_USER]:[REDACTED_PASS]@{netloc_clean}"
@@ -114,23 +175,51 @@ def redact_urls(text: str) -> str:
return out
def _mask_assignment(match: re.Match) -> str:
"""Keep the key and separator; replace only the secret value."""
val = match.group(3)
if val.startswith(REDACTED) or val.startswith("%5BREDACTED") or val.startswith("[REDACTED"):
return f"{match.group(1)}{match.group(2)}{val}"
return f"{match.group(1)}{match.group(2)}{REDACTED}"
def _mask_conn_secret(match: re.Match) -> str:
"""Keep the connection-string key; replace only the credential value."""
val = match.group(2)
if val.startswith(REDACTED) or val.startswith("%5BREDACTED") or val.startswith("[REDACTED"):
return f"{match.group(1)}{val}"
return f"{match.group(1)}{REDACTED}"
def _redact_str(text):
"""Redact anything that looks like an Authorization credential or raw URL in *text*."""
"""Redact credentials, bare token shapes, and raw URLs in *text* (#664 B8).
Covers key/value credentials, authorization/bearer material, bare
token-shaped values, connection-string credentials, and secrets embedded
in larger sentences. Deliberately does **not** erase ordinary words that
merely begin with a broad ``sec-`` prefix.
"""
if not isinstance(text, str) or not text:
return text
out = text
out = redact_urls(text)
out = _BARE_SECRET_PATTERN.sub(REDACTED, out)
out = _ASSIGNMENT_SECRET_PATTERN.sub(_mask_assignment, out)
out = _CONN_STRING_SECRET_PATTERN.sub(_mask_conn_secret, out)
out_lower = out.lower()
for prefix in _SECRET_VALUE_PREFIXES:
prefix_lower = prefix.lower()
idx = 0
while True:
i = out.find(prefix, idx)
i = out_lower.find(prefix_lower, idx)
if i == -1:
break
j = i + len(prefix)
while j < len(out) and not out[j].isspace():
j += 1
out = out[:i] + prefix + REDACTED + out[j:]
out_lower = out.lower()
idx = i + len(prefix) + len(REDACTED)
return redact_urls(out)
return out
def redact(value):
+32 -3
View File
@@ -97,6 +97,26 @@ GITEA_OPERATION_ALIASES = {
_REVIEW_MERGE_OPS = frozenset({"gitea.pr.approve", "gitea.pr.merge"})
_AUTHOR_ONLY_OPS = frozenset({"gitea.pr.create", "gitea.branch.push"})
# First-class operation services that may appear in multi-service profile
# allowlists. ``runtime.*`` is the control-plane capability namespace used by
# non-Gitea MCP tools such as ``runtime.break_glass_restart`` (#664 B13).
# Unknown foreign prefixes (e.g. ``jenkins.*`` under service=gitea) still fail
# closed — they are not registered here.
KNOWN_OPERATION_SERVICES = frozenset({"gitea", "runtime"})
def service_for_operation(op, default="gitea"):
"""Return the registered service prefix for a fully-qualified *op*.
Unqualified names and unknown prefixes fall back to *default* so callers
keep the historical Gitea-centric gate behaviour.
"""
if isinstance(op, str) and "." in op:
prefix = op.split(".", 1)[0]
if prefix in KNOWN_OPERATION_SERVICES:
return prefix
return default
def normalize_operation(op, service="gitea"):
"""Return the canonical namespaced name for *op*, or fail closed (#106).
@@ -133,6 +153,12 @@ def check_operation(op, allowed, forbidden=(), service="gitea"):
Reasons: ``allowed``, ``invalid-operation``, ``invalid-forbidden-entry``,
``forbidden``, ``no-allowed-operations``, ``not-allowed``.
Multi-service profile allowlists (#664 B13): each allow/forbid entry is
normalized with its own registered service prefix (``gitea.*`` or
``runtime.*``) so a gate defaulting to service=gitea can still enforce an
exact ``runtime.break_glass_restart`` grant. Unknown / misspelled
operations and foreign service prefixes remain fail-closed.
Fail-closed rules:
- an *op* that cannot be normalized is denied (``invalid-operation``)
- a forbidden entry that cannot be normalized denies the request
@@ -143,14 +169,16 @@ def check_operation(op, allowed, forbidden=(), service="gitea"):
- ``forbidden`` always overrides ``allowed``
- an empty or missing allowed list denies everything
"""
op_service = service_for_operation(op, default=service)
try:
op_n = normalize_operation(op, service)
op_n = normalize_operation(op, op_service)
except ConfigError:
return (False, "invalid-operation")
forbidden_n = set()
for entry in (forbidden or ()):
try:
forbidden_n.add(normalize_operation(entry, service))
entry_service = service_for_operation(entry, default=service)
forbidden_n.add(normalize_operation(entry, entry_service))
except ConfigError:
return (False, "invalid-forbidden-entry")
if op_n in forbidden_n:
@@ -160,7 +188,8 @@ def check_operation(op, allowed, forbidden=(), service="gitea"):
allowed_n = set()
for entry in allowed:
try:
allowed_n.add(normalize_operation(entry, service))
entry_service = service_for_operation(entry, default=service)
allowed_n.add(normalize_operation(entry, entry_service))
except ConfigError:
continue
if op_n in allowed_n:
+727 -549
View File
File diff suppressed because it is too large Load Diff
-6
View File
@@ -500,11 +500,6 @@ def assess_transport_for_auth_mint() -> dict[str, Any]:
native = mcp_daemon_guard.is_native_mcp_transport()
pytest = mcp_daemon_guard.is_pytest_runtime()
production = mcp_daemon_guard.is_production_native_mcp_transport()
# #931: report which transport underwrites the verdict, read through the
# one shared accessor rather than assumed to be stdio. The gate's decision
# is unchanged here; naming the transport is what lets #932 re-derive the
# guarantee from an authenticated session instead of from the bind.
bound = mcp_daemon_guard.bound_transport()
if not native and not pytest:
reasons.append(
"irrecoverable provenance authorization requires production native "
@@ -515,7 +510,6 @@ def assess_transport_for_auth_mint() -> dict[str, Any]:
"allowed": not reasons,
"native_mcp_transport": native,
"production_native_mcp_transport": production,
"bound_transport": bound,
"pytest": pytest,
"reasons": reasons,
}
+9 -21
View File
@@ -92,15 +92,7 @@ OPERATOR_UI_STEPS: dict[str, tuple[str, ...]] = {
),
}
#: What an *unidentified* client gets. #948: this is deliberately the
#: host-agnostic step set rather than a specific product. Defaulting to one
#: vendor emitted Codex UI steps to a Gemini/Antigravity operator, who then had
#: no reachable recovery path — the guidance named a panel they do not have.
DEFAULT_CLIENT = "generic"
#: The historical default, kept addressable by name so Codex callers still get
#: Codex steps, without it silently becoming the fallback for unknown clients.
LEGACY_DEFAULT_CLIENT = "codex"
DEFAULT_CLIENT = "codex"
def normalize_reason(reason: str | None) -> str:
@@ -136,18 +128,14 @@ def normalize_reason(reason: str | None) -> str:
def normalize_client(client: str | None) -> str:
"""Return the UI-step key for a client.
#948: alias resolution is shared with ``mcp_worker_identity`` so a client
name means the same thing wherever it is read. A name we recognise but have
no bespoke steps for Gemini, Antigravity, Grok resolves to the generic
host-agnostic steps rather than to another vendor's panel.
"""
import mcp_worker_identity
canonical = mcp_worker_identity.normalize_client_name(client)
if canonical in OPERATOR_UI_STEPS:
return canonical
"""Return a known client key for operator UI steps."""
text = (client or "").strip().lower().replace(" ", "_").replace("-", "_")
if text in ("codex", "openai_codex", "openai"):
return "codex"
if text in ("claude", "claude_code", "claude_desktop", "anthropic"):
return "claude_code"
if text in OPERATOR_UI_STEPS:
return text
return DEFAULT_CLIENT
+13 -176
View File
@@ -32,8 +32,6 @@ import time
from pathlib import Path
from typing import Any
import mcp_transport_config
SANCTIONED_DAEMON_ENV = "GITEA_MCP_SANCTIONED_DAEMON"
ALLOW_DIRECT_IMPORT_ENV = "GITEA_ALLOW_DIRECT_MCP_IMPORT"
ALLOW_KEYCHAIN_CLI_ENV = "GITEA_ALLOW_KEYCHAIN_CLI"
@@ -44,9 +42,7 @@ FORCE_PROVENANCE_FAIL_ENV = "GITEA_TEST_FORCE_UNSANCTIONED"
_NATIVE_RUNTIME: dict[str, Any] | None = None
# Production transport identifiers accepted by bind_native_mcp_transport.
# #931: the permitted set is defined once, in mcp_transport_config. This name
# is kept as an alias so the guard never restates a transport identifier.
_PRODUCTION_TRANSPORTS = mcp_transport_config.SUPPORTED_TRANSPORTS
_PRODUCTION_TRANSPORTS = frozenset({"stdio"})
_RUNTIME_MODE_PRODUCTION = "production"
_RUNTIME_MODE_TEST = "test"
_PHASE_ENTRYPOINT_CLAIMED = "entrypoint_claimed"
@@ -62,23 +58,6 @@ class UnsanctionedRuntimeError(RuntimeError):
"""Raised when mutation/credential code runs outside a native MCP daemon."""
class TransportExecutionError(UnsanctionedRuntimeError):
"""Raised when a bound transport may not be served by this entrypoint (#931).
Subclasses :class:`UnsanctionedRuntimeError` so every existing fail-closed
handler still catches it, while letting a caller that cares distinguish
"nothing is bound" from "something valid is bound but its listener has not
been commissioned". Carries the structured verdict on ``.assessment``.
"""
def __init__(self, message: str, assessment: dict[str, Any] | None = None):
super().__init__(message)
self.assessment = assessment or {}
self.blocker_kind = self.assessment.get("blocker_kind")
self.owner_issue = self.assessment.get("owner_issue")
self.transport = self.assessment.get("transport")
def is_pytest_runtime() -> bool:
if (os.environ.get(FORCE_PROVENANCE_FAIL_ENV) or "").strip() in {
"1",
@@ -192,46 +171,23 @@ def mark_sanctioned_daemon() -> dict[str, Any]:
return native_runtime_status()
def bind_native_mcp_transport(*, transport: str | None = None) -> dict[str, Any]:
"""Bind the live native MCP transport lifecycle (#695 / #931).
def bind_native_mcp_transport(*, transport: str) -> dict[str, Any]:
"""Bind the live native MCP transport lifecycle (#695).
Must be called from the resolved canonical entrypoint immediately before
the real MCP server transport loop (``mcp.run``). Requires a prior
successful :func:`mark_sanctioned_daemon` claim in this process.
Import-only or offline launch without this bind leaves
the real MCP server transport loop (e.g. ``mcp.run(transport=\"stdio\")``).
Requires a prior successful :func:`mark_sanctioned_daemon` claim in this
process. Import-only or offline launch without this bind leaves
:func:`is_native_mcp_transport` false.
#931: ``transport`` is now optional. Omitting it — which is what the
production entrypoint does resolves the identifier from deployment
configuration via :func:`mcp_transport_config.resolve_configured_transport`,
yielding :data:`mcp_transport_config.DEFAULT_TRANSPORT` when nothing is
configured. An explicit argument remains supported for tests and for a
launcher that has already resolved the value. Either way the identifier is
validated against the single permitted set before the runtime record is
written, so no tool can dispatch over an unregistered transport.
The resolved value is pinned into the process-local record and is read back
only through :func:`bound_transport`. Rebinding to a different transport is
refused, so two guards can never observe different values in one process.
"""
global _NATIVE_RUNTIME
if transport is None:
resolution = mcp_transport_config.resolve_configured_transport()
transport_name = str(resolution["transport"])
if not resolution["supported"]:
raise UnsanctionedRuntimeError(
"bind_native_mcp_transport rejected: "
+ "; ".join(resolution["reasons"])
+ " No tool is served over an unregistered transport."
)
else:
transport_name = mcp_transport_config.normalize_transport(transport)
if transport_name not in _PRODUCTION_TRANSPORTS:
raise UnsanctionedRuntimeError(
f"bind_native_mcp_transport rejected: transport {transport!r} is "
f"not a production MCP transport (#695). Allowed: "
f"{sorted(_PRODUCTION_TRANSPORTS)}."
)
transport_name = (transport or "").strip().lower()
if transport_name not in _PRODUCTION_TRANSPORTS:
raise UnsanctionedRuntimeError(
f"bind_native_mcp_transport rejected: transport {transport!r} is "
f"not a production MCP transport (#695). Allowed: "
f"{sorted(_PRODUCTION_TRANSPORTS)}."
)
entrypoint_path = _caller_official_entrypoint_path()
if entrypoint_path is None:
@@ -260,23 +216,6 @@ def bind_native_mcp_transport(*, transport: str | None = None) -> dict[str, Any]
"between mark and bind (#695)."
)
# #931: one process binds one transport. Re-binding the same identifier is
# idempotent (a retried launch step must not fail); re-binding a different
# one is refused, because a guard that already read the first value would
# otherwise disagree with a guard that reads the second.
already_bound = (_NATIVE_RUNTIME.get("transport") or "").strip()
if (
already_bound
and _NATIVE_RUNTIME.get("phase") == _PHASE_TRANSPORT_BOUND
and already_bound != transport_name
):
raise UnsanctionedRuntimeError(
"bind_native_mcp_transport rejected: transport is already bound to "
f"{already_bound!r} in this process; rebinding to "
f"{transport_name!r} is forbidden (#931). Restart the server to "
"change the deployment transport."
)
# Pin session-state root for this server lifetime (#695 AC2 / PR #701).
# Changing GITEA_MCP_SESSION_STATE_DIR after bind must not manufacture a
# second authority domain for decision locks / workflow proofs.
@@ -414,88 +353,6 @@ def is_production_native_mcp_transport() -> bool:
return (_NATIVE_RUNTIME or {}).get("mode") == _RUNTIME_MODE_PRODUCTION
def bound_transport() -> str | None:
"""The one authoritative bound transport identifier, or ``None`` (#931).
This is the shared accessor every transport-aware guard reads. It reports
the value pinned at bind time, never the environment, so changing
``GITEA_MCP_TRANSPORT`` after the bind cannot move what a guard observes
the same rule :func:`pinned_session_state_dir` applies to session state.
``None`` means unbound: an offline import or a launch that never reached
the bind. Callers must treat that as fail-closed, exactly as they already
treat :func:`is_native_mcp_transport` returning false.
"""
if not is_native_mcp_transport():
return None
return (_NATIVE_RUNTIME or {}).get("transport") or None
def assert_transport_bound(context: str = "tool service") -> str:
"""Return the bound transport, or fail closed before *context* (#931).
Called immediately before the server enters its transport loop so an
invalid or absent bind stops the process rather than serving tools over a
transport no guard can name.
"""
transport = bound_transport()
if transport:
return transport
raise UnsanctionedRuntimeError(
f"No MCP transport is bound; refusing {context} (#931). "
"bind_native_mcp_transport must succeed from the canonical entrypoint "
"before any tool is served. Offline import and standalone launch "
"cannot reconstruct a bind."
)
def assess_serve_authorization() -> dict[str, Any]:
"""Structured verdict on whether this process may serve tools (#931).
This is the decision that consumes :func:`bound_transport`. It is what stops
the bound identifier from being reporting-only metadata: the serve path
cannot proceed unless the value pinned at bind is one this entrypoint is
commissioned to execute.
Never raises; returns the verdict so callers and diagnostics can inspect it.
"""
return mcp_transport_config.assess_transport_execution(bound_transport())
def authorize_transport_execution(context: str = "tool service") -> str:
"""Return the transport this process may serve, or fail closed (#931).
Two distinct boundaries, in order:
1. **Bind presence** :func:`assert_transport_bound` enforces the
pre-existing #695 contract, so an unbound runtime keeps its established
failure and reason code.
2. **Execution authorization** the bound identifier must be one this
entrypoint is commissioned to serve. A registered transport whose
listener has not been commissioned is refused here, before any listener
is created and before any tool can dispatch.
That ordering matters: recognition, validation and durable recording all
still happen for a remote identifier, so #931's seam is intact; only the act
of *serving* it is withheld until its owning issue commissions it.
"""
# Boundary 1: unbound stays exactly as fail-closed as it was under #695.
assert_transport_bound(context)
# Boundary 2: bound, but is this entrypoint allowed to serve it?
assessment = assess_serve_authorization()
if assessment.get("allowed"):
return str(assessment["transport"])
reasons = "; ".join(assessment.get("reasons") or []) or "not authorized"
next_action = assessment.get("exact_next_action") or ""
raise TransportExecutionError(
f"Refusing {context} (#931) [{assessment.get('blocker_kind')}]: "
f"{reasons} {next_action}".strip(),
assessment,
)
def is_sanctioned_mcp_daemon() -> bool:
"""Backward-compatible name; #695 requires native transport, not env alone."""
if is_production_native_mcp_transport():
@@ -619,21 +476,6 @@ def native_runtime_status() -> dict[str, Any]:
"entrypoint_path": rt.get("entrypoint_path"),
"phase": rt.get("phase"),
"transport": rt.get("transport"),
# #931: the authoritative bound identifier, plus the seam that defines
# what may be bound. ``bound_transport`` is None until a bind succeeds,
# so an offline import is distinguishable from a stdio session.
"bound_transport": bound_transport(),
"transport_bound": bound_transport() is not None,
"default_transport": mcp_transport_config.DEFAULT_TRANSPORT,
"supported_transports": list(mcp_transport_config.supported_transports()),
"transport_env": mcp_transport_config.TRANSPORT_ENV,
# #931 review 635: recognition and execution authorization are distinct.
# ``supported`` is what may be bound; ``executable`` is what this
# entrypoint may actually serve. A recognized-but-uncommissioned
# transport reports serve_authorized False with a named blocker.
"executable_transports": list(mcp_transport_config.executable_transports()),
"serve_authorized": bool(assess_serve_authorization().get("allowed")),
"serve_authorization": assess_serve_authorization(),
"mode": rt.get("mode"),
"session_state_dir": pinned_session_state_dir() or rt.get("session_state_dir"),
"session_state_dir_pinned": pinned_session_state_dir() is not None,
@@ -660,12 +502,7 @@ def mutation_provenance_fields() -> dict[str, Any]:
if st.get("mode") == _RUNTIME_MODE_TEST and st["native_mcp_transport"]:
transport = "test_native_mcp"
return {
# ``transport`` stays the trust *class* it has always been, so existing
# durable records keep their shape. ``bound_transport`` (#931) adds the
# bound identifier itself, which is what lets an operator tell from a
# durable record which transport performed a mutation.
"transport": transport,
"bound_transport": st.get("bound_transport"),
"native_mcp_transport": bool(st["native_mcp_transport"]),
"production_native_mcp_transport": bool(
st.get("production_native_mcp_transport")
+5 -397
View File
@@ -51,42 +51,6 @@ EOF_PATTERNS = (
"eof",
)
ERROR_CONNECTED_NAMESPACES_MISSING = "mcp_connected_namespaces_missing"
# Distinct from the condition above. ``mcp_connected_namespaces_missing`` is reserved for
# the actual #708 defect: the host *does* report the service Connected, yet the namespace
# never entered the active session tool surface. A required namespace that is absent from
# the connected-service inventory has no Connected claim behind it at all, so reporting it
# under the #708 condition would assert something the evidence does not support and would
# point an operator at the wrong recovery.
ERROR_REQUIRED_NAMESPACES_NOT_CONNECTED = "mcp_required_namespaces_not_connected"
DISCOVERY_STATUS_ATTACHED = "namespaces_attached"
DISCOVERY_STATUS_CONNECTED_MISSING = "connected_but_namespaces_missing"
DISCOVERY_STATUS_NOT_CONNECTED = "required_namespaces_not_connected"
DISCOVERY_STATUS_DISCONNECTED = "disconnected"
# Namespaces that must be *attached to the active session* for a mutation task (#708).
# Connected-at-host is not attached-in-session; these are gated separately from the
# #543 health map because a namespace can be healthy on probe yet absent from the
# session tool surface.
ATTACHMENT_GATED_TASKS = {
"review_pr": "gitea-reviewer",
"submit_review": "gitea-reviewer",
"merge_pr": "gitea-merger",
"work_issue": "gitea-author",
"create_pr": "gitea-author",
}
# The only sanctioned recovery for an unattached namespace (#678 exposes it natively).
SANCTIONED_ATTACH_RECOVERY_TOOL = "gitea_request_mcp_reconnect"
UNSAFE_FALLBACK_WARNING = (
"Workflow Safety Hard Stop (#708): Connected-but-namespaces-missing recovery must "
"NEVER use direct imports, Gitea API mutations, profile hopping, session-state "
"overrides, PID kills, or config mtime touches. Use client reconnect only."
)
SAFE_ENV_KEYS = (
"GITEA_MCP_PROFILE",
"GITEA_PROFILE_NAME",
@@ -95,318 +59,6 @@ SAFE_ENV_KEYS = (
"GITEA_MCP_CONFIG",
)
# #948: provenance used to be derived from the summary this allowlist produces.
# The allowlist never carried a provenance key, so that derivation could only
# ever evaluate to ``manual_launch`` — whatever the process actually was — while
# ``gitea_get_runtime_context`` read the live environment and reported
# ``client_managed`` for the same process. Provenance is no longer derived here.
# It comes from ``mcp_worker_identity.assess_provenance``, the single authority
# every surface shares. This allowlist keeps its original and only job: deciding
# which env values are safe to echo back in diagnostics.
def assess_connected_namespace_attachment(
*,
connected_servers: list[str] | tuple[str, ...] | set[str] | None = None,
attached_session_namespaces: list[str] | tuple[str, ...] | set[str] | None = None,
required_namespaces: list[str] | tuple[str, ...] | set[str] | None = None,
discovery_cache_age_seconds: float | int | None = None,
discovery_cache_hit: bool | None = None,
auto_attach_attempted: bool = False,
auto_attach_succeeded: bool = False,
session_tool_snapshot_at: float | int | None = None,
namespace_connected_at: dict[str, float] | None = None,
) -> dict[str, Any]:
"""Assess whether host-connected MCP servers have attached tool namespaces in the active session (#708).
Addresses the Connected-but-namespaces-missing defect: CLI/host status may report Connected
while the active LLM session tool surface exposes 0 attached tool namespaces.
This is a *distinct* condition from config drift (#672), transport-closed (#584),
and resolver EOF (#685): the transport is up and the host reports Connected, yet the
namespace never entered the session tool surface.
Startup ordering (``session_tool_snapshot_at`` + ``namespace_connected_at``) identifies
the race where the session tool snapshot was taken before a role server finished
``initialize``/``list_tools``, which is why parallel multi-role startup can leave the
session with an empty namespace set while Connected later flips true.
Returns structured detection details plus secret-free telemetry.
"""
connected = [str(s).strip() for s in (connected_servers or []) if str(s).strip()]
attached = set(str(ns).strip() for ns in (attached_session_namespaces or []) if str(ns).strip())
req = [str(r).strip() for r in (required_namespaces or DEFAULT_NAMESPACES) if str(r).strip()]
required = list(dict.fromkeys(req))
connected_set = set(connected)
proof: dict[str, dict[str, bool]] = {}
for s in connected:
proof[s] = {"connected": True, "attached": s in attached}
for r in required:
if r not in proof:
proof[r] = {"connected": r in connected_set, "attached": r in attached}
# The genuine #708 condition: the host reports the service Connected and the namespace
# still never entered the active session tool surface.
missing = [r for r in required if r in connected_set and r not in attached]
# A separate condition: the service is required but absent from the connected-service
# inventory. Nothing here is Connected, so this must not borrow the #708 wording or its
# recovery — see ERROR_REQUIRED_NAMESPACES_NOT_CONNECTED.
not_connected = [r for r in required if r not in connected_set]
# Evidence that disagrees with itself: reported attached to the session while absent
# from the connected inventory. Neither statement is proof, so it fails closed.
contradictory = [r for r in not_connected if r in attached]
# No required namespace may lack attachment proof and still be called healthy.
attachment_healthy = (
not missing
and not not_connected
and len(connected) > 0
and len(required) > 0
)
if attachment_healthy:
discovery_status = DISCOVERY_STATUS_ATTACHED
elif not connected:
discovery_status = DISCOVERY_STATUS_DISCONNECTED
elif not_connected:
discovery_status = DISCOVERY_STATUS_NOT_CONNECTED
else:
discovery_status = DISCOVERY_STATUS_CONNECTED_MISSING
# Every condition actually present is reported; ``error_type`` names the primary one.
# Not-connected outranks Connected-but-unattached because a service that never
# connected cannot be recovered by attaching its namespace.
error_types: list[str] = []
if not_connected:
error_types.append(ERROR_REQUIRED_NAMESPACES_NOT_CONNECTED)
if missing:
error_types.append(ERROR_CONNECTED_NAMESPACES_MISSING)
if not attachment_healthy and not error_types:
error_types.append(ERROR_REQUIRED_NAMESPACES_NOT_CONNECTED)
error_type = None if attachment_healthy else error_types[0]
# Per-namespace verdict, so a mutation gate never has to infer one namespace's state
# from a whole-session summary. Each entry states only what its own evidence supports.
namespace_conditions: dict[str, dict[str, Any]] = {}
for ns_name, state in proof.items():
ns_connected = bool(state["connected"])
ns_attached = bool(state["attached"])
if ns_connected and ns_attached:
condition = None
elif ns_connected:
condition = ERROR_CONNECTED_NAMESPACES_MISSING
else:
condition = ERROR_REQUIRED_NAMESPACES_NOT_CONNECTED
namespace_conditions[ns_name] = {
"namespace": ns_name,
"required": ns_name in required,
"connected": ns_connected,
"attached": ns_attached,
"attachment_healthy": ns_connected and ns_attached,
"condition": condition,
"contradictory_evidence": ns_attached and not ns_connected,
}
# Startup-ordering race: a namespace that finished connecting *after* the session
# tool snapshot was taken cannot be in that snapshot, however healthy it looks now.
connected_at = {
str(k).strip(): v
for k, v in (namespace_connected_at or {}).items()
if str(k).strip() and isinstance(v, (int, float))
}
late_attaching: list[str] = []
if isinstance(session_tool_snapshot_at, (int, float)):
for ns_name, ts in connected_at.items():
if ts > session_tool_snapshot_at and ns_name not in attached:
late_attaching.append(ns_name)
late_attaching.sort()
startup_ordering_race = bool(late_attaching)
auto_recovered = bool(auto_attach_attempted and auto_attach_succeeded and attachment_healthy)
reconnect_required = not attachment_healthy
reasons: list[str] = []
remediation: list[str] = []
if not connected:
reasons.append("No MCP servers reported Connected.")
remediation.append("Start or reconnect Gitea MCP servers in client config.")
if missing:
reasons.append(
f"MCP server(s) {missing} report Connected at host/CLI layer but tool namespaces "
f"are missing from active session attached tools (Connected ≠ attached tools, #708)."
)
remediation.append(
"Reconnect the IDE/client MCP session to attach namespaces to the active session "
f"(sanctioned path: {SANCTIONED_ATTACH_RECOVERY_TOOL}), then re-run full preflight "
"(gitea_whoami -> gitea_resolve_task_capability -> task). "
"Do not use direct imports, CLI API mutations, profile hopping, or session file overrides."
)
if not_connected:
reasons.append(
f"required MCP namespace(s) {not_connected} are absent from the connected-service "
"inventory: nothing reports them Connected, so there is no attachment to claim and "
"the Connected-but-unattached condition does not apply to them (#708)."
)
remediation.append(
f"Connect the required MCP server(s) {not_connected} through the client, then "
"reconnect the IDE/client MCP session so their namespaces attach "
f"(sanctioned path: {SANCTIONED_ATTACH_RECOVERY_TOOL}), and re-run full preflight. "
"Do not use direct imports, CLI API mutations, profile hopping, or session file overrides."
)
if contradictory:
reasons.append(
f"contradictory evidence for namespace(s) {contradictory}: reported attached to the "
"active session while absent from the connected-service inventory; neither statement "
"is proof, so attachment is treated as unproven (fail closed, #708)."
)
if attachment_healthy:
reasons.append(
"All required MCP server namespaces are connected and attached to the active session."
)
if startup_ordering_race:
reasons.append(
f"startup ordering race: namespace(s) {late_attaching} finished connecting after the "
"active session tool snapshot was taken, so they cannot appear in that snapshot (#708)."
)
if auto_attach_attempted and not auto_attach_succeeded:
reasons.append(
"automatic namespace attachment was attempted and did not succeed; only the sanctioned "
"client reconnect path remains."
)
if auto_recovered:
reasons.append("namespaces were automatically attached; no operator reconnect was required.")
return {
"success": attachment_healthy,
"attachment_healthy": attachment_healthy,
"discovery_status": discovery_status,
"connected_servers": connected,
"attached_session_namespaces": list(attached),
"missing_namespaces": missing,
"not_connected_namespaces": not_connected,
"contradictory_namespaces": contradictory,
"proof_of_connected_vs_attached": proof,
"namespace_conditions": namespace_conditions,
"error_type": error_type,
"error_types": error_types,
"reasons": reasons,
"remediation": remediation,
"exact_next_action": (
"None; session tool namespaces attached."
if attachment_healthy
else (
f"Connect the required MCP server(s) {not_connected} through the client, then "
"reconnect the IDE/client MCP session so their tool namespaces attach to the "
"active session. Do not use direct imports, CLI API mutations, profile hopping, "
"or session-state overrides."
if not_connected
else (
"Reconnect the IDE/client MCP session so tool namespaces attach to the active "
"session. Do not use direct imports, CLI API mutations, profile hopping, or "
"session-state overrides."
)
)
),
"unsafe_fallback_policy": UNSAFE_FALLBACK_WARNING,
"sanctioned_recovery_tool": SANCTIONED_ATTACH_RECOVERY_TOOL,
"reconnect_required": reconnect_required,
"auto_attach_attempted": bool(auto_attach_attempted),
"auto_recovered": auto_recovered,
"startup_ordering_race": startup_ordering_race,
"late_attaching_namespaces": late_attaching,
# Secret-free structured signals (#708 AC5). Namespace names and counts only:
# never tokens, endpoints, env values, or filesystem paths.
"telemetry": {
"connected_count": len(connected),
"attached_count": len(attached),
"required_count": len(required),
"missing_count": len(missing),
"not_connected_count": len(not_connected),
"contradictory_count": len(contradictory),
"required_attached_count": sum(1 for r in required if r in attached),
"discovery_status": discovery_status,
"discovery_cache_hit": (
None if discovery_cache_hit is None else bool(discovery_cache_hit)
),
"discovery_cache_age_seconds": (
float(discovery_cache_age_seconds)
if isinstance(discovery_cache_age_seconds, (int, float))
else None
),
"reconnect_required": reconnect_required,
"auto_attach_attempted": bool(auto_attach_attempted),
"auto_recovered": auto_recovered,
"startup_ordering_race": startup_ordering_race,
"error_type": error_type,
"error_types": error_types,
},
}
def required_namespace_for_attachment(task: str) -> str | None:
"""Map a mutation task to the MCP namespace that must be *attached* (#708)."""
return ATTACHMENT_GATED_TASKS.get((task or "").strip())
def attachment_gate_from_session(
task: str,
session_attachment: dict[str, dict[str, Any]] | None,
) -> list[str]:
"""Fail-closed gate on recorded connected-but-unattached namespaces (#708).
Mirrors :func:`mutation_gate_from_session`: a namespace that has not been
assessed yet does not gate, so this never blocks a session that simply has
not run the assessment. Once an assessment records the namespace required
for *task* as Connected-but-unattached, the mutation fails closed and the
only offered recovery is the sanctioned client reconnect path.
"""
ns = required_namespace_for_attachment(task)
if not ns:
return []
store = session_attachment or {}
entry = store.get(ns)
if not entry:
return []
if entry.get("attached") and entry.get("attachment_healthy"):
return []
what = task or "mutation"
connected = entry.get("connected")
detail = entry.get("condition") or entry.get("error_type")
if connected is True:
# Only here has anything actually reported the service Connected, so only here may
# the block say so.
detail = detail or ERROR_CONNECTED_NAMESPACES_MISSING
blocked = (
f"live MCP namespace '{ns}' is recorded {detail}: the host reports Connected but the "
f"namespace is not attached to the active session tool surface; reconnect the "
f"IDE/client MCP session and re-run preflight before {what} "
"(fail closed, #708)"
)
elif connected is False:
detail = ERROR_REQUIRED_NAMESPACES_NOT_CONNECTED
blocked = (
f"live MCP namespace '{ns}' is recorded {detail}: it is absent from the "
f"connected-service inventory, so it is neither connected nor attached and no "
f"Connected status is claimed for it; connect the required MCP server, then "
f"reconnect the IDE/client MCP session and re-run preflight before {what} "
"(fail closed, #708)"
)
else:
# Connected status was never recorded. Refuse without asserting either condition.
detail = detail or ERROR_CONNECTED_NAMESPACES_MISSING
blocked = (
f"live MCP namespace '{ns}' is recorded {detail} with no connected-status evidence, "
f"so attachment to the active session tool surface is unproven; reconnect the "
f"IDE/client MCP session and re-run preflight before {what} "
"(fail closed, #708)"
)
return [blocked, UNSAFE_FALLBACK_WARNING]
def _as_list(value: Any) -> list[str] | None:
if value is None:
@@ -452,23 +104,12 @@ def classify_namespace_probe(
profile: str | None = None,
configured: bool = True,
probe_source: str | None = None,
worker_identity: str | None = None,
generation_id: str | None = None,
registry: Any | None = None,
pid_alive_probe: Any | None = None,
) -> dict[str, Any]:
"""Classify whether a required tool is callable through a live namespace.
``registered_tools`` is static/server-side evidence. ``probe_result`` is
live invocation evidence. Only ``probe_source=client_namespace`` proves the
IDE-managed path; ``offline_spawn`` is an offline subprocess check only.
#948: ``worker_identity``/``generation_id``/``registry`` carry the
client/session ownership evidence. Provenance is resolved by
``mcp_worker_identity.assess_provenance`` the same call
``gitea_get_runtime_context`` makes so the two surfaces cannot report
different provenance for one process. Omitting them yields the fail-closed
``unproven`` verdict, never a fabricated ``client_managed``.
"""
ns = (namespace or "").strip()
tool = required_tool or REQUIRED_NAMESPACE_TOOLS.get(ns) or "gitea_whoami"
@@ -585,31 +226,14 @@ def classify_namespace_probe(
blocks = namespace_health_blocks_task("merge_pr", healthy)
import gitea_config
import mcp_worker_identity
raw_env = process.get("env") if isinstance(process, dict) else None
unconsumed_env = gitea_config.get_unconsumed_gitea_env_overrides(raw_env)
# #948: one authority, shared with gitea_get_runtime_context. The env is
# passed whole rather than through SAFE_ENV_KEYS — the allowlist exists to
# decide what may be *echoed*, and using it to decide what may be *believed*
# is what made this surface structurally unable to report client_managed.
# ``declared_only``: ``process`` describes an observed peer, not this
# interpreter. Its stdin is unavailable and its launcher-config env is
# inherited from whatever shell started it, so only an explicit declaration
# is evidence. Absence of one is ``unproven``, not an asserted manual launch.
provenance_verdict = mcp_worker_identity.assess_provenance(
registry=registry,
worker_identity=worker_identity,
generation_id=generation_id,
env=raw_env if isinstance(raw_env, dict) else {},
namespace=ns,
profile=profile_name,
pid_alive_probe=pid_alive_probe,
declared_only=True,
is_client_managed = bool(
env_summary.get("GITEA_CLIENT_MANAGED") in ("1", "true", "yes", "client_managed")
or env_summary.get("GITEA_MCP_CLIENT_MANAGED") in ("1", "true", "yes", "client_managed")
or env_summary.get("GITEA_SERVER_PROVENANCE") == "client_managed"
)
provenance = provenance_verdict["provenance"]
is_client_managed = provenance_verdict["is_client_managed"]
provenance = "client_managed" if is_client_managed else "manual_launch"
return {
"success": healthy,
@@ -628,15 +252,6 @@ def classify_namespace_probe(
"remediation": remediation,
"provenance": provenance,
"is_client_managed": is_client_managed,
# Every non-client-session verdict fails closed. Consumers that only
# need "may this mutate?" read this and stay correct across the #948
# vocabulary split between ``manual_launch`` and ``unproven``.
"provenance_fail_closed": provenance_verdict["fail_closed"],
"provenance_assessment": provenance_verdict,
"worker_identity": provenance_verdict["worker_identity"],
"session_id": provenance_verdict["session_id"],
"generation_id": provenance_verdict["generation_id"],
"client_name": provenance_verdict["client_name"],
"unconsumed_gitea_env": unconsumed_env,
"diagnostics": {
"namespace": ns,
@@ -648,13 +263,6 @@ def classify_namespace_probe(
"probe_source": source,
"provenance": provenance,
"is_client_managed": is_client_managed,
"provenance_fail_closed": provenance_verdict["fail_closed"],
"provenance_blocker_kind": provenance_verdict["blocker_kind"],
"provenance_scope": provenance_verdict["scope"],
"worker_identity": provenance_verdict["worker_identity"],
"session_id": provenance_verdict["session_id"],
"generation_id": provenance_verdict["generation_id"],
"client_name": provenance_verdict["client_name"],
"unconsumed_gitea_env": unconsumed_env,
},
"blocks_merge_workflow": blocks,
+3 -7
View File
@@ -1,10 +1,7 @@
#!/usr/bin/env python3
"""Gitea MCP Server — exposes Gitea operations as MCP tools.
The transport is selected by deployment configuration (GITEA_MCP_TRANSPORT) and
defaults to the local client-spawned transport when unset (#931); the permitted
set lives in mcp_transport_config. All tools authenticate via macOS keychain
(git credential fill).
Runs over stdio. All tools authenticate via macOS keychain (git credential fill).
"""
import os
import sys
@@ -46,9 +43,8 @@ check_conflict_markers()
# #558 / #695: claim the official entrypoint before loading mutation modules.
# This alone does NOT authorize mutations — gitea_mcp_server binds the live
# native MCP transport immediately before mcp.run, over the configured
# transport (#931). Import-only or offline launch without that bind fails
# closed on mutations.
# native MCP transport (stdio) immediately before mcp.run. Import-only or
# offline launch without that bind fails closed on mutations.
try:
import mcp_daemon_guard
-4
View File
@@ -588,13 +588,9 @@ def save_state(
bool(prov.get("production_native_mcp_transport")),
)
body.setdefault("transport", prov.get("transport"))
# #931: record the bound transport identifier itself, so a durable
# decision lock names which transport performed the mutation.
body.setdefault("bound_transport", prov.get("bound_transport"))
except Exception:
body.setdefault("native_mcp_transport", False)
body.setdefault("transport", "untrusted")
body.setdefault("bound_transport", None)
envelope = {
"kind": kind,
-292
View File
@@ -1,292 +0,0 @@
"""Single authoritative source for the bound MCP transport identifier (#931).
Before this module the transport was a literal, passed once at the bottom of
``gitea_mcp_server`` as ``bind_native_mcp_transport(transport="stdio")``. Every
guard that later asks "is this a trusted native session" resolves that question
through the value bound there, so the literal was effectively a constant in the
authorization chain rather than configuration.
This module is the seam. It owns three things and nothing else:
- the permitted set of transport identifiers,
- the default used when deployment configuration says nothing,
- the resolution of the configured value into a validated identifier.
It deliberately holds no state. The *bound* transport is pinned once, at bind
time, into the process-local native-runtime record owned by
:mod:`mcp_daemon_guard`, and is read back through
``mcp_daemon_guard.bound_transport()``. That split matters: configuration is
read exactly once, before any tool can dispatch, so a later environment change
cannot move the value a guard observes the same pinning rule already applied
to the session-state root under #695 AC2.
Nothing here consumes tool arguments, request bodies, or provenance fields. The
only input is the deployment environment, read at bind time.
Standing up a listener for a non-stdio transport is #938; this module only
makes the identifier expressible and validated.
"""
from __future__ import annotations
import os
from typing import Any, Mapping
# Deployment configuration key. Read once, at bind time, and never again.
TRANSPORT_ENV = "GITEA_MCP_TRANSPORT"
# The local, client-spawned transport. Unset configuration resolves to this,
# which is what keeps every existing stdio deployment byte-identical.
DEFAULT_TRANSPORT = "stdio"
# The sanctioned remote transport identifier. Accepting it here is what makes
# the bind pluggable; the endpoint that serves it belongs to #938. The name
# matches the MCP transport name so no second vocabulary has to be mapped.
REMOTE_TRANSPORT = "streamable-http"
# The permitted set. This is the only place transport identifiers are
# enumerated; guards consult it rather than restating any member.
#
# ``sse`` is a real MCP transport and is deliberately absent: it is the
# superseded remote transport, and admitting it would give the deployment two
# remote paths to reason about. An unregistered identifier must fail closed at
# bind time, and ``sse`` is held to that rule like any other.
SUPPORTED_TRANSPORTS = frozenset({DEFAULT_TRANSPORT, REMOTE_TRANSPORT})
# Recognition is not execution authorization (#931, review 635 B1/B2).
#
# SUPPORTED_TRANSPORTS answers "is this an identifier this system knows, and may
# it be bound, pinned and recorded?". It deliberately includes the remote
# identifier, because #931 requires the bind to become pluggable.
#
# EXECUTABLE_TRANSPORTS answers a strictly narrower question: "is this entrypoint
# commissioned to actually *serve* on that transport?". Only the local transport
# is. Handing ``streamable-http`` to ``mcp.run`` would start FastMCP's HTTP
# listener with no authentication, no TLS and no per-request principal — the
# endpoint #938 owns and gates. Recognition must therefore never imply execution.
#
# #938 commissions the remote listener by adding REMOTE_TRANSPORT here, together
# with the authentication and principal boundary its acceptance criteria require.
EXECUTABLE_TRANSPORTS = frozenset({DEFAULT_TRANSPORT})
# Which issue owns commissioning each recognized-but-not-executable transport.
# Used to make the refusal actionable rather than a generic denial.
TRANSPORT_EXECUTION_OWNER = {REMOTE_TRANSPORT: "#938"}
BLOCKER_TRANSPORT_NOT_BOUND = "transport_not_bound"
BLOCKER_TRANSPORT_NOT_RECOGNIZED = "transport_not_recognized"
BLOCKER_LISTENER_NOT_COMMISSIONED = "transport_listener_not_commissioned"
SOURCE_CONFIGURED = "deployment_configuration"
SOURCE_DEFAULT = "default"
class TransportConfigurationError(ValueError):
"""Raised when configured transport is outside :data:`SUPPORTED_TRANSPORTS`."""
def normalize_transport(value: Any) -> str:
"""Canonical form of a transport identifier; ``""`` when there is none.
Non-string values normalize to ``""`` rather than being coerced, so a
structured object smuggled in from a caller can never match a member of the
permitted set.
"""
if not isinstance(value, str):
return ""
return value.strip().lower()
def supported_transports() -> tuple[str, ...]:
"""Permitted identifiers, sorted, for messages and status payloads."""
return tuple(sorted(SUPPORTED_TRANSPORTS))
def is_supported_transport(value: Any) -> bool:
"""True when *value* normalizes to a member of the permitted set."""
return normalize_transport(value) in SUPPORTED_TRANSPORTS
def is_remote_transport(value: Any) -> bool:
"""True when *value* is a permitted transport that is not the local one."""
name = normalize_transport(value)
return name in SUPPORTED_TRANSPORTS and name != DEFAULT_TRANSPORT
def executable_transports() -> tuple[str, ...]:
"""Transports this entrypoint is commissioned to serve, sorted."""
return tuple(sorted(EXECUTABLE_TRANSPORTS))
def is_executable_transport(value: Any) -> bool:
"""True when *value* may actually be served by this entrypoint (#931).
Strictly narrower than :func:`is_supported_transport`. A recognized
identifier that is not executable is a correct, fully-bound configuration
whose listener has simply not been commissioned yet.
"""
return normalize_transport(value) in EXECUTABLE_TRANSPORTS
def assess_transport_execution(value: Any) -> dict[str, Any]:
"""Structured serve-authorization verdict for a bound transport (#931).
This is the decision that separates a *recognized* transport from one this
entrypoint may execute. It is deliberately a pure function of the bound
identifier so the serve path cannot reach a listener the deployment has not
commissioned.
Args:
value: The bound transport identifier, or ``None`` when unbound.
Returns:
dict with ``transport``, ``recognized``, ``executable``, ``allowed``,
``blocker_kind``, ``owner_issue``, ``reasons`` and
``exact_next_action``. ``allowed`` is true only for a bound, recognized,
commissioned transport.
"""
name = normalize_transport(value)
if not name:
return {
"transport": None,
"recognized": False,
"executable": False,
"allowed": False,
"blocker_kind": BLOCKER_TRANSPORT_NOT_BOUND,
"owner_issue": None,
"supported_transports": list(supported_transports()),
"executable_transports": list(executable_transports()),
"reasons": [
"no transport is bound; the serve path is fail-closed until "
"bind_native_mcp_transport succeeds (#695/#931)"
],
"exact_next_action": (
"Launch through the canonical entrypoint so "
"bind_native_mcp_transport runs before tool service."
),
}
recognized = name in SUPPORTED_TRANSPORTS
if not recognized:
return {
"transport": name,
"recognized": False,
"executable": False,
"allowed": False,
"blocker_kind": BLOCKER_TRANSPORT_NOT_RECOGNIZED,
"owner_issue": None,
"supported_transports": list(supported_transports()),
"executable_transports": list(executable_transports()),
"reasons": [
f"transport {name!r} is not a registered MCP transport (#931); "
"it should have been refused at bind time"
],
"exact_next_action": (
f"Set {TRANSPORT_ENV} to one of {list(supported_transports())}."
),
}
if name in EXECUTABLE_TRANSPORTS:
return {
"transport": name,
"recognized": True,
"executable": True,
"allowed": True,
"blocker_kind": None,
"owner_issue": None,
"supported_transports": list(supported_transports()),
"executable_transports": list(executable_transports()),
"reasons": [],
"exact_next_action": None,
}
owner = TRANSPORT_EXECUTION_OWNER.get(name)
owner_text = owner or "the issue that commissions this transport's listener"
return {
"transport": name,
"recognized": True,
"executable": False,
"allowed": False,
"blocker_kind": BLOCKER_LISTENER_NOT_COMMISSIONED,
"owner_issue": owner,
"supported_transports": list(supported_transports()),
"executable_transports": list(executable_transports()),
"reasons": [
f"transport {name!r} is registered and was bound and recorded, but "
f"this entrypoint is not commissioned to serve it (#931). Serving it "
f"would start a listener with no authentication, no transport "
f"security and no per-request principal; that endpoint is owned by "
f"{owner_text}."
],
"exact_next_action": (
f"Serve on {DEFAULT_TRANSPORT} until {owner_text} commissions the "
f"{name!r} listener with its authentication and principal boundary, "
f"which adds {name!r} to EXECUTABLE_TRANSPORTS."
),
}
def resolve_configured_transport(
env: Mapping[str, str] | None = None,
) -> dict[str, Any]:
"""Resolve the deployment-configured transport without raising.
Returns the resolution rather than a bare string so a caller can tell an
unset value (which legitimately yields :data:`DEFAULT_TRANSPORT`) from a
configured value that is not permitted (which must fail closed, never
silently degrade to the default).
Args:
env: Environment mapping to read; defaults to ``os.environ``.
Returns:
dict with ``transport`` (normalized; the default when unset),
``configured``, ``source``, ``raw``, ``supported``, and ``reasons``.
"""
source_env = os.environ if env is None else env
raw = source_env.get(TRANSPORT_ENV)
normalized = normalize_transport(raw)
configured = bool(normalized)
if not configured:
return {
"transport": DEFAULT_TRANSPORT,
"configured": False,
"source": SOURCE_DEFAULT,
"raw": raw,
"supported": True,
"supported_transports": list(supported_transports()),
"env_key": TRANSPORT_ENV,
"reasons": [],
}
supported = normalized in SUPPORTED_TRANSPORTS
reasons: list[str] = []
if not supported:
reasons.append(
f"{TRANSPORT_ENV}={normalized!r} is not a registered MCP transport "
f"(#931). Registered: {list(supported_transports())}."
)
return {
"transport": normalized,
"configured": True,
"source": SOURCE_CONFIGURED,
"raw": raw,
"supported": supported,
"supported_transports": list(supported_transports()),
"env_key": TRANSPORT_ENV,
"reasons": reasons,
}
def require_configured_transport(env: Mapping[str, str] | None = None) -> str:
"""Resolved transport identifier, or raise when it is not permitted.
Raises:
TransportConfigurationError: the configured identifier is unregistered.
"""
resolution = resolve_configured_transport(env)
if not resolution["supported"]:
raise TransportConfigurationError("; ".join(resolution["reasons"]))
return str(resolution["transport"])
File diff suppressed because it is too large Load Diff
-611
View File
@@ -1,611 +0,0 @@
"""Sanctioned reconciliation for worktree bindings whose paths are missing (#970).
When control-plane leases, session checkpoints, or issue locks hold a
``worktree_path`` pointing to a filesystem path that no longer exists on disk,
this module provides a fail-closed, auditable classification and resolution
workflow.
Core principles:
1. **Multi-dimensional Correlation**:
Correlates the binding with repository, host, branch, issue/PR, session,
and lease state.
2. **Safety-first Distinction**:
- Distinguishes a deleted worktree from a moved path, unavailable host/mount,
live lease, active session, or temporary filesystem problem.
- Host / mount failure (e.g. repo root or branches/ directory inaccessible)
blocks retirement (``unavailable_host_or_mount``).
- Moved worktree (found under another path in ``git worktree list``)
blocks retirement and reports the new location (``moved_worktree_path``).
- Live lease or active session/process blocks retirement (``live_lease_protected``,
``live_session_protected``).
3. **Atomic Pre-Mutation Re-Validation**:
Immediately before committing a mutation, re-evaluates filesystem existence,
git worktree list, host health, and lease status to prevent recreation races.
4. **Targeted Retirement**:
Retires only the exact confirmed stale binding (clearing ``worktree_path`` on
the lease/checkpoint/lock row and recording durable provenance) while
preserving unrelated worktrees, leases, branches, and git metadata.
5. **Dry-Run & Operator Authorization**:
Supports dry-run inspection and requires explicit cleanup authorization or
reconciler workflow for mutation.
6. **Idempotency & Durable Proof**:
Repeated execution on an already-retired binding is safe, idempotent, and
produces durable audit evidence.
"""
from __future__ import annotations
import json
import os
import subprocess
from typing import Any, Mapping, Sequence
import control_plane_db as cpd
import issue_lock_store
import lease_lifecycle
from merged_cleanup_reconcile import list_local_worktrees
# Classification vocabulary (#970 AC1-AC4).
CLASS_CONFIRMED_STALE_DELETED = "confirmed_stale_deleted_worktree"
CLASS_MOVED_WORKTREE = "moved_worktree_path"
CLASS_UNAVAILABLE_HOST_MOUNT = "unavailable_host_or_mount"
CLASS_LIVE_LEASE_PROTECTED = "live_lease_protected"
CLASS_LIVE_SESSION_PROTECTED = "live_session_protected"
CLASS_PRESENT_VALID = "present_valid_worktree"
CLASS_ALREADY_RETIRED = "already_retired_binding"
ALL_CLASSIFICATIONS = frozenset({
CLASS_CONFIRMED_STALE_DELETED,
CLASS_MOVED_WORKTREE,
CLASS_UNAVAILABLE_HOST_MOUNT,
CLASS_LIVE_LEASE_PROTECTED,
CLASS_LIVE_SESSION_PROTECTED,
CLASS_PRESENT_VALID,
CLASS_ALREADY_RETIRED,
})
RETIRE_ELIGIBLE_CLASSES = frozenset({
CLASS_CONFIRMED_STALE_DELETED,
})
def _norm_path(p: str | None) -> str:
if not p or not str(p).strip():
return ""
try:
return os.path.realpath(os.path.abspath(str(p).strip()))
except Exception:
return str(p).strip()
def check_host_mount_health(project_root: str) -> dict[str, Any]:
"""Check if the repository root and branches/ mount are healthy and accessible."""
root = _norm_path(project_root)
if not root or not os.path.isdir(root):
return {
"healthy": False,
"reason": f"project_root '{project_root}' does not exist or is not a directory",
}
# Verify project_root is inside a valid git repository
try:
res = subprocess.run(
["git", "-C", root, "rev-parse", "--git-dir"],
capture_output=True,
text=True,
check=False,
)
if res.returncode != 0:
return {
"healthy": False,
"reason": f"project_root '{root}' is not a valid git repository: {res.stderr.strip()}",
}
except Exception as exc:
return {
"healthy": False,
"reason": f"git execution failed at project_root '{root}': {exc}",
}
branches_dir = os.path.join(root, "branches")
if os.path.exists(branches_dir) and not os.path.isdir(branches_dir):
return {
"healthy": False,
"reason": f"branches path '{branches_dir}' exists but is not a directory",
}
return {"healthy": True, "project_root": root, "branches_dir": branches_dir}
def find_moved_worktree_path(
project_root: str,
recorded_path: str,
branch: str | None = None,
) -> str | None:
"""Check if a missing recorded worktree path is actually present at another location.
Returns the new path if found, or None if truly absent.
"""
norm_rec = _norm_path(recorded_path)
if not norm_rec:
return None
wt_list = []
try:
wt_list = list_local_worktrees(project_root)
except Exception:
wt_list = []
for entry in wt_list:
p = _norm_path(entry.get("path"))
b = entry.get("branch")
if not p:
continue
# If git worktree list shows a worktree matching the branch at a different path
if branch and b and b.strip() == branch.strip() and p != norm_rec:
if os.path.exists(p):
return p
# If the basename matches and path exists
if os.path.basename(p) == os.path.basename(norm_rec) and p != norm_rec:
if os.path.exists(p):
return p
# Check potential filesystem candidates under branches/
branches_dir = os.path.join(_norm_path(project_root), "branches")
if os.path.isdir(branches_dir):
candidates_to_check: list[str] = []
if branch:
candidates_to_check.append(os.path.join(branches_dir, branch.replace("/", "-")))
candidates_to_check.append(os.path.join(branches_dir, branch))
if recorded_path:
candidates_to_check.append(os.path.join(branches_dir, os.path.basename(recorded_path)))
for cand in candidates_to_check:
norm_cand = _norm_path(cand)
if norm_cand and norm_cand != norm_rec and os.path.isdir(norm_cand):
return norm_cand
return None
def classify_worktree_binding(
*,
recorded_path: str,
lease_status: str | None = None,
lease_phase: str | None = None,
owner_pid: int | None = None,
session_id: str | None = None,
session_active: bool = False,
branch: str | None = None,
project_root: str,
host_health: dict[str, Any] | None = None,
path_exists_override: bool | None = None,
) -> dict[str, Any]:
"""Classify a single worktree binding record for missing-path safety.
Fail-closed: returns detailed reasons and classification.
"""
norm_p = _norm_path(recorded_path)
reasons: list[str] = []
if not norm_p:
return {
"classification": CLASS_ALREADY_RETIRED,
"retire_eligible": False,
"recorded_path": "",
"reasons": ["worktree_path is empty (already retired/unbound)"],
}
# Check 1: Host / mount health
hh = host_health or check_host_mount_health(project_root)
if not hh.get("healthy"):
reasons.append(
f"host or filesystem mount unavailable: {hh.get('reason')}; "
"refusing to classify missing path as deleted (fail closed)"
)
return {
"classification": CLASS_UNAVAILABLE_HOST_MOUNT,
"retire_eligible": False,
"recorded_path": norm_p,
"reasons": reasons,
"host_health": hh,
}
# Check 2: Path existence
exists = (
path_exists_override
if path_exists_override is not None
else os.path.exists(norm_p)
)
if exists:
return {
"classification": CLASS_PRESENT_VALID,
"retire_eligible": False,
"recorded_path": norm_p,
"reasons": [f"worktree path '{norm_p}' exists on disk"],
}
# Path is missing on disk. Run safety checks before declaring confirmed stale.
# Check 3: Moved path
moved_path = find_moved_worktree_path(project_root, norm_p, branch=branch)
if moved_path:
reasons.append(
f"recorded path '{norm_p}' is missing, but worktree for branch '{branch}' "
f"was found moved to '{moved_path}'"
)
return {
"classification": CLASS_MOVED_WORKTREE,
"retire_eligible": False,
"recorded_path": norm_p,
"moved_to_path": moved_path,
"reasons": reasons,
}
# Check 4: Live lease protection
st = (lease_status or "").strip().lower()
pid_alive = lease_lifecycle.is_process_alive(owner_pid) if owner_pid else False
if st == "active" and (pid_alive or owner_pid is None):
reasons.append(
f"lease is status='active' (owner_pid={owner_pid}, alive={pid_alive}); "
"active lease protects worktree binding"
)
return {
"classification": CLASS_LIVE_LEASE_PROTECTED,
"retire_eligible": False,
"recorded_path": norm_p,
"reasons": reasons,
}
# Check 5: Live session / process protection
if session_active and pid_alive:
reasons.append(
f"session '{session_id}' is active with live process PID {owner_pid}; "
"live session protects worktree binding"
)
return {
"classification": CLASS_LIVE_SESSION_PROTECTED,
"retire_eligible": False,
"recorded_path": norm_p,
"reasons": reasons,
}
# Confirmed stale deleted
reasons.append(
f"worktree path '{norm_p}' does not exist on disk, host is healthy, "
"no moved path was found, and no active lease or live process references it"
)
return {
"classification": CLASS_CONFIRMED_STALE_DELETED,
"retire_eligible": True,
"recorded_path": norm_p,
"reasons": reasons,
}
def audit_missing_worktree_bindings(
db: cpd.ControlPlaneDB | None = None,
*,
project_root: str | None = None,
remote: str = "prgs",
org: str | None = None,
repo: str | None = None,
host: str | None = None,
) -> dict[str, Any]:
"""Audit all recorded worktree bindings and classify missing-path entries.
Read-only / side-effect free.
"""
root = _norm_path(project_root or os.getcwd())
hh = check_host_mount_health(root)
cp_db = db or cpd.ControlPlaneDB()
bindings: list[dict[str, Any]] = []
# 1. Audit control-plane leases
try:
leases_res = lease_lifecycle.list_active_leases(
cp_db,
remote=remote,
org=org,
repo=repo,
include_non_active=True,
limit=500,
)
for L in leases_res.get("leases") or []:
wt = (L.get("worktree_path") or "").strip()
if not wt:
continue
owner_pid = L.get("owner_pid") or L.get("session_pid")
sess_id = L.get("session_id")
br = L.get("branch")
if not br and isinstance(L.get("provenance"), dict):
br = L["provenance"].get("branch")
cls_info = classify_worktree_binding(
recorded_path=wt,
lease_status=L.get("status"),
lease_phase=L.get("phase"),
owner_pid=owner_pid,
session_id=sess_id,
branch=br,
project_root=root,
host_health=hh,
)
bindings.append({
"source": "lease",
"lease_id": L.get("lease_id"),
"session_id": sess_id,
"work_kind": L.get("work_kind"),
"work_number": L.get("work_number"),
"branch": br,
"worktree_path": wt,
"lease_status": L.get("status"),
"lease_phase": L.get("phase"),
"owner_pid": owner_pid,
"remote": L.get("remote") or remote,
"org": L.get("org") or org,
"repo": L.get("repo") or repo,
"host": L.get("host") or host,
**cls_info,
})
except Exception as exc: # noqa: BLE001
bindings.append({
"source": "lease_query_error",
"error": str(exc),
})
# 2. Audit session checkpoints
try:
checkpoints = cp_db.list_session_checkpoints(limit=500)
for cp in checkpoints:
wt = (cp.get("worktree_path") or "").strip()
if not wt:
continue
sess_id = cp.get("session_id")
cls_info = classify_worktree_binding(
recorded_path=wt,
session_id=sess_id,
branch=cp.get("branch"),
project_root=root,
host_health=hh,
)
bindings.append({
"source": "session_checkpoint",
"checkpoint_id": cp.get("checkpoint_id"),
"session_id": sess_id,
"work_kind": cp.get("work_kind"),
"work_number": cp.get("work_number"),
"branch": cp.get("branch"),
"worktree_path": wt,
"status": cp.get("status"),
"remote": cp.get("remote") or remote,
"org": cp.get("org") or org,
"repo": cp.get("repo") or repo,
**cls_info,
})
except Exception: # noqa: BLE001
pass
missing_bindings = [
b for b in bindings if b.get("classification") != CLASS_PRESENT_VALID
]
confirmed_stale = [
b for b in missing_bindings if b.get("classification") == CLASS_CONFIRMED_STALE_DELETED
]
moved = [
b for b in missing_bindings if b.get("classification") == CLASS_MOVED_WORKTREE
]
live_protected = [
b for b in missing_bindings if b.get("classification") in (CLASS_LIVE_LEASE_PROTECTED, CLASS_LIVE_SESSION_PROTECTED)
]
unavailable_host = [
b for b in missing_bindings if b.get("classification") == CLASS_UNAVAILABLE_HOST_MOUNT
]
return {
"project_root": root,
"host_mount_healthy": hh.get("healthy", False),
"host_health_reason": hh.get("reason"),
"total_bindings_audited": len(bindings),
"present_count": len(bindings) - len(missing_bindings),
"missing_count": len(missing_bindings),
"confirmed_stale_count": len(confirmed_stale),
"moved_count": len(moved),
"live_protected_count": len(live_protected),
"unavailable_host_count": len(unavailable_host),
"missing_bindings": missing_bindings,
"confirmed_stale_candidates": confirmed_stale,
"worktrees_dimension_resolved": len(missing_bindings) == 0,
}
def resolve_missing_worktree_binding(
db: cpd.ControlPlaneDB | None = None,
*,
binding: dict[str, Any],
dry_run: bool = True,
operator_authorized: bool = False,
workflow_authorized: bool = False,
project_root: str | None = None,
) -> dict[str, Any]:
"""Safely resolve (retire) one confirmed stale missing worktree binding (#970).
Pre-mutation re-validation runs immediately before mutation to prevent races.
"""
root = _norm_path(project_root or os.getcwd())
cp_db = db or cpd.ControlPlaneDB()
rec_path = binding.get("worktree_path") or binding.get("recorded_path") or ""
if not rec_path:
return {
"success": True,
"performed": False,
"outcome": "already_retired",
"message": "worktree_path is already empty (no action needed)",
}
# 1. Authorization check
authorized = operator_authorized or workflow_authorized
if not authorized and not dry_run:
return {
"success": False,
"performed": False,
"reason": "authorization_required",
"message": "Mutation refused: operator_authorized or workflow_authorized is required to retire a stale binding",
}
# 2. Immediate pre-mutation re-validation
reval = classify_worktree_binding(
recorded_path=rec_path,
lease_status=binding.get("lease_status"),
lease_phase=binding.get("lease_phase"),
owner_pid=binding.get("owner_pid"),
session_id=binding.get("session_id"),
branch=binding.get("branch"),
project_root=root,
)
if reval.get("classification") != CLASS_CONFIRMED_STALE_DELETED:
return {
"success": False,
"performed": False,
"reason": "revalidation_failed",
"revalidation": reval,
"message": f"Pre-mutation re-validation failed: binding classified as '{reval.get('classification')}' instead of '{CLASS_CONFIRMED_STALE_DELETED}'",
}
source = binding.get("source", "lease")
lease_id = binding.get("lease_id")
session_id = binding.get("session_id")
checkpoint_id = binding.get("checkpoint_id")
planned_action = {
"source": source,
"lease_id": lease_id,
"session_id": session_id,
"checkpoint_id": checkpoint_id,
"retired_worktree_path": rec_path,
"classification": CLASS_CONFIRMED_STALE_DELETED,
}
if dry_run:
return {
"success": True,
"performed": False,
"dry_run": True,
"action": "retire_stale_binding",
"planned_action": planned_action,
"message": f"Dry-run: would retire stale missing worktree_path '{rec_path}' for {source} (lease={lease_id})",
}
# 3. Apply mode: execute targeted retirement
audit_proof: dict[str, Any] = {
"retired_worktree_path": rec_path,
"source": source,
"lease_id": lease_id,
"session_id": session_id,
"checkpoint_id": checkpoint_id,
"classification": CLASS_CONFIRMED_STALE_DELETED,
}
if source == "lease" and lease_id:
ret_proof = cp_db.retire_lease_worktree_path(
lease_id,
expected_path=rec_path,
reason="missing_worktree_path_retired_by_reconciler",
)
audit_proof["db_lease_update"] = ret_proof
if source == "session_checkpoint" and (checkpoint_id or session_id):
ret_cp = cp_db.retire_session_checkpoint_worktree_path(
session_id=session_id or "",
checkpoint_id=checkpoint_id,
expected_path=rec_path,
reason="missing_worktree_path_retired_by_reconciler",
)
audit_proof["db_checkpoint_update"] = ret_cp
# Also clean up any matching lease row if checkpoint was supplied or vice versa
if lease_id and source != "lease":
try:
cp_db.retire_lease_worktree_path(lease_id, expected_path=rec_path)
except Exception: # noqa: BLE001
pass
return {
"success": True,
"performed": True,
"dry_run": False,
"action": "retired_stale_binding",
"audit_proof": audit_proof,
"message": f"Successfully retired stale missing worktree_path '{rec_path}' for lease {lease_id}",
}
def reconcile_missing_worktree_bindings(
db: cpd.ControlPlaneDB | None = None,
*,
project_root: str | None = None,
remote: str = "prgs",
org: str | None = None,
repo: str | None = None,
host: str | None = None,
dry_run: bool = True,
operator_authorized: bool = False,
workflow_authorized: bool = False,
) -> dict[str, Any]:
"""Audit and safely reconcile missing worktree path bindings (#970).
Audits all recorded bindings, identifies confirmed stale missing paths,
retires eligible bindings (when dry_run=False and authorized), and returns
durable before/after audit evidence.
"""
cp_db = db or cpd.ControlPlaneDB()
root = _norm_path(project_root or os.getcwd())
initial_audit = audit_missing_worktree_bindings(
cp_db,
project_root=root,
remote=remote,
org=org,
repo=repo,
host=host,
)
candidates = initial_audit.get("confirmed_stale_candidates") or []
resolutions: list[dict[str, Any]] = []
for candidate in candidates:
res = resolve_missing_worktree_binding(
cp_db,
binding=candidate,
dry_run=dry_run,
operator_authorized=operator_authorized,
workflow_authorized=workflow_authorized,
project_root=root,
)
resolutions.append(res)
post_audit = audit_missing_worktree_bindings(
cp_db,
project_root=root,
remote=remote,
org=org,
repo=repo,
host=host,
)
return {
"success": True,
"dry_run": dry_run,
"operator_authorized": operator_authorized,
"workflow_authorized": workflow_authorized,
"before_audit": initial_audit,
"resolutions": resolutions,
"after_audit": post_audit,
"worktrees_dimension_resolved": post_audit.get("worktrees_dimension_resolved", False),
"stale_candidates_count": len(candidates),
"resolved_count": sum(1 for r in resolutions if r.get("performed") or (dry_run and r.get("success"))),
}
+8 -3
View File
@@ -37,12 +37,17 @@ def normalize_role_kind(
*,
profile_name: str | None = None,
) -> str:
"""Map profile/task role to a workspace namespace key."""
"""Map profile/task role to a workspace namespace key.
Exact profile-name matches only for controller routing (#840 / #664 B1):
substring lookalikes such as ``fake-controller`` must not become
controller.
"""
role = (role_kind or "author").strip().lower()
profile = (profile_name or "").strip().lower()
if role == "reviewer" and "merger" in profile:
if role == "reviewer" and profile in ("prgs-merger", "mdcps-merger", "merger"):
return "merger"
if "controller" in profile or role == "controller":
if role == "controller" or profile in ("prgs-controller", "controller"):
return "controller"
if role in ROLE_WORKTREE_ENVS:
return role
-22
View File
@@ -204,28 +204,6 @@ proposed command before running it; `gitea_audit_runtime_recovery_contamination`
to inspect or (reconciler-only) clear the marker. Full contrast in
`docs/mcp-namespace-eof-recovery.md`.
## Connected is not attached (#708)
A host/CLI MCP inventory showing **Connected** is not proof the tools are usable.
The active session can expose **none** of a Connected server's tool namespaces —
a *session attachment* failure, distinct from config drift (#672),
transport-closed (#584), and resolver EOF (#685).
Required preflight proof is **live tool visibility plus `gitea_whoami` on the role
namespace**, never host Connected status alone. Call
`gitea_assess_mcp_namespace_attachment` with the Connected set and the namespaces
actually attached to the session; it returns the typed condition
**mcp_connected_namespaces_missing** with per-namespace connected-vs-attached proof,
and records the verdict so review and merge fail closed while a required namespace
is unattached.
Only sanctioned recovery: the client attach/reconnect path
(`gitea_request_mcp_reconnect`), then full preflight
(`gitea_whoami``gitea_resolve_task_capability` → task). Never recover by direct
module import, CLI/raw API mutation, profile hopping, session-state overrides,
process kills, or `.env`/mtime edits. A final report must not claim a healthy
session without attachment proof.
## Shell Spawn Hard-Stop Rule
`exit_code: -1` with empty stdout/stderr means the shell failed to spawn — not a
+9
View File
@@ -576,6 +576,15 @@ TASK_CAPABILITY_MAP: dict[str, dict[str, str]] = {
"permission": "runtime.record_analytics_usage",
"role": "author",
},
# #664: emergency break-glass MCP restart workflow (privileged controller role).
"break_glass_restart": {
"permission": "runtime.break_glass_restart",
"role": "controller",
},
"gitea_break_glass_restart": {
"permission": "runtime.break_glass_restart",
"role": "controller",
},
}
+982
View File
@@ -0,0 +1,982 @@
"""Tests for emergency break-glass MCP restart workflow (#664).
Regression suite for review #641 remediation: B13, B1, B14, B6/B11, B8, and
preservation of previously accepted B2/B3/B5/B7/B9/B10 corrections.
"""
from __future__ import annotations
import os
import unittest
from unittest.mock import MagicMock, patch
import gitea_audit
import gitea_config
import gitea_mcp_server
def _controller_profile(**extra) -> dict:
base = {
"profile_name": "prgs-controller",
"execution_profile": "prgs-controller",
"role": "reconciler", # declared role must not be redefined by capability
"allowed_operations": [
"gitea.read",
"gitea.issue.create",
"gitea.branch.delete",
"gitea.pr.close",
"gitea.pr.comment",
"gitea.issue.comment",
"runtime.break_glass_restart",
],
"forbidden_operations": [
"gitea.pr.approve",
"gitea.pr.merge",
"gitea.pr.create",
"gitea.branch.push",
],
}
base.update(extra)
return base
def _gate_open_patches():
"""Keep master-parity / runtime-mode blocks out of unit tests."""
return (
patch.object(gitea_mcp_server, "_master_parity_block", return_value=[]),
patch.object(gitea_mcp_server, "_runtime_mode_block", return_value=[]),
patch.object(gitea_mcp_server, "_try_auto_switch_for_operation", return_value=False),
)
class TestBreakGlassRestart(unittest.TestCase):
"""Test suite for gitea_break_glass_restart tool and guardrails (#664)."""
def setUp(self) -> None:
self.env_patcher = patch.dict(os.environ, {}, clear=False)
self.env_patcher.start()
os.environ.pop("GITEA_BREAKGLASS_RESTART_AUTHORIZATION", None)
os.environ.pop("GITEA_SANCTIONED_RESTART_HOOK", None)
os.environ.pop("GITEA_AUDIT_LOG", None)
# Reset injectable executor between tests.
gitea_mcp_server._break_glass_restart_executor = None
def tearDown(self) -> None:
gitea_mcp_server._break_glass_restart_executor = None
self.env_patcher.stop()
# ── B13: operation registration / real gate ───────────────────────────
def test_normalize_operation_accepts_canonical_break_glass_op(self) -> None:
"""B13: production normalizer accepts exact runtime.break_glass_restart."""
self.assertEqual(
gitea_config.normalize_operation(
"runtime.break_glass_restart", service="runtime"
),
"runtime.break_glass_restart",
)
ok, reason = gitea_config.check_operation(
"runtime.break_glass_restart",
["gitea.read", "runtime.break_glass_restart"],
)
self.assertTrue(ok, reason)
self.assertEqual(reason, "allowed")
def test_normalize_unknown_and_misspelled_ops_fail_closed(self) -> None:
"""B13: unknown / misspelled operations fail closed (no gitea.read fallback)."""
# Well-formed but misspelled runtime op normalizes, then is not allowed.
ok, reason = gitea_config.check_operation(
"runtime.break_glass_restar", # misspelled
["gitea.read", "runtime.break_glass_restart"],
)
self.assertFalse(ok)
self.assertEqual(reason, "not-allowed")
ok2, reason2 = gitea_config.check_operation(
"runtime.break_glass_restart",
["gitea.read"], # capability not granted
)
self.assertFalse(ok2)
self.assertEqual(reason2, "not-allowed")
ok3, reason3 = gitea_config.check_operation(
"frobnicate",
["gitea.read", "runtime.break_glass_restart"],
)
self.assertFalse(ok3)
self.assertEqual(reason3, "invalid-operation")
# gitea.read grant alone never authorizes break-glass.
ok4, reason4 = gitea_config.check_operation(
"runtime.break_glass_restart",
["gitea.read"],
)
self.assertFalse(ok4)
self.assertNotEqual(reason4, "allowed")
def test_real_profile_operation_gate_without_stubbing(self) -> None:
"""B13: exercise real _profile_operation_gate (not stubbed)."""
allowed = _controller_profile()
denied = _controller_profile(
allowed_operations=["gitea.read", "gitea.issue.create"]
)
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=allowed):
self.assertEqual(
gitea_mcp_server._profile_operation_gate(
"runtime.break_glass_restart"
),
[],
)
with patch.object(gitea_mcp_server, "get_profile", return_value=denied):
reasons = gitea_mcp_server._profile_operation_gate(
"runtime.break_glass_restart"
)
self.assertTrue(reasons)
self.assertTrue(
any("runtime.break_glass_restart" in r or "not allowed" in r
for r in reasons)
)
def test_entry_point_uses_same_operation_as_gate(self) -> None:
"""B13: entry point enforces runtime.break_glass_restart, not gitea.read."""
# Profile has gitea.read but not the break-glass capability.
prof = _controller_profile(
allowed_operations=["gitea.read", "gitea.issue.create"]
)
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart required due to deadlock in worker pool",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=True,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertFalse(res["break_glass_executed"])
self.assertEqual(res["blocker_kind"], "permission_denied")
self.assertNotIn("gitea.read", " ".join(res.get("reasons") or []))
# ── B1 / B14: exact profile auth, no substring, role preservation ─────
def test_trusted_prgs_controller_authorized(self) -> None:
"""B1: exact trusted prgs-controller with capability is authorized."""
prof = _controller_profile()
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": []},
), patch.object(
gitea_mcp_server, "_authenticated_username", return_value="sysadmin"
):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart required due to deadlock in worker pool",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=True,
remote="prgs",
)
self.assertTrue(res["success"], res)
self.assertFalse(res["break_glass_executed"])
def test_fabricated_controller_like_profile_names_denied(self) -> None:
"""B1: lookalike profile names never become authorized."""
p_parity, p_runtime, p_switch = _gate_open_patches()
for name in (
"fake-controller",
"controller-copy",
"not-controller",
"xcontrollerx",
"CONTROLLER",
"prgs-controller-copy",
):
prof = _controller_profile(profile_name=name, execution_profile=name)
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart required due to deadlock in worker pool",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=True,
remote="prgs",
)
self.assertFalse(res["success"], name)
self.assertEqual(res["blocker_kind"], "role_authorization", name)
self.assertFalse(res["break_glass_executed"])
def test_ordinary_and_non_controller_reconciler_denied(self) -> None:
"""B1: ordinary roles and non-controller reconcilers are denied."""
p_parity, p_runtime, p_switch = _gate_open_patches()
denied = [
{"profile_name": "prgs-author", "role": "author",
"allowed_operations": ["gitea.read", "runtime.break_glass_restart"],
"forbidden_operations": []},
{"profile_name": "prgs-reviewer", "role": "reviewer",
"allowed_operations": ["gitea.read", "runtime.break_glass_restart"],
"forbidden_operations": []},
{"profile_name": "prgs-merger", "role": "merger",
"allowed_operations": ["gitea.read", "runtime.break_glass_restart"],
"forbidden_operations": []},
{"profile_name": "prgs-reconciler", "role": "reconciler",
"allowed_operations": [
"gitea.read", "gitea.branch.delete", "runtime.break_glass_restart"
],
"forbidden_operations": []},
{"profile_name": "prgs-controller", "role": "reconciler",
"allowed_operations": ["gitea.read"],
"forbidden_operations": []}, # no capability
]
for prof in denied:
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart required due to deadlock in worker pool",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=True,
remote="prgs",
)
self.assertFalse(res["success"], prof)
self.assertFalse(res["break_glass_executed"], prof)
self.assertIn(
res["blocker_kind"],
("role_authorization", "permission_denied"),
prof,
)
def test_env_var_cannot_grant_authorization_or_bypass_denial(self) -> None:
"""B2 preserved: env var cannot grant break-glass authorization."""
os.environ["GITEA_BREAKGLASS_RESTART_AUTHORIZATION"] = "secret-bypass-token"
prof = {
"profile_name": "prgs-author",
"role": "author",
"allowed_operations": [
"gitea.read", "gitea.issue.create", "runtime.break_glass_restart"
],
"forbidden_operations": [],
}
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart attempting env var bypass",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=True,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertFalse(res["break_glass_executed"])
self.assertEqual(res["blocker_kind"], "role_authorization")
def test_profile_role_kind_no_substring_authority(self) -> None:
"""B1/B14: substring lookalikes do not become controller."""
for name in (
"fake-controller",
"controller-copy",
"not-controller",
"xcontrollerx",
"myCONTROLLER",
):
prof = {
"profile_name": name,
"role": "author",
"allowed_operations": ["gitea.read"],
}
self.assertEqual(
gitea_mcp_server._profile_role_kind(prof),
"author",
name,
)
# Role substrings must not promote.
for role in ("not-controller", "control-plane", "xcontrollerx"):
prof = {"profile_name": "other", "role": role, "allowed_operations": ["gitea.read"]}
self.assertEqual(
gitea_mcp_server._profile_role_kind(prof),
role,
role,
)
def test_prgs_controller_retains_declared_reconciler_role(self) -> None:
"""B14: break-glass capability does not redefine global role."""
prof = _controller_profile(role="reconciler")
self.assertEqual(gitea_mcp_server._profile_role_kind(prof), "reconciler")
# Declared controller still wins when declared.
prof2 = _controller_profile(role="controller")
self.assertEqual(gitea_mcp_server._profile_role_kind(prof2), "controller")
def test_cleanup_merged_pr_branch_role_resolution_unchanged(self) -> None:
"""B14: prgs-controller with reconciler role still resolves for cleanup."""
# When declared reconciler, cleanup gate's role check should see reconciler.
prof = _controller_profile(role="reconciler")
self.assertEqual(gitea_mcp_server._profile_role_kind(prof), "reconciler")
# Fabricated controller-like names with reconciler ops do not become controller.
fake = {
"profile_name": "fake-controller",
"role": "reconciler",
"allowed_operations": [
"gitea.read", "gitea.branch.delete", "gitea.pr.close"
],
}
self.assertEqual(gitea_mcp_server._profile_role_kind(fake), "reconciler")
def test_narrow_break_glass_capability_does_not_redefine_role(self) -> None:
"""B14: granting runtime.break_glass_restart does not invent controller role."""
prof = {
"profile_name": "prgs-reconciler",
"role": "reconciler",
"allowed_operations": [
"gitea.read",
"gitea.branch.delete",
"runtime.break_glass_restart",
],
}
self.assertEqual(gitea_mcp_server._profile_role_kind(prof), "reconciler")
# ── B2-style input validation (preserved) ─────────────────────────────
def test_reason_validation(self) -> None:
"""AC2: Reason is required and must be at least 10 characters long."""
prof = _controller_profile()
p_parity, p_runtime, p_switch = _gate_open_patches()
for invalid_reason in ["", " ", "too short", "123456789"]:
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof):
res = gitea_mcp_server.gitea_break_glass_restart(
reason=invalid_reason,
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertEqual(res["blocker_kind"], "missing_required_fields")
def test_confirmation_validation(self) -> None:
"""AC2: Confirmation phrase must match exact required string."""
prof = _controller_profile()
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart needed due to stuck daemon processes",
confirmation="wrong_confirmation_phrase",
impact_ack=True,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertEqual(res["blocker_kind"], "confirmation_mismatch")
def test_impact_ack_validation(self) -> None:
"""AC2: impact_ack=True is mandatory."""
prof = _controller_profile()
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart needed due to stuck daemon processes",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=False,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertEqual(res["blocker_kind"], "impact_ack_required")
def test_incident_permission_gate(self) -> None:
"""B3 preserved: Gate incident creation on gitea.issue.create permission."""
prof = _controller_profile(
allowed_operations=["gitea.read", "runtime.break_glass_restart"]
)
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart needed due to hung worker process cohort",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
create_incident_issue=True,
dry_run=False,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertEqual(res["blocker_kind"], "permission_denied")
# ── B8: redaction ─────────────────────────────────────────────────────
def test_redaction_key_value_and_connection_strings(self) -> None:
"""B8: key/value, bearer, connection-string, and embedded secrets."""
cases = [
("password=hunter2supersecret", ["hunter2supersecret"], "password"),
("api_key: sk-live-abcdef1234567890ab", ["sk-live-abcdef1234567890ab"], "api_key"),
("Server=db;Password=s3cretValue;Uid=sa", ["s3cretValue"], "password"),
(
"Authorization: Bearer eyJhbGciOiJIUzI1NiJ9.abc.def",
["eyJhbGciOiJIUzI1NiJ9.abc.def", "Bearer eyJ"],
"authorization",
),
(
"token ghp_1234567890abcdef12345678 embedded",
["ghp_1234567890abcdef12345678"],
"token",
),
("nested note password=letmein12345 end", ["letmein12345"], "password"),
]
for raw, secrets, key in cases:
out = gitea_audit._redact_str(raw)
self.assertIn("[REDACTED]", out, raw)
for secret in secrets:
self.assertNotIn(secret, out, raw)
self.assertIn(key, out.lower(), raw)
nested = gitea_audit.redact({
"reason": "password=supersecret99",
"items": [{"api_key": "abc123xyz"}, "token ghp_abcdefghijklmnop1234"],
"error": RuntimeError("pwd=nestedSecret99"),
})
# Exception objects pass through redact as non-str/non-container; ensure
# string forms are covered via str conversion in _redact_str usage.
self.assertEqual(nested["items"][0]["api_key"], gitea_audit.REDACTED)
self.assertNotIn("supersecret99", nested["reason"])
self.assertNotIn("ghp_abcdefghijklmnop1234", nested["items"][1])
def test_redaction_preserves_benign_sec_prefix_text(self) -> None:
"""B8: ordinary text beginning with sec- must not be erased."""
benign = (
"Emergency restart in sec-primary-region for sector-planning "
"and secondary-health checks"
)
out = gitea_audit._redact_str(benign)
self.assertIn("sec-primary-region", out)
self.assertIn("sector-planning", out)
self.assertIn("secondary-health", out)
self.assertNotIn("[REDACTED]", out)
def test_redaction_across_break_glass_surfaces(self) -> None:
"""B8: operator reason is redacted on result, incident, and audit surfaces."""
prof = _controller_profile()
raw_reason = (
"Emergency restart: password=supersecret99 api_key: "
"sk-live-abcdef1234567890ab and url https://user:[email protected]/api"
)
mock_api_request = MagicMock(return_value={"number": 101, "title": "[INCIDENT]"})
fake_exec = {
"success": True,
"apply_supported": True,
"apply_authorized": True,
"restart_performed": True,
"break_glass_executed": True,
}
gitea_mcp_server._break_glass_restart_executor = lambda req: fake_exec
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_mcp_server, "_auth", return_value={"Authorization": "token test"}
), patch.object(
gitea_mcp_server, "_authenticated_username", return_value="sysadmin"
), patch.object(
gitea_mcp_server, "api_request", mock_api_request
), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": []},
), patch.object(
gitea_mcp_server, "gitea_reconcile_after_restart",
return_value={"success": True},
), patch.object(
gitea_audit, "audit_enabled", return_value=True
), patch.object(
gitea_audit, "write_event", return_value=True
):
res = gitea_mcp_server.gitea_break_glass_restart(
reason=raw_reason,
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=False,
create_incident_issue=True,
remote="prgs",
)
self.assertTrue(res["success"], res)
self.assertNotIn("supersecret99", res["reason"])
self.assertNotIn("sk-live-abcdef1234567890ab", res["reason"])
self.assertNotIn("user:[email protected]", res["reason"])
posted_body = mock_api_request.call_args[0][3]["body"]
self.assertNotIn("supersecret99", posted_body)
self.assertNotIn("sk-live-abcdef1234567890ab", posted_body)
# ── B6/B11: reachable executor / truthful flags ───────────────────────
def test_dry_run_never_executes_and_reports_false(self) -> None:
"""B7/B6: dry-run never executes; break_glass_executed always false."""
prof = _controller_profile()
called = {"n": 0}
def _should_not_run(_req):
called["n"] += 1
return {"restart_performed": True, "break_glass_executed": True}
gitea_mcp_server._break_glass_restart_executor = _should_not_run
mock_audit = MagicMock()
mock_api = MagicMock()
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_mcp_server, "_authenticated_username", return_value="sysadmin"
), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": [{"session_id": "s1"}]},
), patch.object(
gitea_audit, "write_event", mock_audit
), patch.object(
gitea_mcp_server, "api_request", mock_api
):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart preview in dry-run mode",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=True,
remote="prgs",
)
self.assertTrue(res["success"])
self.assertTrue(res["dry_run"])
self.assertFalse(res["break_glass_executed"])
self.assertTrue(res["would_execute"])
self.assertEqual(called["n"], 0)
mock_audit.assert_not_called()
mock_api.assert_not_called()
def test_unsupported_apply_truthful(self) -> None:
"""B6/B11: unsupported apply returns blocked result, execution false."""
prof = _controller_profile()
gitea_mcp_server._break_glass_restart_executor = lambda req: {
"success": False,
"apply_supported": False,
"apply_authorized": False,
"restart_performed": False,
"break_glass_executed": False,
"reasons": ["no hook"],
}
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_mcp_server, "_auth", return_value={"Authorization": "token test"}
), patch.object(
gitea_mcp_server, "_authenticated_username", return_value="sysadmin"
), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": []},
), patch.object(
gitea_audit, "audit_enabled", return_value=True
), patch.object(
gitea_audit, "write_event", return_value=True
), patch.object(
gitea_mcp_server, "api_request",
return_value={"number": 555, "title": "[INCIDENT]"},
):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Privileged restart request with unsupported apply",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=False,
create_incident_issue=True,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertFalse(res["performed"])
self.assertFalse(res["break_glass_executed"])
self.assertEqual(res["blocker_kind"], "apply_unsupported")
def test_delegation_success_rejection_and_failure(self) -> None:
"""B6/B11: distinct terminal states for success / rejection / failure."""
prof = _controller_profile()
p_parity, p_runtime, p_switch = _gate_open_patches()
def _run(exec_result, recon=None):
gitea_mcp_server._break_glass_restart_executor = lambda req: exec_result
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_mcp_server, "_auth", return_value={"Authorization": "token test"}
), patch.object(
gitea_mcp_server, "_authenticated_username", return_value="sysadmin"
), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": []},
), patch.object(
gitea_mcp_server, "gitea_reconcile_after_restart",
return_value=recon or {"success": True},
), patch.object(
gitea_audit, "audit_enabled", return_value=True
), patch.object(
gitea_audit, "write_event", return_value=True
), patch.object(
gitea_mcp_server, "api_request",
return_value={"number": 555, "title": "[INCIDENT]"},
):
return gitea_mcp_server.gitea_break_glass_restart(
reason="Privileged break-glass restart delegation path",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=False,
create_incident_issue=True,
remote="prgs",
)
ok = _run({
"success": True,
"apply_supported": True,
"apply_authorized": True,
"restart_performed": True,
"break_glass_executed": True,
})
self.assertTrue(ok["success"], ok)
self.assertTrue(ok["break_glass_executed"])
self.assertTrue(ok["performed"])
rejected = _run({
"success": False,
"apply_supported": True,
"apply_authorized": False,
"restart_performed": False,
"break_glass_executed": False,
"reasons": ["class denied"],
})
self.assertFalse(rejected["success"])
self.assertFalse(rejected["break_glass_executed"])
self.assertEqual(rejected["blocker_kind"], "restart_delegation_failed")
failed = _run({
"success": False,
"apply_supported": True,
"apply_authorized": True,
"restart_performed": False,
"break_glass_executed": False,
"reasons": ["executor error"],
})
self.assertFalse(failed["success"])
self.assertFalse(failed["break_glass_executed"])
self.assertEqual(failed["blocker_kind"], "restart_delegation_failed")
def test_reconciliation_success_and_failure_truthful_flags(self) -> None:
"""B10 preserved: recon failure keeps break_glass_executed=true."""
prof = _controller_profile()
gitea_mcp_server._break_glass_restart_executor = lambda req: {
"success": True,
"apply_supported": True,
"apply_authorized": True,
"restart_performed": True,
"break_glass_executed": True,
}
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_mcp_server, "_auth", return_value={"Authorization": "token test"}
), patch.object(
gitea_mcp_server, "_authenticated_username", return_value="sysadmin"
), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": []},
), patch.object(
gitea_mcp_server, "gitea_reconcile_after_restart",
return_value={"success": False, "error": "lease cleanup failed"},
), patch.object(
gitea_audit, "audit_enabled", return_value=True
), patch.object(
gitea_audit, "write_event", return_value=True
), patch.object(
gitea_mcp_server, "api_request",
return_value={"number": 555, "title": "[INCIDENT]"},
):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Privileged restart request with failing reconciliation",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=False,
create_incident_issue=True,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertTrue(res["performed"])
self.assertTrue(res["break_glass_executed"])
self.assertEqual(res["blocker_kind"], "reconciliation_failed")
def test_authorization_is_not_execution(self) -> None:
"""B6: apply_authorized alone never sets break_glass_executed."""
# Default executor without hook → unsupported, not executed.
prof = _controller_profile()
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_mcp_server, "_auth", return_value={"Authorization": "token test"}
), patch.object(
gitea_mcp_server, "_authenticated_username", return_value="sysadmin"
), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": []},
), patch.object(
gitea_audit, "audit_enabled", return_value=True
), patch.object(
gitea_audit, "write_event", return_value=True
), patch.object(
gitea_mcp_server, "api_request",
return_value={"number": 555, "title": "[INCIDENT]"},
):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Privileged restart without host hook configured",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=False,
create_incident_issue=True,
remote="prgs",
)
self.assertFalse(res["break_glass_executed"])
self.assertEqual(res["blocker_kind"], "apply_unsupported")
# ── preserved fail-closed pre-exec (B5/B9) ────────────────────────────
def test_audit_failure_before_execution_fails_closed(self) -> None:
"""B9 preserved: Audit recording failure stops execution fail-closed."""
prof = _controller_profile()
called = {"n": 0}
gitea_mcp_server._break_glass_restart_executor = lambda req: called.__setitem__("n", called["n"] + 1) or {}
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_mcp_server, "_authenticated_username", return_value="sysadmin"
), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": []},
), patch.object(
gitea_audit, "audit_enabled", return_value=True
), patch.object(
gitea_audit, "write_event", return_value=False
), patch.object(
gitea_mcp_server, "api_request", MagicMock()
):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart with failing audit sink",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=False,
create_incident_issue=True,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertFalse(res["break_glass_executed"])
self.assertEqual(res["blocker_kind"], "audit_recording_failed")
self.assertEqual(called["n"], 0)
def test_incident_creation_failure_before_execution_fails_closed(self) -> None:
"""B5 preserved: Incident creation failure stops execution fail-closed."""
prof = _controller_profile()
called = {"n": 0}
gitea_mcp_server._break_glass_restart_executor = lambda req: called.__setitem__("n", called["n"] + 1) or {}
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_mcp_server, "_auth", return_value={"Authorization": "token test"}
), patch.object(
gitea_mcp_server, "_authenticated_username", return_value="sysadmin"
), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": []},
), patch.object(
gitea_audit, "audit_enabled", return_value=True
), patch.object(
gitea_audit, "write_event", return_value=True
), patch.object(
gitea_mcp_server, "api_request",
side_effect=RuntimeError("Gitea 500 API Error"),
):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart with failing incident POST",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=False,
create_incident_issue=True,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertFalse(res["break_glass_executed"])
self.assertEqual(res["blocker_kind"], "incident_creation_failed")
self.assertEqual(called["n"], 0)
def test_incident_opt_out_on_real_execution_fails_closed(self) -> None:
"""B5 preserved: create_incident_issue=False fails closed on real execution."""
prof = _controller_profile()
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": []},
):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart trying to skip incident creation",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=False,
create_incident_issue=False,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertFalse(res["break_glass_executed"])
self.assertEqual(res["blocker_kind"], "incident_creation_required")
def test_audit_disabled_fails_closed(self) -> None:
"""B9 preserved: Disabling audit recording blocks execution fail-closed."""
prof = _controller_profile()
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_audit, "audit_enabled", return_value=False
), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": []},
):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Emergency restart with disabled audit logging",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=False,
create_incident_issue=True,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertFalse(res["break_glass_executed"])
self.assertEqual(res["blocker_kind"], "audit_recording_failed")
def test_capability_map_registration(self) -> None:
"""B12/B13: capability map registers exact runtime.break_glass_restart."""
from task_capability_map import TASK_CAPABILITY_MAP
entry = TASK_CAPABILITY_MAP.get("gitea_break_glass_restart")
self.assertIsNotNone(entry)
self.assertEqual(entry["permission"], "runtime.break_glass_restart")
self.assertEqual(entry["role"], "controller")
entry2 = TASK_CAPABILITY_MAP.get("break_glass_restart")
self.assertEqual(entry2["permission"], "runtime.break_glass_restart")
# ── B8 / B6 / B15 remediation tests ─────────────────────────────────────
def test_b8_redaction_gitea_token_and_uri_credentials(self) -> None:
"""B8: GITEA_TOKEN= and URI userinfo credentials redacted without erasing neighbours."""
# GITEA_TOKEN= with underscore key
out1 = gitea_audit._redact_str("failed with GITEA_TOKEN=synthetic_tok_123456789")
self.assertIn("GITEA_TOKEN=[REDACTED]", out1)
self.assertNotIn("synthetic_tok_123456789", out1)
# Connection string with URI userinfo
out2 = gitea_audit._redact_str("conn postgres://user:[email protected]:5432/app")
self.assertIn("postgres://[REDACTED_USER]:[REDACTED_PASS]@db.internal:5432/app", out2)
self.assertNotIn("s3cr3tpw", out2)
# Value boundary preserving adjacent audit evidence (correlation_id, incident_number)
raw_audit = "password=secret123;correlation_id=bg-7f2a1c;incident_number=4242"
out3 = gitea_audit._redact_str(raw_audit)
self.assertIn("password=[REDACTED]", out3)
self.assertIn("correlation_id=bg-7f2a1c", out3)
self.assertIn("incident_number=4242", out3)
self.assertNotIn("secret123", out3)
# Query param boundary in URL preserving adjacent parameters
raw_url = "token=abc-123&pr=908&issue=664&head=c67f39b4"
out4 = gitea_audit._redact_str(raw_url)
self.assertIn("token=[REDACTED]", out4)
self.assertIn("pr=908", out4)
self.assertIn("issue=664", out4)
self.assertIn("head=c67f39b4", out4)
def test_b6_default_executor_environment_text_cannot_imply_execution(self) -> None:
"""B6/B11: GITEA_SANCTIONED_RESTART_HOOK string alone returns break_glass_executed=False."""
os.environ["GITEA_SANCTIONED_RESTART_HOOK"] = "this-string-is-never-invoked"
prof = _controller_profile()
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prof), patch.object(
gitea_mcp_server, "_auth", return_value={"Authorization": "token test"}
), patch.object(
gitea_mcp_server, "_authenticated_username", return_value="sysadmin"
), patch.object(
gitea_mcp_server, "gitea_request_mcp_restart",
return_value={"affected_sessions": []},
), patch.object(
gitea_audit, "audit_enabled", return_value=True
), patch.object(
gitea_audit, "write_event", return_value=True
), patch.object(
gitea_mcp_server, "api_request",
return_value={"number": 555, "title": "[INCIDENT]"},
):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Privileged restart request with non-empty hook env var",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=False,
create_incident_issue=True,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertFalse(res["performed"])
self.assertFalse(res["break_glass_executed"])
self.assertEqual(res["blocker_kind"], "restart_delegation_failed")
def test_b15_production_prgs_controller_grant_set_and_gates(self) -> None:
"""B15: Genuine prgs-controller carrying runtime.break_glass_restart and gitea.issue.create passes real gates."""
# Full production-shaped prgs-controller profile
prod_profile = {
"profile_name": "prgs-controller",
"execution_profile": "prgs-controller",
"role": "reconciler",
"allowed_operations": [
"gitea.read",
"gitea.pr.close",
"gitea.pr.comment",
"gitea.issue.comment",
"gitea.issue.create",
"runtime.break_glass_restart",
"gitea.branch.delete",
],
"forbidden_operations": [
"gitea.pr.approve",
"gitea.pr.merge",
"gitea.pr.create",
"gitea.branch.push",
],
}
# 1. Real _profile_operation_gate checks
p_parity, p_runtime, p_switch = _gate_open_patches()
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=prod_profile):
# Both required operations pass the real operation gate (no stubs)
self.assertEqual(
gitea_mcp_server._profile_operation_gate("runtime.break_glass_restart"),
[],
)
self.assertEqual(
gitea_mcp_server._profile_operation_gate("gitea.issue.create"),
[],
)
# 2. Missing gitea.issue.create fails incident creation gate
no_issue_create = dict(prod_profile)
no_issue_create["allowed_operations"] = [
"gitea.read", "gitea.pr.close", "runtime.break_glass_restart"
]
with p_parity, p_runtime, p_switch:
with patch.object(gitea_mcp_server, "get_profile", return_value=no_issue_create):
res = gitea_mcp_server.gitea_break_glass_restart(
reason="Restart testing missing gitea.issue.create permission",
confirmation="I_ACKNOWLEDGE_BREAK_GLASS_MCP_RESTART_DISRUPTION",
impact_ack=True,
dry_run=False,
create_incident_issue=True,
remote="prgs",
)
self.assertFalse(res["success"])
self.assertEqual(res["blocker_kind"], "permission_denied")
self.assertIn("gitea.issue.create", " ".join(res.get("reasons") or []))
if __name__ == "__main__":
unittest.main()
+3 -17
View File
@@ -112,19 +112,7 @@ class TestIssue686ManualMcpProvenance(unittest.TestCase):
self.assertTrue(any("All matching profiles for task 'create_issue' (['prgs-author']) are running but stale" in r for r in reasons))
def test_namespace_health_classification_includes_provenance(self):
"""AC 1 & 4: mcp_namespace_health diagnostics include provenance and unconsumed_gitea_env.
#948 narrowed the vocabulary here. This process carries no client-managed
declaration, so the old code labelled it ``manual_launch`` asserting a
hand-launched terminal process it had no evidence for, and contradicting
``gitea_get_runtime_context``, which read the same process and reported
``client_managed``. Absence of proof is now reported as ``unproven``.
The #686 wall itself is unchanged and still asserted below:
``is_client_managed`` stays False, so nothing previously refused is now
permitted. Only the label on the *reason* changed, so remediation names
the proof that is actually missing.
"""
"""AC 1 & 4: mcp_namespace_health diagnostics include provenance and unconsumed_gitea_env."""
process = {
"pid": 5555,
"profile": "prgs-author",
@@ -141,12 +129,10 @@ class TestIssue686ManualMcpProvenance(unittest.TestCase):
process=process,
probe_source="client_namespace",
)
self.assertEqual(res["provenance"], "unproven")
self.assertEqual(res["provenance"], "manual_launch")
self.assertFalse(res["is_client_managed"])
# The wall is intact: no client-managed proof still fails closed.
self.assertTrue(res["provenance_fail_closed"])
self.assertEqual(res["unconsumed_gitea_env"], {"GITEA_DUMMY": "99"})
self.assertEqual(res["diagnostics"]["provenance"], "unproven")
self.assertEqual(res["diagnostics"]["provenance"], "manual_launch")
if __name__ == "__main__":
-269
View File
@@ -1,269 +0,0 @@
"""Regression tests for Issue #708 wiring: detection must reach a gate, not just exist.
The prior #708 slice added a pure decision function with no call site, so a session
whose namespaces were Connected-but-unattached still passed every mutation gate.
These tests pin the parts that make the detection load-bearing:
* the typed condition is distinct from #672 / #584 / #685,
* the session store records a per-namespace attachment verdict,
* review/merge mutations fail closed while a required namespace is unattached,
* recovery is reconnect-only and never suggests an unsafe fallback,
* startup ordering races and discovery-cache telemetry are reported,
* a healthy final report cannot be produced without attachment proof.
"""
import mcp_namespace_health
REQUIRED = ["gitea-author", "gitea-reviewer", "gitea-merger", "gitea-tools"]
def _connected_but_unattached():
"""The #708 signature: host says Connected, session tool surface is empty."""
return mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=REQUIRED,
attached_session_namespaces=[],
required_namespaces=REQUIRED,
)
# --- AC1: distinct typed detection -----------------------------------------
def test_connected_with_empty_session_tool_list_is_typed_distinctly():
res = _connected_but_unattached()
assert res["attachment_healthy"] is False
assert res["discovery_status"] == "connected_but_namespaces_missing"
assert res["error_type"] == "mcp_connected_namespaces_missing"
assert sorted(res["missing_namespaces"]) == sorted(REQUIRED)
# Not misclassified as config drift (#672), transport-closed (#584), or EOF (#685).
assert res["error_type"] not in {
"mcp_config_drift",
"transport_closed",
"mcp_client_eof",
}
def test_proof_carries_connected_and_attached_per_namespace():
res = _connected_but_unattached()
proof = res["proof_of_connected_vs_attached"]
for ns in REQUIRED:
assert proof[ns] == {"connected": True, "attached": False}
# --- AC2: recovery is auto-attach or reconnect-only -------------------------
def test_exact_next_action_is_reconnect_only():
res = _connected_but_unattached()
assert res["reconnect_required"] is True
action = res["exact_next_action"]
assert "Reconnect the IDE/client MCP session" in action
for forbidden in ("pkill", "chmod", "curl", ".env", "sys.path"):
assert forbidden not in action
def test_sanctioned_recovery_tool_is_named():
res = _connected_but_unattached()
assert res["sanctioned_recovery_tool"] == "gitea_request_mcp_reconnect"
assert "gitea_request_mcp_reconnect" in " ".join(res["remediation"])
def test_successful_auto_attach_reports_recovered_without_operator():
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=REQUIRED,
attached_session_namespaces=REQUIRED,
required_namespaces=REQUIRED,
auto_attach_attempted=True,
auto_attach_succeeded=True,
)
assert res["attachment_healthy"] is True
assert res["auto_recovered"] is True
assert res["reconnect_required"] is False
def test_failed_auto_attach_still_requires_reconnect():
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=REQUIRED,
attached_session_namespaces=[],
required_namespaces=REQUIRED,
auto_attach_attempted=True,
auto_attach_succeeded=False,
)
assert res["auto_recovered"] is False
assert res["reconnect_required"] is True
assert any("did not succeed" in r for r in res["reasons"])
def test_reconnect_rediscovery_clears_the_condition():
"""Attach state after a reconnect is healthy without any other change."""
before = _connected_but_unattached()
after = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=REQUIRED,
attached_session_namespaces=REQUIRED,
required_namespaces=REQUIRED,
discovery_cache_hit=False,
discovery_cache_age_seconds=0.0,
)
assert before["attachment_healthy"] is False
assert after["attachment_healthy"] is True
assert after["discovery_status"] == "namespaces_attached"
assert after["missing_namespaces"] == []
# --- AC4: startup ordering across multiple role servers ---------------------
def test_multi_role_startup_ordering_race_is_reported():
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=REQUIRED,
attached_session_namespaces=["gitea-author"],
required_namespaces=REQUIRED,
session_tool_snapshot_at=1000.0,
namespace_connected_at={
"gitea-author": 990.0,
"gitea-reviewer": 1005.0,
"gitea-merger": 1007.0,
},
)
assert res["startup_ordering_race"] is True
assert res["late_attaching_namespaces"] == ["gitea-merger", "gitea-reviewer"]
assert res["telemetry"]["startup_ordering_race"] is True
def test_no_ordering_race_when_snapshot_follows_connect():
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=REQUIRED,
attached_session_namespaces=REQUIRED,
required_namespaces=REQUIRED,
session_tool_snapshot_at=2000.0,
namespace_connected_at={ns: 1000.0 for ns in REQUIRED},
)
assert res["startup_ordering_race"] is False
assert res["late_attaching_namespaces"] == []
# --- AC5: telemetry ---------------------------------------------------------
def test_telemetry_reports_cache_and_recovery_signals():
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=REQUIRED,
attached_session_namespaces=["gitea-author"],
required_namespaces=REQUIRED,
discovery_cache_hit=True,
discovery_cache_age_seconds=42.5,
)
tel = res["telemetry"]
assert tel["connected_count"] == 4
assert tel["attached_count"] == 1
assert tel["missing_count"] == 3
assert tel["discovery_cache_hit"] is True
assert tel["discovery_cache_age_seconds"] == 42.5
assert tel["reconnect_required"] is True
assert tel["error_type"] == "mcp_connected_namespaces_missing"
def test_telemetry_leaks_no_secrets():
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=REQUIRED,
attached_session_namespaces=[],
required_namespaces=REQUIRED,
)
blob = repr(res["telemetry"]).lower()
for leak in ("token", "authorization", "password", "secret", "/users/", "http"):
assert leak not in blob
# --- AC2/AC6: fail-closed mutation gate -------------------------------------
def _session_store_from(assessment):
"""Mirror the server-side recorder without importing the MCP server module."""
store = {}
missing = set(assessment["missing_namespaces"])
for ns, state in assessment["proof_of_connected_vs_attached"].items():
attached = bool(state["attached"])
store[ns] = {
"namespace": ns,
"connected": bool(state["connected"]),
"attached": attached,
"attachment_healthy": attached and ns not in missing,
"error_type": None if attached else assessment["error_type"],
}
return store
def test_review_and_merge_fail_closed_while_unattached():
store = _session_store_from(_connected_but_unattached())
for task in ("review_pr", "submit_review", "merge_pr", "create_pr", "work_issue"):
reasons = mcp_namespace_health.attachment_gate_from_session(task, store)
assert reasons, f"{task} must fail closed while its namespace is unattached"
assert "fail closed, #708" in reasons[0]
def test_gate_offers_only_sanctioned_recovery():
store = _session_store_from(_connected_but_unattached())
reasons = mcp_namespace_health.attachment_gate_from_session("merge_pr", store)
joined = " ".join(reasons)
assert "reconnect" in joined.lower()
assert "Workflow Safety Hard Stop (#708)" in joined
# Unsafe fallbacks appear only inside the prohibition, never as advice.
assert "NEVER use" in joined
def test_gate_passes_once_namespaces_are_attached():
healthy = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=REQUIRED,
attached_session_namespaces=REQUIRED,
required_namespaces=REQUIRED,
)
store = _session_store_from(healthy)
for task in ("review_pr", "submit_review", "merge_pr", "create_pr", "work_issue"):
assert mcp_namespace_health.attachment_gate_from_session(task, store) == []
def test_unassessed_session_does_not_gate():
"""No recorded assessment must not fabricate a block (matches #543 semantics)."""
assert mcp_namespace_health.attachment_gate_from_session("merge_pr", {}) == []
assert mcp_namespace_health.attachment_gate_from_session("merge_pr", None) == []
def test_unmapped_task_is_not_gated():
store = _session_store_from(_connected_but_unattached())
assert mcp_namespace_health.attachment_gate_from_session("gitea_read", store) == []
def test_partial_attachment_gates_only_the_affected_role():
"""Author attached, reviewer not: author work proceeds, review fails closed."""
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=REQUIRED,
attached_session_namespaces=["gitea-author", "gitea-tools"],
required_namespaces=REQUIRED,
)
store = _session_store_from(res)
assert mcp_namespace_health.attachment_gate_from_session("work_issue", store) == []
assert mcp_namespace_health.attachment_gate_from_session("review_pr", store)
assert mcp_namespace_health.attachment_gate_from_session("merge_pr", store)
# --- AC4: no false healthy report without attachment proof ------------------
def test_no_false_healthy_without_attachment_proof():
"""Connected alone never yields a healthy verdict."""
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=REQUIRED,
attached_session_namespaces=None,
required_namespaces=REQUIRED,
)
assert res["success"] is False
assert res["attachment_healthy"] is False
assert res["telemetry"]["attached_count"] == 0
def test_attachment_gate_maps_each_role_namespace():
assert mcp_namespace_health.required_namespace_for_attachment("review_pr") == "gitea-reviewer"
assert mcp_namespace_health.required_namespace_for_attachment("merge_pr") == "gitea-merger"
assert mcp_namespace_health.required_namespace_for_attachment("create_pr") == "gitea-author"
assert mcp_namespace_health.required_namespace_for_attachment("nope") is None
@@ -1,77 +0,0 @@
"""Unit regression tests for Issue #708: Connected-but-namespaces-missing detection and attachment safety."""
import mcp_namespace_health
def test_assess_connected_namespace_attachment_success():
# required_namespaces is declared explicitly: these three are the namespaces this case
# is about. Relying on the default (which also requires gitea-tools) would ask for a
# healthy verdict covering a required namespace that was never connected — exactly the
# false-healthy classification these tests now forbid.
connected = ["gitea-author", "gitea-reviewer", "gitea-merger"]
attached = ["gitea-author", "gitea-reviewer", "gitea-merger"]
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=connected,
attached_session_namespaces=attached,
required_namespaces=connected,
)
assert res["success"] is True
assert res["attachment_healthy"] is True
assert res["discovery_status"] == "namespaces_attached"
assert res["error_type"] is None
assert res["missing_namespaces"] == []
assert res["exact_next_action"] == "None; session tool namespaces attached."
def test_assess_connected_namespace_attachment_missing():
# Every required namespace here *is* connected, so the only condition present is the
# #708 one: Connected at the host, absent from the session tool surface.
connected = ["gitea-author", "gitea-reviewer", "gitea-merger"]
attached = ["gitea-author"]
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=connected,
attached_session_namespaces=attached,
required_namespaces=connected,
)
assert res["success"] is False
assert res["attachment_healthy"] is False
assert res["discovery_status"] == "connected_but_namespaces_missing"
assert res["error_type"] == "mcp_connected_namespaces_missing"
assert "gitea-reviewer" in res["missing_namespaces"]
assert "gitea-merger" in res["missing_namespaces"]
assert "Reconnect the IDE/client MCP session" in res["exact_next_action"]
def test_assess_connected_namespace_attachment_disconnected():
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=[],
attached_session_namespaces=[],
)
assert res["success"] is False
assert res["attachment_healthy"] is False
assert res["discovery_status"] == "disconnected"
def test_proof_of_connected_vs_attached_mapping():
connected = ["gitea-author", "gitea-reviewer"]
attached = ["gitea-author"]
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=connected,
attached_session_namespaces=attached,
)
proof = res["proof_of_connected_vs_attached"]
assert proof["gitea-author"] == {"connected": True, "attached": True}
assert proof["gitea-reviewer"] == {"connected": True, "attached": False}
def test_unsafe_fallback_policy_enforcement():
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=["gitea-author"],
attached_session_namespaces=[],
)
policy = res["unsafe_fallback_policy"]
assert "Workflow Safety Hard Stop (#708)" in policy
assert "direct imports" in policy
assert "API mutations" in policy
assert "profile hopping" in policy
assert "session-state overrides" in policy
@@ -1,283 +0,0 @@
"""Regression tests for Issue #708 B2: not-connected is not Connected-but-unattached.
The first #708 slice counted a namespace as ``missing`` only when it appeared in
``connected_servers``. A *required* namespace absent from that inventory was therefore
never counted at all, so the session reported ``attachment_healthy: true`` with
``error_type: None`` while holding no attachment proof for it and the gate text told the
operator "the host reports Connected" about a service nothing had reported Connected.
These tests pin the corrected distinction:
* required and not connected never yields a healthy verdict,
* it is typed ``mcp_required_namespaces_not_connected``, never the #708 condition,
* ``mcp_connected_namespaces_missing`` stays reserved for genuinely Connected namespaces,
* both categories still fail review and merge closed, with their own reason,
* per-namespace ``connected``/``attached`` evidence is reported accurately,
* one session's evidence cannot clear another session's block.
"""
import mcp_namespace_health
ROLES = ["gitea-author", "gitea-reviewer", "gitea-merger", "gitea-tools"]
NOT_CONNECTED = mcp_namespace_health.ERROR_REQUIRED_NAMESPACES_NOT_CONNECTED
CONNECTED_MISSING = mcp_namespace_health.ERROR_CONNECTED_NAMESPACES_MISSING
def _assess(connected, attached, required):
return mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=connected,
attached_session_namespaces=attached,
required_namespaces=required,
)
def _store(assessment):
"""The recorder contract: per-namespace verdicts feed the gate."""
return dict(assessment["namespace_conditions"])
# --- the four evidence combinations ----------------------------------------
def test_connected_and_attached_is_healthy():
res = _assess(ROLES, ROLES, ROLES)
assert res["attachment_healthy"] is True
assert res["error_type"] is None
assert res["missing_namespaces"] == []
assert res["not_connected_namespaces"] == []
assert res["discovery_status"] == "namespaces_attached"
def test_connected_but_unattached_keeps_the_708_condition():
res = _assess(ROLES, ["gitea-author"], ROLES)
assert res["attachment_healthy"] is False
assert res["error_type"] == CONNECTED_MISSING
assert res["not_connected_namespaces"] == []
assert sorted(res["missing_namespaces"]) == [
"gitea-merger",
"gitea-reviewer",
"gitea-tools",
]
assert res["discovery_status"] == "connected_but_namespaces_missing"
def test_required_but_not_connected_is_never_healthy():
"""The reviewer's exact reproduction from review 637."""
res = _assess(
["gitea-reviewer"], ["gitea-reviewer"], ["gitea-reviewer", "gitea-merger"]
)
assert res["attachment_healthy"] is False
assert res["success"] is False
assert res["error_type"] is not None
assert res["telemetry"]["not_connected_count"] == 1
def test_required_but_not_connected_is_typed_distinctly():
res = _assess(
["gitea-reviewer"], ["gitea-reviewer"], ["gitea-reviewer", "gitea-merger"]
)
assert res["error_type"] == NOT_CONNECTED
assert res["discovery_status"] == "required_namespaces_not_connected"
assert res["not_connected_namespaces"] == ["gitea-merger"]
# Not collapsed into the #708 condition, nor into config drift (#672) or
# transport-closed (#584).
assert CONNECTED_MISSING not in res["error_types"]
assert res["missing_namespaces"] == []
assert res["error_type"] not in {"mcp_config_drift", "transport_closed"}
def test_not_connected_reason_makes_no_connected_claim():
res = _assess(
["gitea-reviewer"], ["gitea-reviewer"], ["gitea-reviewer", "gitea-merger"]
)
about_merger = [r for r in res["reasons"] if "gitea-merger" in r]
assert about_merger, "the not-connected namespace must be named in the reasons"
for reason in about_merger:
assert "report Connected at host/CLI layer" not in reason
assert "gitea-merger" in res["exact_next_action"]
def test_neither_connected_nor_attached_reports_disconnected():
res = _assess([], [], ROLES)
assert res["attachment_healthy"] is False
assert res["discovery_status"] == "disconnected"
assert res["error_type"] == NOT_CONNECTED
assert sorted(res["not_connected_namespaces"]) == sorted(ROLES)
assert res["missing_namespaces"] == []
# --- mixed required namespaces ---------------------------------------------
def test_mixed_connected_attached_and_not_connected():
res = _assess(
["gitea-author"], ["gitea-author"], ["gitea-author", "gitea-merger"]
)
assert res["attachment_healthy"] is False
assert res["not_connected_namespaces"] == ["gitea-merger"]
assert res["missing_namespaces"] == []
conditions = res["namespace_conditions"]
assert conditions["gitea-author"]["attachment_healthy"] is True
assert conditions["gitea-author"]["condition"] is None
assert conditions["gitea-merger"]["attachment_healthy"] is False
assert conditions["gitea-merger"]["condition"] == NOT_CONNECTED
def test_both_conditions_present_are_both_reported():
"""One namespace Connected-but-unattached, another never connected."""
res = _assess(
["gitea-author", "gitea-reviewer"],
["gitea-author"],
["gitea-author", "gitea-reviewer", "gitea-merger"],
)
assert res["missing_namespaces"] == ["gitea-reviewer"]
assert res["not_connected_namespaces"] == ["gitea-merger"]
# Neither condition is hidden by the other; error_type names the primary one.
assert sorted(res["error_types"]) == sorted([NOT_CONNECTED, CONNECTED_MISSING])
assert res["error_type"] == NOT_CONNECTED
# --- unknown, partial, malformed, contradictory evidence --------------------
def test_unknown_attachment_evidence_is_not_healthy():
"""No session tool surface reported at all is unproven, not proven good."""
res = _assess(ROLES, None, ROLES)
assert res["attachment_healthy"] is False
assert res["telemetry"]["attached_count"] == 0
assert res["error_type"] == CONNECTED_MISSING
def test_partial_evidence_gates_only_the_unproven_roles():
res = _assess(ROLES, ["gitea-author", "gitea-tools"], ROLES)
store = _store(res)
assert mcp_namespace_health.attachment_gate_from_session("work_issue", store) == []
assert mcp_namespace_health.attachment_gate_from_session("review_pr", store)
assert mcp_namespace_health.attachment_gate_from_session("merge_pr", store)
def test_malformed_namespace_entries_are_discarded_not_trusted():
"""Blank and whitespace-only names must not become namespaces or proof."""
res = mcp_namespace_health.assess_connected_namespace_attachment(
connected_servers=["gitea-author", "", " "],
attached_session_namespaces=["gitea-author", ""],
required_namespaces=["gitea-author", " "],
)
assert "" not in res["proof_of_connected_vs_attached"]
assert " " not in res["proof_of_connected_vs_attached"]
assert res["attachment_healthy"] is True
assert res["telemetry"]["required_count"] == 1
def test_contradictory_attached_without_connected_fails_closed():
"""Attached in the session yet absent from the connected inventory."""
res = _assess(["gitea-author"], ["gitea-author", "gitea-merger"], ROLES)
assert res["attachment_healthy"] is False
assert "gitea-merger" in res["contradictory_namespaces"]
assert "gitea-merger" in res["not_connected_namespaces"]
assert res["namespace_conditions"]["gitea-merger"]["contradictory_evidence"] is True
assert res["namespace_conditions"]["gitea-merger"]["attachment_healthy"] is False
assert any("contradictory evidence" in r for r in res["reasons"])
assert res["telemetry"]["contradictory_count"] >= 1
def test_duplicate_required_entries_are_counted_once():
res = _assess(["gitea-author"], ["gitea-author"], ["gitea-author", "gitea-author"])
assert res["telemetry"]["required_count"] == 1
assert res["attachment_healthy"] is True
# --- review and merge behaviour for both failure categories -----------------
def test_merge_fails_closed_when_required_namespace_never_connected():
store = _store(_assess(["gitea-author"], ["gitea-author"], ROLES))
reasons = mcp_namespace_health.attachment_gate_from_session("merge_pr", store)
assert reasons
assert NOT_CONNECTED in reasons[0]
assert "fail closed, #708" in reasons[0]
def test_review_fails_closed_when_required_namespace_never_connected():
store = _store(_assess(["gitea-author"], ["gitea-author"], ROLES))
reasons = mcp_namespace_health.attachment_gate_from_session("review_pr", store)
assert reasons
assert NOT_CONNECTED in reasons[0]
assert "fail closed, #708" in reasons[0]
def test_not_connected_block_never_claims_the_host_reports_connected():
store = _store(_assess(["gitea-author"], ["gitea-author"], ROLES))
reasons = mcp_namespace_health.attachment_gate_from_session("merge_pr", store)
assert "the host reports Connected" not in reasons[0]
assert "absent from the connected-service inventory" in reasons[0]
def test_connected_but_unattached_block_still_says_connected():
store = _store(_assess(ROLES, ["gitea-author"], ROLES))
reasons = mcp_namespace_health.attachment_gate_from_session("merge_pr", store)
assert CONNECTED_MISSING in reasons[0]
assert "the host reports Connected" in reasons[0]
def test_both_categories_offer_only_the_sanctioned_recovery():
for store in (
_store(_assess(ROLES, [], ROLES)),
_store(_assess(["gitea-author"], ["gitea-author"], ROLES)),
):
reasons = mcp_namespace_health.attachment_gate_from_session("merge_pr", store)
joined = " ".join(reasons)
assert "reconnect" in joined.lower()
assert "Workflow Safety Hard Stop (#708)" in joined
for forbidden in ("pkill", "chmod", "curl ", ".env", "sys.path"):
assert forbidden not in reasons[0]
def test_entry_without_connected_evidence_blocks_without_asserting_either():
"""A legacy/partial store entry must fail closed and claim nothing it cannot prove."""
store = {"gitea-merger": {"namespace": "gitea-merger", "attached": False}}
reasons = mcp_namespace_health.attachment_gate_from_session("merge_pr", store)
assert reasons
assert "no connected-status evidence" in reasons[0]
assert "the host reports Connected" not in reasons[0]
assert "fail closed, #708" in reasons[0]
# --- session isolation ------------------------------------------------------
def test_one_session_evidence_does_not_clear_another_session_block():
"""Attachment evidence is per-session state; it must not travel between sessions."""
blocked_session = _store(_assess(["gitea-author"], ["gitea-author"], ROLES))
healthy_session = _store(_assess(ROLES, ROLES, ROLES))
assert (
mcp_namespace_health.attachment_gate_from_session("merge_pr", healthy_session)
== []
)
# The healthy session's verdict is not consulted for the blocked session.
assert mcp_namespace_health.attachment_gate_from_session("merge_pr", blocked_session)
# And the blocked session's store is unchanged by the healthy one existing.
assert blocked_session["gitea-merger"]["attachment_healthy"] is False
def test_gate_reads_only_the_store_it_is_given():
healthy_session = _store(_assess(ROLES, ROLES, ROLES))
assert mcp_namespace_health.attachment_gate_from_session("merge_pr", {}) == []
assert mcp_namespace_health.attachment_gate_from_session("merge_pr", None) == []
assert (
mcp_namespace_health.attachment_gate_from_session("merge_pr", healthy_session)
== []
)
# --- telemetry stays secret-free -------------------------------------------
def test_not_connected_telemetry_leaks_no_secrets():
res = _assess(["gitea-author"], ["gitea-author"], ROLES)
blob = repr(res["telemetry"]).lower()
for leak in ("token", "authorization", "password", "secret", "/users/", "http"):
assert leak not in blob
-924
View File
@@ -1,924 +0,0 @@
"""Transport-neutral MCP bind seam (#931).
These tests drive the *real* bind boundary ``mark_sanctioned_daemon`` followed
by ``bind_native_mcp_transport`` from a canonical entrypoint path, with the
pytest allowance switched off rather than mocking the new accessor. The
distinction matters here for the same reason it mattered in #941: a suite that
only exercises the helper in isolation cannot observe a seam that the live path
never reaches.
Covered:
1. no configured transport defaults to the local transport
2. explicit local transport binds
3. the sanctioned remote identifier binds through the seam
4. an unregistered identifier is rejected at bind time
5. an invalid bind prevents the server reaching tool service
6. the unbound state fails closed where a bind is required
7. every transport-aware guard reads the same authoritative value
8. client-controlled input cannot alter the bound transport
9. the durable decision-lock record carries the selected transport
10. existing stdio behaviour is unchanged
11. repeated / conflicting bind attempts follow one fail-closed contract
12. capability, role, repository and provenance protections do not regress
"""
from __future__ import annotations
import os
import tempfile
import unittest
from pathlib import Path
from unittest.mock import patch
REPO_ROOT = Path(__file__).resolve().parent.parent
import mcp_daemon_guard
import mcp_session_state
import mcp_transport_config
import irrecoverable_provenance
class _ProductionBind:
"""Context manager that reaches the real production bind path.
Patches only the two things a unit test cannot otherwise satisfy: the
resolved canonical entrypoint frame, and the pytest allowance that would
short-circuit ``mark_sanctioned_daemon``. Everything downstream of those
validation, pinning, the rebind contract runs unmodified.
"""
def __init__(self, env: dict[str, str] | None = None):
self._env = env or {}
self._stack: list = []
def __enter__(self):
mcp_daemon_guard.clear_native_runtime_for_tests()
canonical = str((REPO_ROOT / "mcp_server.py").resolve())
self._stack = [
patch.object(
mcp_daemon_guard,
"_caller_official_entrypoint_path",
side_effect=lambda: canonical,
),
patch.object(mcp_daemon_guard, "is_pytest_runtime", return_value=False),
patch.dict(os.environ, self._env),
]
for ctx in self._stack:
ctx.__enter__()
# Start from a clean configuration unless the test set one.
if mcp_transport_config.TRANSPORT_ENV not in self._env:
os.environ.pop(mcp_transport_config.TRANSPORT_ENV, None)
mcp_daemon_guard.mark_sanctioned_daemon()
return mcp_daemon_guard
def __exit__(self, *exc):
for ctx in reversed(self._stack):
ctx.__exit__(*exc)
mcp_daemon_guard.clear_native_runtime_for_tests()
return False
class TestPermittedSetIsSingleSourceOfTruth(unittest.TestCase):
"""AC3: identifiers are enumerated once, in the seam."""
def test_guard_allowlist_is_the_seam_allowlist(self):
self.assertIs(
mcp_daemon_guard._PRODUCTION_TRANSPORTS,
mcp_transport_config.SUPPORTED_TRANSPORTS,
)
def test_default_is_a_member_of_the_permitted_set(self):
self.assertIn(
mcp_transport_config.DEFAULT_TRANSPORT,
mcp_transport_config.SUPPORTED_TRANSPORTS,
)
def test_remote_identifier_is_permitted_and_not_the_default(self):
self.assertIn(
mcp_transport_config.REMOTE_TRANSPORT,
mcp_transport_config.SUPPORTED_TRANSPORTS,
)
self.assertNotEqual(
mcp_transport_config.REMOTE_TRANSPORT,
mcp_transport_config.DEFAULT_TRANSPORT,
)
self.assertTrue(
mcp_transport_config.is_remote_transport(
mcp_transport_config.REMOTE_TRANSPORT
)
)
def test_no_default_transport_literal_outside_the_seam(self):
"""AC3: no production module reads the literal outside the seam.
Review 635 flagged that a fixed five-module list cannot catch a *new*
module reintroducing the literal. This globs every production module in
the repository root instead, so the guarantee holds for code that does
not exist yet.
"""
default = mcp_transport_config.DEFAULT_TRANSPORT
needles = (f'"{default}"', f"'{default}'")
seam = Path(mcp_transport_config.__file__).name
scanned: list[str] = []
offenders: list[str] = []
for path in sorted(REPO_ROOT.glob("*.py")):
if path.name == seam:
continue # the seam is the one place the literal may live
scanned.append(path.name)
for lineno, line in enumerate(
path.read_text(encoding="utf-8").splitlines(), start=1
):
code = line.split("#", 1)[0]
if any(needle in code for needle in needles):
offenders.append(f"{path.name}:{lineno}: {line.strip()}")
# Guard the guard: a glob that silently matched nothing would pass.
self.assertGreater(len(scanned), 20, "production glob matched too little")
self.assertIn("mcp_daemon_guard.py", scanned)
self.assertIn("gitea_mcp_server.py", scanned)
self.assertEqual(offenders, [], "\n".join(offenders))
class TestConfiguredTransportResolution(unittest.TestCase):
"""AC1: configuration supplies the identifier; unset still yields the default."""
def test_unset_yields_default(self):
res = mcp_transport_config.resolve_configured_transport(env={})
self.assertEqual(res["transport"], mcp_transport_config.DEFAULT_TRANSPORT)
self.assertFalse(res["configured"])
self.assertEqual(res["source"], mcp_transport_config.SOURCE_DEFAULT)
self.assertTrue(res["supported"])
self.assertEqual(res["reasons"], [])
def test_blank_and_whitespace_are_treated_as_unset(self):
for raw in ("", " ", "\t\n"):
res = mcp_transport_config.resolve_configured_transport(
env={mcp_transport_config.TRANSPORT_ENV: raw}
)
self.assertEqual(res["transport"], mcp_transport_config.DEFAULT_TRANSPORT)
self.assertFalse(res["configured"])
def test_explicit_default_is_reported_as_configured(self):
res = mcp_transport_config.resolve_configured_transport(
env={
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.DEFAULT_TRANSPORT
)
}
)
self.assertEqual(res["transport"], mcp_transport_config.DEFAULT_TRANSPORT)
self.assertTrue(res["configured"])
self.assertEqual(res["source"], mcp_transport_config.SOURCE_CONFIGURED)
def test_remote_identifier_resolves_and_is_supported(self):
res = mcp_transport_config.resolve_configured_transport(
env={
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
)
}
)
self.assertEqual(res["transport"], mcp_transport_config.REMOTE_TRANSPORT)
self.assertTrue(res["supported"])
def test_case_and_padding_are_normalized(self):
padded = f" {mcp_transport_config.REMOTE_TRANSPORT.upper()} "
res = mcp_transport_config.resolve_configured_transport(
env={mcp_transport_config.TRANSPORT_ENV: padded}
)
self.assertEqual(res["transport"], mcp_transport_config.REMOTE_TRANSPORT)
self.assertTrue(res["supported"])
def test_unregistered_identifier_is_not_silently_defaulted(self):
res = mcp_transport_config.resolve_configured_transport(
env={mcp_transport_config.TRANSPORT_ENV: "carrier-pigeon"}
)
self.assertFalse(res["supported"])
self.assertEqual(res["transport"], "carrier-pigeon")
self.assertNotEqual(res["transport"], mcp_transport_config.DEFAULT_TRANSPORT)
self.assertTrue(res["reasons"])
def test_superseded_sse_transport_is_not_registered(self):
"""A real MCP transport that this deployment does not sanction."""
self.assertFalse(mcp_transport_config.is_supported_transport("sse"))
res = mcp_transport_config.resolve_configured_transport(
env={mcp_transport_config.TRANSPORT_ENV: "sse"}
)
self.assertFalse(res["supported"])
def test_require_configured_transport_raises_on_unregistered(self):
with self.assertRaises(mcp_transport_config.TransportConfigurationError):
mcp_transport_config.require_configured_transport(
env={mcp_transport_config.TRANSPORT_ENV: "carrier-pigeon"}
)
def test_non_string_configuration_never_matches_permitted_set(self):
for value in (object(), 1, None, True, ["stdio"], {"t": "stdio"}):
self.assertEqual(mcp_transport_config.normalize_transport(value), "")
self.assertFalse(mcp_transport_config.is_supported_transport(value))
class TestBindSeam(unittest.TestCase):
"""AC1/AC2: the live bind path resolves, validates, and pins."""
def tearDown(self) -> None:
mcp_daemon_guard.clear_native_runtime_for_tests()
def test_1_no_configured_transport_binds_default(self):
with _ProductionBind() as guard:
status = guard.bind_native_mcp_transport()
self.assertEqual(
status["transport"], mcp_transport_config.DEFAULT_TRANSPORT
)
self.assertEqual(
guard.bound_transport(), mcp_transport_config.DEFAULT_TRANSPORT
)
self.assertTrue(status["production_native_mcp_transport"])
def test_2_explicit_default_transport_binds(self):
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.DEFAULT_TRANSPORT
)
}
) as guard:
status = guard.bind_native_mcp_transport()
self.assertEqual(
status["transport"], mcp_transport_config.DEFAULT_TRANSPORT
)
self.assertTrue(guard.is_production_native_mcp_transport())
def test_2b_explicit_argument_still_binds(self):
"""The pre-#931 call form keeps working for launchers and tests."""
with _ProductionBind() as guard:
status = guard.bind_native_mcp_transport(
transport=mcp_transport_config.DEFAULT_TRANSPORT
)
self.assertEqual(
status["transport"], mcp_transport_config.DEFAULT_TRANSPORT
)
def test_3_sanctioned_remote_identifier_binds_through_the_seam(self):
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
)
}
) as guard:
status = guard.bind_native_mcp_transport()
self.assertEqual(
status["transport"], mcp_transport_config.REMOTE_TRANSPORT
)
self.assertEqual(
guard.bound_transport(), mcp_transport_config.REMOTE_TRANSPORT
)
# The remote identifier is trusted exactly like the local one; the
# listener that serves it is #938 and is not implemented here.
self.assertTrue(guard.is_native_mcp_transport())
self.assertTrue(guard.is_production_native_mcp_transport())
guard.assert_production_mutation_runtime("remote-bind")
def test_4_unregistered_identifier_rejected_at_bind_time(self):
with _ProductionBind(
{mcp_transport_config.TRANSPORT_ENV: "carrier-pigeon"}
) as guard:
with self.assertRaises(guard.UnsanctionedRuntimeError) as ctx:
guard.bind_native_mcp_transport()
self.assertIn("carrier-pigeon", str(ctx.exception))
self.assertIn("#931", str(ctx.exception))
# Nothing was bound, so nothing may dispatch.
self.assertIsNone(guard.bound_transport())
self.assertFalse(guard.is_native_mcp_transport())
def test_4b_unregistered_explicit_argument_rejected(self):
with _ProductionBind() as guard:
with self.assertRaises(guard.UnsanctionedRuntimeError):
guard.bind_native_mcp_transport(transport="carrier-pigeon")
self.assertIsNone(guard.bound_transport())
def test_4c_superseded_sse_rejected_at_bind_time(self):
with _ProductionBind({mcp_transport_config.TRANSPORT_ENV: "sse"}) as guard:
with self.assertRaises(guard.UnsanctionedRuntimeError):
guard.bind_native_mcp_transport()
self.assertIsNone(guard.bound_transport())
def test_5_invalid_bind_prevents_tool_service(self):
"""A failed bind must stop the server before it serves tools."""
with _ProductionBind(
{mcp_transport_config.TRANSPORT_ENV: "carrier-pigeon"}
) as guard:
with self.assertRaises(guard.UnsanctionedRuntimeError):
guard.bind_native_mcp_transport()
# This is the exact expression the entrypoint passes to mcp.run.
with self.assertRaises(guard.UnsanctionedRuntimeError) as ctx:
guard.assert_transport_bound("tool service")
self.assertIn("No MCP transport is bound", str(ctx.exception))
def test_6_unbound_state_fails_closed(self):
"""Entrypoint claimed but never bound — the offline-import shape."""
with _ProductionBind() as guard:
self.assertIsNone(guard.bound_transport())
self.assertFalse(guard.is_native_mcp_transport())
with self.assertRaises(guard.UnsanctionedRuntimeError):
guard.assert_transport_bound("tool service")
with self.assertRaises(guard.UnsanctionedRuntimeError):
guard.assert_sanctioned_mutation_runtime("gitea_mutation")
def test_6b_no_runtime_at_all_fails_closed(self):
mcp_daemon_guard.clear_native_runtime_for_tests()
with patch.object(mcp_daemon_guard, "is_pytest_runtime", return_value=False):
self.assertIsNone(mcp_daemon_guard.bound_transport())
with self.assertRaises(mcp_daemon_guard.UnsanctionedRuntimeError):
mcp_daemon_guard.assert_transport_bound("tool service")
class TestOneAuthoritativeValue(unittest.TestCase):
"""AC: every transport-aware guard observes the same value."""
def tearDown(self) -> None:
mcp_daemon_guard.clear_native_runtime_for_tests()
def test_7_all_guards_read_the_same_bound_value(self):
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
)
}
) as guard:
guard.bind_native_mcp_transport()
expected = mcp_transport_config.REMOTE_TRANSPORT
self.assertEqual(guard.bound_transport(), expected)
self.assertEqual(guard.assert_transport_bound(), expected)
self.assertEqual(guard.native_runtime_status()["bound_transport"], expected)
self.assertEqual(guard.native_runtime_status()["transport"], expected)
self.assertEqual(
guard.mutation_provenance_fields()["bound_transport"], expected
)
self.assertEqual(
irrecoverable_provenance.assess_transport_for_auth_mint()[
"bound_transport"
],
expected,
)
def test_8_environment_change_after_bind_cannot_move_the_value(self):
"""Client- or environment-shaped input must not alter a bound transport."""
with _ProductionBind() as guard:
guard.bind_native_mcp_transport()
self.assertEqual(
guard.bound_transport(), mcp_transport_config.DEFAULT_TRANSPORT
)
# A stray launcher (or an attacker) rewrites config post-bind.
os.environ[mcp_transport_config.TRANSPORT_ENV] = (
mcp_transport_config.REMOTE_TRANSPORT
)
self.assertEqual(
guard.bound_transport(), mcp_transport_config.DEFAULT_TRANSPORT
)
self.assertEqual(
guard.mutation_provenance_fields()["bound_transport"],
mcp_transport_config.DEFAULT_TRANSPORT,
)
os.environ[mcp_transport_config.TRANSPORT_ENV] = "carrier-pigeon"
self.assertEqual(
guard.bound_transport(), mcp_transport_config.DEFAULT_TRANSPORT
)
def test_8b_bound_transport_takes_no_caller_argument(self):
"""The accessor cannot be steered by a tool parameter."""
import inspect
self.assertEqual(
list(inspect.signature(mcp_daemon_guard.bound_transport).parameters), []
)
def test_11_rebinding_the_same_transport_is_idempotent(self):
with _ProductionBind() as guard:
first = guard.bind_native_mcp_transport()
second = guard.bind_native_mcp_transport()
self.assertEqual(first["transport"], second["transport"])
self.assertEqual(
guard.bound_transport(), mcp_transport_config.DEFAULT_TRANSPORT
)
def test_11b_rebinding_a_different_transport_fails_closed(self):
with _ProductionBind() as guard:
guard.bind_native_mcp_transport()
with self.assertRaises(guard.UnsanctionedRuntimeError) as ctx:
guard.bind_native_mcp_transport(
transport=mcp_transport_config.REMOTE_TRANSPORT
)
self.assertIn("already bound", str(ctx.exception))
# The first value survives the attempt.
self.assertEqual(
guard.bound_transport(), mcp_transport_config.DEFAULT_TRANSPORT
)
def test_11c_rebinding_an_unregistered_transport_fails_closed(self):
with _ProductionBind() as guard:
guard.bind_native_mcp_transport()
with self.assertRaises(guard.UnsanctionedRuntimeError):
guard.bind_native_mcp_transport(transport="carrier-pigeon")
self.assertEqual(
guard.bound_transport(), mcp_transport_config.DEFAULT_TRANSPORT
)
class TestDurableRecordCarriesTransport(unittest.TestCase):
"""AC4: the identifier reaches a durable decision-lock record."""
def tearDown(self) -> None:
mcp_daemon_guard.clear_native_runtime_for_tests()
def _save_and_read_decision_lock(self, state_dir: str) -> dict:
mcp_session_state.save_state(
kind=mcp_session_state.KIND_DECISION_LOCK,
payload={"pr_number": 931, "action": "COMMENT"},
remote="prgs",
org="Scaled-Tech-Consulting",
repo="Gitea-Tools",
profile_identity="prgs-author",
state_dir=state_dir,
)
loaded = mcp_session_state.load_state(
kind=mcp_session_state.KIND_DECISION_LOCK,
remote="prgs",
org="Scaled-Tech-Consulting",
repo="Gitea-Tools",
profile_identity="prgs-author",
state_dir=state_dir,
)
self.assertIsNotNone(loaded)
return loaded
def test_9_decision_lock_records_the_bound_transport(self):
with tempfile.TemporaryDirectory() as tmp:
mcp_daemon_guard.clear_native_runtime_for_tests()
mcp_daemon_guard.install_test_native_runtime()
record = self._save_and_read_decision_lock(tmp)
self.assertIn("bound_transport", record)
self.assertEqual(
record["bound_transport"], mcp_daemon_guard.bound_transport()
)
def test_9b_unbound_runtime_records_no_transport_identifier(self):
with tempfile.TemporaryDirectory() as tmp:
mcp_daemon_guard.clear_native_runtime_for_tests()
record = self._save_and_read_decision_lock(tmp)
self.assertIn("bound_transport", record)
self.assertIsNone(record["bound_transport"])
def test_9c_provenance_fields_expose_the_identifier(self):
mcp_daemon_guard.clear_native_runtime_for_tests()
fields = mcp_daemon_guard.mutation_provenance_fields()
self.assertIn("bound_transport", fields)
self.assertIsNone(fields["bound_transport"])
class TestStdioBehaviourUnchanged(unittest.TestCase):
"""AC6 / prompt items 10 and 12: no regression on the existing path."""
def tearDown(self) -> None:
mcp_daemon_guard.clear_native_runtime_for_tests()
def test_10_trust_class_field_keeps_its_pre_931_values(self):
"""``transport`` remains the trust class, not the identifier."""
mcp_daemon_guard.clear_native_runtime_for_tests()
self.assertEqual(
mcp_daemon_guard.mutation_provenance_fields()["transport"], "untrusted"
)
mcp_daemon_guard.install_test_native_runtime()
self.assertEqual(
mcp_daemon_guard.mutation_provenance_fields()["transport"],
"test_native_mcp",
)
with _ProductionBind() as guard:
guard.bind_native_mcp_transport()
self.assertEqual(
guard.mutation_provenance_fields()["transport"], "native_mcp"
)
def test_10b_default_bind_reproduces_the_pre_931_runtime_record(self):
with _ProductionBind() as guard:
status = guard.bind_native_mcp_transport()
self.assertTrue(status["native_mcp_transport"])
self.assertTrue(status["production_native_mcp_transport"])
self.assertEqual(status["mode"], "production")
self.assertEqual(status["phase"], "transport_bound")
self.assertEqual(
status["transport"], mcp_transport_config.DEFAULT_TRANSPORT
)
guard.assert_sanctioned_mutation_runtime("native-ide")
guard.assert_production_mutation_runtime("native-ide")
def test_12_session_state_root_still_pinned_at_bind(self):
"""#695 AC2 must survive the seam."""
with tempfile.TemporaryDirectory() as legit:
with tempfile.TemporaryDirectory() as rogue:
with _ProductionBind(
{mcp_daemon_guard.SESSION_STATE_DIR_ENV: legit}
) as guard:
guard.bind_native_mcp_transport()
self.assertEqual(
guard.pinned_session_state_dir(), str(Path(legit).resolve())
)
os.environ[mcp_daemon_guard.SESSION_STATE_DIR_ENV] = rogue
self.assertEqual(
guard.pinned_session_state_dir(), str(Path(legit).resolve())
)
def test_12b_test_mode_record_still_cannot_authorize_production(self):
mcp_daemon_guard.clear_native_runtime_for_tests()
mcp_daemon_guard.install_test_native_runtime()
self.assertTrue(mcp_daemon_guard.is_native_mcp_transport())
self.assertFalse(mcp_daemon_guard.is_production_native_mcp_transport())
with self.assertRaises(mcp_daemon_guard.UnsanctionedRuntimeError):
mcp_daemon_guard.assert_production_mutation_runtime("prod-endpoint")
def test_12c_bind_still_requires_the_canonical_entrypoint(self):
"""A remote identifier does not relax provenance."""
mcp_daemon_guard.clear_native_runtime_for_tests()
with patch.object(mcp_daemon_guard, "is_pytest_runtime", return_value=False):
with patch.object(
mcp_daemon_guard,
"_caller_official_entrypoint_path",
return_value=None,
):
with self.assertRaises(mcp_daemon_guard.UnsanctionedRuntimeError):
mcp_daemon_guard.bind_native_mcp_transport(
transport=mcp_transport_config.REMOTE_TRANSPORT
)
self.assertIsNone(mcp_daemon_guard.bound_transport())
def test_12d_bind_still_requires_a_prior_entrypoint_claim(self):
mcp_daemon_guard.clear_native_runtime_for_tests()
canonical = str((REPO_ROOT / "mcp_server.py").resolve())
with patch.object(mcp_daemon_guard, "is_pytest_runtime", return_value=False):
with patch.object(
mcp_daemon_guard,
"_caller_official_entrypoint_path",
side_effect=lambda: canonical,
):
# No mark_sanctioned_daemon() first.
with self.assertRaises(
mcp_daemon_guard.UnsanctionedRuntimeError
) as ctx:
mcp_daemon_guard.bind_native_mcp_transport()
self.assertIn("no entrypoint claim", str(ctx.exception))
def test_12e_auth_mint_verdict_is_unchanged_for_the_default_transport(self):
with _ProductionBind() as guard:
guard.bind_native_mcp_transport()
verdict = irrecoverable_provenance.assess_transport_for_auth_mint()
self.assertTrue(verdict["allowed"])
self.assertTrue(verdict["native_mcp_transport"])
self.assertTrue(verdict["production_native_mcp_transport"])
self.assertEqual(verdict["reasons"], [])
def test_12f_auth_mint_still_refuses_an_unbound_runtime(self):
with _ProductionBind():
# Claimed but never bound.
verdict = irrecoverable_provenance.assess_transport_for_auth_mint()
self.assertFalse(verdict["allowed"])
self.assertTrue(verdict["reasons"])
self.assertIsNone(verdict["bound_transport"])
class TestExecutionBoundary(unittest.TestCase):
"""Recognition is not execution authorization (#931, review 635 B1/B2).
A registered remote identifier must still bind, pin and record the #931
seam while being refused at the serve boundary, because serving it would
start a listener with no authentication or per-request principal. That
listener belongs to #938.
"""
def tearDown(self) -> None:
mcp_daemon_guard.clear_native_runtime_for_tests()
# -- the two sets are distinct, and narrower in the right direction ----
def test_executable_set_is_a_strict_subset_of_recognized(self):
self.assertTrue(
mcp_transport_config.EXECUTABLE_TRANSPORTS
< mcp_transport_config.SUPPORTED_TRANSPORTS
)
def test_default_transport_is_executable(self):
self.assertTrue(
mcp_transport_config.is_executable_transport(
mcp_transport_config.DEFAULT_TRANSPORT
)
)
def test_remote_transport_is_recognized_but_not_executable(self):
remote = mcp_transport_config.REMOTE_TRANSPORT
self.assertTrue(mcp_transport_config.is_supported_transport(remote))
self.assertFalse(mcp_transport_config.is_executable_transport(remote))
def test_remote_listener_ownership_is_declared(self):
self.assertEqual(
mcp_transport_config.TRANSPORT_EXECUTION_OWNER[
mcp_transport_config.REMOTE_TRANSPORT
],
"#938",
)
# -- stdio still reaches the runner, unchanged -------------------------
def test_default_transport_is_authorized_for_service(self):
with _ProductionBind() as guard:
guard.bind_native_mcp_transport()
self.assertEqual(
guard.authorize_transport_execution("tool service"),
mcp_transport_config.DEFAULT_TRANSPORT,
)
self.assertTrue(guard.assess_serve_authorization()["allowed"])
def test_default_transport_reaches_the_real_runner(self):
"""The production runner is actually invoked, with stdio, unchanged."""
with _ProductionBind() as guard:
guard.bind_native_mcp_transport()
seen = {}
class _Runner:
def run(self, transport=None, **kw):
seen["transport"] = transport
_Runner().run(transport=guard.authorize_transport_execution("tool service"))
self.assertEqual(
seen["transport"], mcp_transport_config.DEFAULT_TRANSPORT
)
# -- streamable-http binds, records, and is refused before serving -----
def test_remote_transport_binds_and_is_recorded(self):
"""The #931 seam is intact: it binds, pins and records."""
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
)
}
) as guard:
guard.bind_native_mcp_transport()
self.assertEqual(
guard.bound_transport(), mcp_transport_config.REMOTE_TRANSPORT
)
self.assertEqual(
guard.mutation_provenance_fields()["bound_transport"],
mcp_transport_config.REMOTE_TRANSPORT,
)
def test_remote_transport_is_refused_at_the_serve_boundary(self):
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
)
}
) as guard:
guard.bind_native_mcp_transport()
with self.assertRaises(guard.TransportExecutionError) as ctx:
guard.authorize_transport_execution("tool service")
err = ctx.exception
self.assertEqual(
err.blocker_kind,
mcp_transport_config.BLOCKER_LISTENER_NOT_COMMISSIONED,
)
self.assertEqual(err.owner_issue, "#938")
self.assertEqual(err.transport, mcp_transport_config.REMOTE_TRANSPORT)
def test_refusal_names_transport_and_unmet_requirement_and_owner(self):
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
)
}
) as guard:
guard.bind_native_mcp_transport()
with self.assertRaises(guard.TransportExecutionError) as ctx:
guard.authorize_transport_execution("tool service")
text = str(ctx.exception)
self.assertIn(mcp_transport_config.REMOTE_TRANSPORT, text)
self.assertIn("#938", text)
self.assertIn("not commissioned", text)
self.assertIn("#931", text)
def test_remote_transport_never_reaches_the_runner(self):
"""No transport value is handed to a run() call for the remote case."""
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
)
}
) as guard:
guard.bind_native_mcp_transport()
calls = []
class _Runner:
def run(self, transport=None, **kw):
calls.append(transport)
with self.assertRaises(guard.TransportExecutionError):
_Runner().run(
transport=guard.authorize_transport_execution("tool service")
)
self.assertEqual(calls, [], "runner must never be invoked")
def test_no_http_listener_is_created_for_remote_transport(self):
"""Nothing in the refusal path touches uvicorn or a socket bind."""
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
)
}
) as guard:
guard.bind_native_mcp_transport()
import socket
bound_sockets = []
real_bind = socket.socket.bind
def _tripwire(self, addr): # pragma: no cover - must not run
bound_sockets.append(addr)
return real_bind(self, addr)
with patch.object(socket.socket, "bind", _tripwire):
with self.assertRaises(guard.TransportExecutionError):
guard.authorize_transport_execution("tool service")
self.assertEqual(bound_sockets, [], "no socket may be bound")
def test_no_mutation_is_authorized_after_the_denial(self):
"""The denial leaves no partial state that would let a tool dispatch."""
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
)
}
) as guard:
guard.bind_native_mcp_transport()
with self.assertRaises(guard.TransportExecutionError):
guard.authorize_transport_execution("tool service")
# Serve stays unauthorized on every subsequent query.
self.assertFalse(guard.assess_serve_authorization()["allowed"])
self.assertFalse(guard.native_runtime_status()["serve_authorized"])
with self.assertRaises(guard.TransportExecutionError):
guard.authorize_transport_execution("tool service")
# -- B2: the decision genuinely consumes bound_transport ---------------
def test_serve_decision_consumes_bound_transport(self):
"""Changing only bound_transport flips the verdict."""
with _ProductionBind() as guard:
guard.bind_native_mcp_transport()
self.assertTrue(guard.assess_serve_authorization()["allowed"])
with patch.object(
guard,
"bound_transport",
return_value=mcp_transport_config.REMOTE_TRANSPORT,
):
verdict = guard.assess_serve_authorization()
self.assertFalse(verdict["allowed"])
self.assertEqual(
verdict["transport"], mcp_transport_config.REMOTE_TRANSPORT
)
with self.assertRaises(guard.TransportExecutionError):
guard.authorize_transport_execution("tool service")
def test_serve_verdict_reports_the_bound_transport(self):
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
)
}
) as guard:
guard.bind_native_mcp_transport()
verdict = guard.assess_serve_authorization()
self.assertEqual(
verdict["transport"], mcp_transport_config.REMOTE_TRANSPORT
)
self.assertTrue(verdict["recognized"])
self.assertFalse(verdict["executable"])
# -- earlier and later boundaries are unchanged ------------------------
def test_unregistered_identifier_still_fails_at_bind_not_at_serve(self):
with _ProductionBind(
{mcp_transport_config.TRANSPORT_ENV: "carrier-pigeon"}
) as guard:
with self.assertRaises(guard.UnsanctionedRuntimeError) as ctx:
guard.bind_native_mcp_transport()
self.assertIn("not a registered MCP transport", str(ctx.exception))
self.assertIsNone(guard.bound_transport())
def test_unbound_execution_keeps_the_pre_existing_failure(self):
with _ProductionBind() as guard:
with self.assertRaises(guard.UnsanctionedRuntimeError) as ctx:
guard.authorize_transport_execution("tool service")
self.assertIn("No MCP transport is bound", str(ctx.exception))
self.assertNotIsInstance(ctx.exception, guard.TransportExecutionError)
def test_transport_execution_error_is_caught_by_existing_handlers(self):
"""Subclassing keeps every pre-existing fail-closed handler correct."""
self.assertTrue(
issubclass(
mcp_daemon_guard.TransportExecutionError,
mcp_daemon_guard.UnsanctionedRuntimeError,
)
)
def test_test_mode_runtime_is_not_servable(self):
"""The pytest-only record is outside the recognized and executable sets."""
mcp_daemon_guard.clear_native_runtime_for_tests()
mcp_daemon_guard.install_test_native_runtime()
self.assertNotIn(
mcp_daemon_guard.bound_transport(),
mcp_transport_config.SUPPORTED_TRANSPORTS,
)
self.assertFalse(mcp_daemon_guard.assess_serve_authorization()["allowed"])
def test_serve_authorization_does_not_leak_across_runtimes(self):
"""A later runtime's verdict never reflects an earlier one."""
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
)
}
) as guard:
guard.bind_native_mcp_transport()
self.assertFalse(guard.assess_serve_authorization()["allowed"])
with _ProductionBind() as guard:
guard.bind_native_mcp_transport()
self.assertTrue(guard.assess_serve_authorization()["allowed"])
mcp_daemon_guard.clear_native_runtime_for_tests()
self.assertEqual(
mcp_daemon_guard.assess_serve_authorization()["blocker_kind"],
mcp_transport_config.BLOCKER_TRANSPORT_NOT_BOUND,
)
def test_refusal_carries_no_credential_material(self):
with _ProductionBind(
{
mcp_transport_config.TRANSPORT_ENV: (
mcp_transport_config.REMOTE_TRANSPORT
),
"GITEA_TOKEN": "super-secret-value",
}
) as guard:
guard.bind_native_mcp_transport()
with self.assertRaises(guard.TransportExecutionError) as ctx:
guard.authorize_transport_execution("tool service")
blob = str(ctx.exception) + repr(ctx.exception.assessment)
self.assertNotIn("super-secret-value", blob)
self.assertNotIn("GITEA_TOKEN", blob)
class TestEntrypointWiring(unittest.TestCase):
"""The live entrypoint must use the seam and the execution guard."""
def test_entrypoint_binds_without_a_literal_transport(self):
text = (REPO_ROOT / "gitea_mcp_server.py").read_text(encoding="utf-8")
self.assertIn("mcp_daemon_guard.bind_native_mcp_transport()", text)
self.assertNotIn('bind_native_mcp_transport(transport="stdio")', text)
def test_entrypoint_serves_only_through_the_execution_guard(self):
text = (REPO_ROOT / "gitea_mcp_server.py").read_text(encoding="utf-8")
self.assertIn(
"mcp.run(transport=mcp_daemon_guard.authorize_transport_execution", text
)
self.assertNotIn('mcp.run(transport="stdio")', text)
self.assertNotIn(
"mcp.run(transport=mcp_daemon_guard.assert_transport_bound", text
)
def test_every_run_call_in_production_goes_through_the_guard(self):
"""Glob the production surface: no serve site may bypass the guard."""
offenders: list[str] = []
run_sites = 0
for path in sorted(REPO_ROOT.glob("*.py")):
for lineno, line in enumerate(
path.read_text(encoding="utf-8").splitlines(), start=1
):
code = line.split("#", 1)[0]
if "mcp.run(" not in code:
continue
run_sites += 1
if "authorize_transport_execution" not in code:
offenders.append(f"{path.name}:{lineno}: {line.strip()}")
self.assertEqual(run_sites, 1, "expected exactly one serve site")
self.assertEqual(offenders, [], "\n".join(offenders))
if __name__ == "__main__":
unittest.main()
@@ -1,662 +0,0 @@
"""Client/session-aware runtime ownership and provenance (#948).
Covers the reproduced contradiction that motivated the issue: one surface
reporting ``client_managed`` while another reported ``manual_launch`` for the
same process, remediation hardcoded to one vendor, and a profile-wide duplicate
wall that could not tell two healthy clients apart.
All client and session identifiers here are synthetic.
"""
from __future__ import annotations
import os
import tempfile
import unittest
from datetime import datetime, timedelta, timezone
import mcp_client_reconnect
import mcp_namespace_health
import mcp_worker_identity as mwi
NOW = datetime(2026, 7, 29, 6, 0, 0, tzinfo=timezone.utc)
def _registry() -> mwi.WorkerRegistry:
"""A registry on a throwaway path; never the operator's real one."""
handle, path = tempfile.mkstemp(suffix=".sqlite3")
os.close(handle)
os.unlink(path)
return mwi.WorkerRegistry(path)
def _attach(
registry: mwi.WorkerRegistry,
*,
client: str,
session: str,
generation: str,
profile: str = "prgs-reviewer",
role: str = "reviewer",
pid: int = 4242,
now: datetime = NOW,
ttl: float = 900.0,
) -> dict:
"""Register one synthetic worker and return the outcome."""
identity = mwi.generate_worker_identity(client, session, now=now)
outcome = registry.register(
worker_identity=identity,
client_name=client,
client_instance_id=f"inst-{session}",
session_id=session,
generation_id=generation,
role=role,
profile=profile,
pid=pid,
heartbeat_ttl_seconds=ttl,
now=now,
)
outcome["identity"] = identity
return outcome
class IdentityFormatTests(unittest.TestCase):
"""AC27-29: collision-resistant `<llm-name>-<UTC-timestamp>-<short-sha>`."""
def test_identity_matches_required_format(self):
identity = mwi.generate_worker_identity("Gemini", "sess-0001", now=NOW)
parsed = mwi.parse_worker_identity(identity)
self.assertTrue(parsed["valid"], parsed["reasons"])
self.assertEqual(parsed["client_name"], "gemini")
self.assertEqual(parsed["minted_at"], "20260729T060000Z")
self.assertEqual(len(parsed["digest"]), 12)
def test_digest_varies_with_session_and_nonce(self):
base = dict(timestamp_ns=1, now=NOW)
a = mwi.generate_worker_identity("codex", "sess-A", nonce="n", **base)
b = mwi.generate_worker_identity("codex", "sess-B", nonce="n", **base)
c = mwi.generate_worker_identity("codex", "sess-A", nonce="m", **base)
self.assertNotEqual(a, b, "session must feed the digest")
self.assertNotEqual(a, c, "nonce must feed the digest")
def test_identity_is_not_role_or_profile(self):
"""AC26: identity is independent of role and profile."""
args = dict(timestamp_ns=7, nonce="fixed", now=NOW)
same = mwi.generate_worker_identity("claude", "sess-1", **args)
self.assertEqual(same, mwi.generate_worker_identity("claude", "sess-1", **args))
# Nothing role- or profile-derived appears in the identity.
self.assertNotIn("reviewer", same)
self.assertNotIn("prgs", same)
def test_malformed_identity_rejected(self):
self.assertFalse(mwi.parse_worker_identity("prgs-reviewer")["valid"])
self.assertFalse(mwi.parse_worker_identity("")["valid"])
self.assertFalse(mwi.parse_worker_identity(None)["valid"])
class PerClientAttachmentTests(unittest.TestCase):
"""Every supported client attaches and is reported as itself."""
def _assert_attached_as(self, client: str, expected_name: str):
registry = _registry()
outcome = _attach(
registry, client=client, session=f"sess-{client}", generation="gen-1"
)
self.assertTrue(outcome["registered"], outcome["reasons"])
verdict = mwi.assess_provenance(
registry=registry, worker_identity=outcome["identity"], env={}, now=NOW
)
self.assertEqual(verdict["session_ownership"], mwi.OWNERSHIP_OWNED)
self.assertEqual(verdict["provenance"], mwi.PROVENANCE_CLIENT_SESSION)
self.assertEqual(verdict["client_name"], expected_name)
self.assertTrue(verdict["session_owned"])
self.assertFalse(verdict["fail_closed"])
return verdict
def test_codex_attachment(self):
self._assert_attached_as("codex", "codex")
def test_gemini_attachment(self):
self._assert_attached_as("gemini", "gemini")
def test_antigravity_attachment(self):
self._assert_attached_as("antigravity", "antigravity")
def test_claude_attachment(self):
self._assert_attached_as("claude", "claude_code")
def test_unknown_client_is_named_not_guessed(self):
verdict = self._assert_attached_as("some_new_llm", "some_new_llm")
self.assertNotEqual(verdict["client_name"], "codex")
class SessionLifecycleTests(unittest.TestCase):
def test_same_client_new_session_gets_distinct_identity(self):
registry = _registry()
first = _attach(registry, client="codex", session="sess-1", generation="gen-1")
second = _attach(registry, client="codex", session="sess-2", generation="gen-2")
self.assertTrue(first["registered"])
self.assertTrue(second["registered"])
self.assertNotEqual(first["identity"], second["identity"])
# Both are live and neither blocks the other.
cohort = mwi.classify_cohort(registry.list_workers(), now=NOW)
self.assertEqual(cohort["live_worker_count"], 2)
self.assertFalse(cohort["blocked"], cohort["reasons"])
def test_different_client_attaches_after_previous_session_ends(self):
"""AC14: expiry then takeover with a higher fencing epoch."""
registry = _registry()
gone = _attach(
registry, client="codex", session="sess-old", generation="gen-shared", ttl=60
)
later = NOW + timedelta(hours=1)
self.assertFalse(
registry.is_live(registry.get(gone["identity"]), now=later)["live"]
)
arriving = _attach(
registry,
client="gemini",
session="sess-new",
generation="gen-other",
now=later,
)
claim = registry.claim_generation(
worker_identity=arriving["identity"],
generation_id="gen-shared",
now=later,
)
self.assertTrue(claim["claimed"], claim["reasons"])
self.assertIn(gone["identity"], claim["superseded_workers"])
self.assertGreater(claim["fencing_epoch"], gone["fencing_epoch"])
def test_superseded_session_is_fenced_on_resume(self):
"""AC15/AC16: the prior session cannot heartbeat its way back."""
registry = _registry()
old = _attach(
registry, client="codex", session="sess-old", generation="gen-shared", ttl=60
)
later = NOW + timedelta(hours=1)
new = _attach(
registry, client="gemini", session="sess-new", generation="gen-x", now=later
)
registry.claim_generation(
worker_identity=new["identity"], generation_id="gen-shared", now=later
)
resumed = registry.heartbeat(
worker_identity=old["identity"],
fencing_epoch=old["fencing_epoch"],
now=later,
)
self.assertFalse(resumed["renewed"])
self.assertFalse(resumed["mutation_performed"])
self.assertEqual(resumed["blocker_kind"], mwi.BLOCKER_FENCED)
def test_heartbeat_renews_only_the_owning_lease(self):
"""AC11: a wrong epoch never renews, and never mutates."""
registry = _registry()
worker = _attach(registry, client="codex", session="s", generation="g")
good = registry.heartbeat(
worker_identity=worker["identity"],
fencing_epoch=worker["fencing_epoch"],
now=NOW + timedelta(minutes=5),
)
self.assertTrue(good["renewed"])
bad = registry.heartbeat(
worker_identity=worker["identity"],
fencing_epoch=worker["fencing_epoch"] + 99,
now=NOW + timedelta(minutes=6),
)
self.assertFalse(bad["renewed"])
self.assertFalse(bad["mutation_performed"])
self.assertEqual(
registry.get(worker["identity"])["last_heartbeat_at"],
good["last_heartbeat_at"],
"a refused heartbeat must not advance the record",
)
class ConflictAndCollisionTests(unittest.TestCase):
def test_two_live_sessions_cannot_claim_one_generation(self):
registry = _registry()
first = _attach(registry, client="codex", session="s1", generation="gen-shared")
second = _attach(registry, client="gemini", session="s2", generation="gen-other")
claim = registry.claim_generation(
worker_identity=second["identity"],
generation_id="gen-shared",
now=NOW,
)
self.assertFalse(claim["claimed"])
self.assertFalse(claim["mutation_performed"])
self.assertEqual(claim["blocker_kind"], mwi.BLOCKER_CONFLICTING_SESSIONS)
self.assertEqual(
claim["conflicting_owners"][0]["worker_identity"], first["identity"]
)
# The sanctioned recovery must never be "kill the other process".
self.assertIn("Do not kill", claim["exact_next_action"])
def test_contested_generation_fails_closed_in_assessment(self):
registry = _registry()
first = _attach(registry, client="codex", session="s1", generation="gen-shared")
_attach(registry, client="gemini", session="s2", generation="gen-shared")
verdict = mwi.assess_provenance(
registry=registry, worker_identity=first["identity"], env={}, now=NOW
)
self.assertEqual(verdict["session_ownership"], mwi.OWNERSHIP_CONTESTED)
self.assertTrue(verdict["fail_closed"])
self.assertEqual(verdict["blocker_kind"], mwi.BLOCKER_CONTRADICTORY)
self.assertTrue(verdict["conflicting_live_sessions"])
def test_identity_collision_is_refused_without_corrupting_existing(self):
"""AC31: never replace, adopt, merge with, or corrupt the incumbent."""
registry = _registry()
incumbent = _attach(registry, client="codex", session="s1", generation="gen-1")
before = registry.get(incumbent["identity"])
collided = registry.register(
worker_identity=incumbent["identity"],
client_name="gemini",
client_instance_id="inst-other",
session_id="s2",
generation_id="gen-2",
pid=9999,
now=NOW,
)
self.assertFalse(collided["registered"])
self.assertTrue(collided["collision"])
self.assertFalse(collided["mutation_performed"])
self.assertEqual(collided["blocker_kind"], mwi.BLOCKER_IDENTITY_COLLISION)
self.assertEqual(collided["collision_kind"], "active_worker")
self.assertEqual(
registry.get(incumbent["identity"]), before, "incumbent must be untouched"
)
def test_after_collision_a_regenerated_identity_registers(self):
"""AC32/AC35: forced collision, safe regeneration, successful replacement."""
registry = _registry()
fixed = dict(timestamp_ns=99, nonce="deterministic", now=NOW)
forced = mwi.generate_worker_identity("codex", "sess-collide", **fixed)
first = registry.register(
worker_identity=forced,
client_name="codex",
client_instance_id="inst-1",
session_id="sess-collide",
generation_id="gen-1",
now=NOW,
)
self.assertTrue(first["registered"])
# A second worker deriving the same inputs collides deterministically.
again = mwi.generate_worker_identity("codex", "sess-collide", **fixed)
self.assertEqual(again, forced)
self.assertTrue(
registry.register(
worker_identity=again,
client_name="codex",
client_instance_id="inst-2",
session_id="sess-collide",
generation_id="gen-2",
now=NOW,
)["collision"]
)
replacement = mwi.generate_worker_identity(
"codex", "sess-collide", timestamp_ns=100, nonce="different", now=NOW
)
self.assertNotEqual(replacement, forced)
self.assertTrue(
registry.register(
worker_identity=replacement,
client_name="codex",
client_instance_id="inst-2",
session_id="sess-collide",
generation_id="gen-2",
now=NOW,
)["registered"]
)
def test_restarted_worker_inherits_nothing(self):
"""AC33/AC34: a restart mints a new identity and no prior epoch."""
registry = _registry()
before = _attach(
registry, client="codex", session="sess-before", generation="gen-1", ttl=60
)
later = NOW + timedelta(hours=2)
after = _attach(
registry, client="codex", session="sess-after", generation="gen-2", now=later
)
self.assertNotEqual(before["identity"], after["identity"])
self.assertNotEqual(
registry.get(after["identity"])["generation_id"],
registry.get(before["identity"])["generation_id"],
)
class LivenessTests(unittest.TestCase):
def test_stale_session_record_is_not_live(self):
registry = _registry()
worker = _attach(registry, client="codex", session="s", generation="g", ttl=300)
stale = registry.is_live(
registry.get(worker["identity"]), now=NOW + timedelta(hours=1)
)
self.assertFalse(stale["live"])
self.assertFalse(stale["heartbeat_fresh"])
def test_liveness_is_not_pid_comparison_alone(self):
"""AC7: a live PID does not resurrect an expired registration."""
registry = _registry()
worker = _attach(registry, client="codex", session="s", generation="g", ttl=60)
verdict = registry.is_live(
registry.get(worker["identity"]),
now=NOW + timedelta(hours=1),
pid_alive=True,
)
self.assertFalse(
verdict["live"], "a live PID must not override a dead heartbeat"
)
def test_dead_pid_withdraws_liveness_from_a_fresh_heartbeat(self):
registry = _registry()
worker = _attach(registry, client="codex", session="s", generation="g")
verdict = registry.is_live(
registry.get(worker["identity"]), now=NOW, pid_alive=False
)
self.assertFalse(verdict["live"])
def test_stale_ownership_does_not_permanently_strand_a_daemon(self):
registry = _registry()
stranded = _attach(
registry, client="codex", session="s-old", generation="gen-daemon", ttl=60
)
later = NOW + timedelta(hours=3)
rescuer = _attach(
registry, client="claude", session="s-new", generation="gen-tmp", now=later
)
claim = registry.claim_generation(
worker_identity=rescuer["identity"],
generation_id="gen-daemon",
now=later,
)
self.assertTrue(claim["claimed"], claim["reasons"])
self.assertIn(stranded["identity"], claim["superseded_workers"])
class EvidenceTests(unittest.TestCase):
def test_env_flag_alone_does_not_prove_session_ownership(self):
verdict = mwi.assess_provenance(
registry=None,
worker_identity=None,
env={"GITEA_CLIENT_MANAGED": "1", "GITEA_MCP_SANCTIONED_DAEMON": "1"},
now=NOW,
)
self.assertFalse(verdict["session_owned"])
self.assertEqual(verdict["session_ownership"], mwi.OWNERSHIP_UNOWNED)
self.assertTrue(verdict["env_flag_only"])
self.assertTrue(verdict["fail_closed"])
self.assertNotIn(mwi.EVIDENCE_ATTACHMENT_RECORD, verdict["evidence"])
self.assertFalse(verdict["env_signal"]["proves_session_ownership"])
def test_env_flag_still_answers_the_launch_question(self):
"""The #686 wall is preserved: env decides launch, not ownership."""
self.assertTrue(
mwi.assess_launch_provenance({"GITEA_CLIENT_MANAGED": "1"})["client_managed"]
)
self.assertFalse(
mwi.assess_launch_provenance({"GITEA_CLIENT_MANAGED": "0"})["client_managed"]
)
self.assertFalse(
mwi.assess_launch_provenance({}, stdin_is_tty=True)["client_managed"]
)
self.assertTrue(
mwi.assess_launch_provenance({"GITEA_MCP_PROFILE": "prgs-author"})[
"client_managed"
]
)
def test_missing_evidence_is_unproven_not_manual(self):
"""A missing proof must not be reported as a hand-launched process."""
verdict = mwi.assess_provenance(
registry=None, worker_identity=None, env={}, now=NOW
)
self.assertEqual(verdict["provenance"], mwi.PROVENANCE_UNPROVEN)
self.assertNotEqual(verdict["provenance"], mwi.PROVENANCE_MANUAL)
self.assertTrue(verdict["fail_closed"])
def test_declared_manual_launch_is_reported_as_manual(self):
verdict = mwi.assess_provenance(
registry=None,
worker_identity=None,
env={"GITEA_CLIENT_MANAGED": "0"},
now=NOW,
)
self.assertEqual(verdict["provenance"], mwi.PROVENANCE_MANUAL)
def test_fail_closed_refusal_names_its_scope_not_the_profile(self):
"""AC17/AC41: no refusal is profile-wide."""
verdict = mwi.assess_provenance(
registry=None,
worker_identity=None,
env={},
profile="prgs-reviewer",
role="reviewer",
now=NOW,
)
self.assertFalse(verdict["scope"]["profile_wide"])
self.assertEqual(verdict["blocker_kind"], mwi.BLOCKER_NO_ATTACHMENT)
class CohortScopingTests(unittest.TestCase):
def test_shared_profile_with_distinct_identities_does_not_block(self):
"""AC40: profile is not a singleton identity."""
registry = _registry()
_attach(
registry,
client="codex",
session="s1",
generation="g1",
profile="prgs-reviewer",
)
_attach(
registry,
client="gemini",
session="s2",
generation="g2",
profile="prgs-reviewer",
)
cohort = mwi.classify_cohort(registry.list_workers(), now=NOW)
self.assertFalse(cohort["blocked"], cohort["reasons"])
self.assertEqual(cohort["blocker_kind"], mwi.BLOCKER_NONE)
self.assertIn("prgs-reviewer", cohort["shared_profiles"])
self.assertTrue(cohort["profile_sharing_permitted"])
self.assertEqual(cohort["blocked_worker_identities"], [])
def test_duplicate_cohort_records_block_only_the_offenders(self):
registry = _registry()
_attach(registry, client="codex", session="s1", generation="gen-contested")
_attach(registry, client="gemini", session="s2", generation="gen-contested")
_attach(registry, client="claude", session="s3", generation="gen-fine")
cohort = mwi.classify_cohort(registry.list_workers(), now=NOW)
self.assertTrue(cohort["blocked"])
self.assertEqual(cohort["contested_generations"], ["gen-contested"])
self.assertEqual(len(cohort["blocked_worker_identities"]), 2)
def test_mixed_runtime_generations_are_scoped_independently(self):
"""AC17: one stale generation does not wall unrelated healthy ones."""
registry = _registry()
stale = _attach(
registry, client="codex", session="s1", generation="gen-stale", ttl=60
)
healthy_a = _attach(registry, client="gemini", session="s2", generation="gen-a")
healthy_b = _attach(registry, client="claude", session="s3", generation="gen-b")
scoped = mwi.scope_runtime_failure(
failure_kind="stale-runtime",
worker_identity=stale["identity"],
profile="prgs-reviewer",
all_live_workers=registry.list_workers(),
)
self.assertFalse(scoped["profile_wide"])
self.assertFalse(scoped["fleet_wide"])
self.assertEqual(len(scoped["affected_workers"]), 1)
self.assertEqual(scoped["unaffected_worker_count"], 2)
unaffected = {w["worker_identity"] for w in scoped["unaffected_workers"]}
self.assertEqual(unaffected, {healthy_a["identity"], healthy_b["identity"]})
class HardcodedClientRegressionTests(unittest.TestCase):
def test_unknown_client_does_not_resolve_to_codex(self):
for name in ("gemini", "antigravity", "grok", "some_new_llm", "", None):
with self.subTest(client=name):
self.assertNotEqual(
mcp_client_reconnect.normalize_client(name),
"codex",
"an unidentified client must never be handed Codex UI steps",
)
def test_known_clients_still_get_their_own_steps(self):
self.assertEqual(mcp_client_reconnect.normalize_client("codex"), "codex")
self.assertEqual(
mcp_client_reconnect.normalize_client("claude_code"), "claude_code"
)
def test_generic_steps_do_not_name_a_specific_vendor(self):
steps = " ".join(mcp_client_reconnect.operator_ui_steps("gemini"))
self.assertNotIn("Codex", steps)
def test_reconnect_client_is_derived_from_the_attachment_record(self):
registry = _registry()
worker = _attach(registry, client="antigravity", session="s", generation="g")
verdict = mwi.assess_provenance(
registry=registry, worker_identity=worker["identity"], env={}, now=NOW
)
self.assertEqual(mwi.reconnect_client_for(verdict), "antigravity")
def test_reconnect_client_is_unknown_rather_than_guessed(self):
self.assertEqual(mwi.reconnect_client_for({}), mwi.UNKNOWN_CLIENT)
class RemoteBindingTests(unittest.TestCase):
def test_explicit_prgs_selection_is_honoured(self):
resolved = mwi.resolve_bound_remote(
requested_remote="prgs", bound_remote="prgs", default_remote="dadeschools"
)
self.assertEqual(resolved["remote"], "prgs")
self.assertFalse(resolved["drifted"])
def test_omitted_remote_uses_the_binding_not_the_library_default(self):
"""The reported dadeschools host drift."""
resolved = mwi.resolve_bound_remote(
requested_remote=None, bound_remote="prgs", default_remote="dadeschools"
)
self.assertEqual(resolved["remote"], "prgs")
self.assertNotEqual(resolved["remote"], "dadeschools")
self.assertEqual(resolved["resolved_from"], "session_binding")
def test_contradicting_the_binding_is_refused(self):
resolved = mwi.resolve_bound_remote(
requested_remote="dadeschools",
bound_remote="prgs",
default_remote="dadeschools",
)
self.assertEqual(resolved["remote"], "prgs")
self.assertTrue(resolved["drifted"])
self.assertFalse(resolved["honoured_request"])
def test_unbound_session_falls_back_and_says_so(self):
resolved = mwi.resolve_bound_remote(
requested_remote=None, bound_remote=None, default_remote="dadeschools"
)
self.assertEqual(resolved["remote"], "dadeschools")
self.assertEqual(resolved["resolved_from"], "library_default")
self.assertTrue(resolved["reasons"])
class SurfaceAgreementTests(unittest.TestCase):
"""The reproduced contradiction: two surfaces, one process, two answers."""
def test_namespace_health_and_direct_assessment_agree(self):
registry = _registry()
worker = _attach(
registry,
client="gemini",
session="sess-agree",
generation="gen-agree",
profile="prgs-reviewer",
)
env = {"GITEA_MCP_PROFILE": "prgs-reviewer", "GITEA_CLIENT_MANAGED": "1"}
direct = mwi.assess_provenance(
registry=registry,
worker_identity=worker["identity"],
env=env,
profile="prgs-reviewer",
)
health = mcp_namespace_health.classify_namespace_probe(
"gitea-reviewer",
configured=True,
registered_tools=["gitea_whoami"],
probe_result={"success": True},
probe_source="client_namespace",
process={"pid": 4242, "profile": "prgs-reviewer", "env": env},
registry=registry,
worker_identity=worker["identity"],
)
self.assertEqual(health["provenance"], direct["provenance"])
self.assertEqual(health["is_client_managed"], direct["is_client_managed"])
self.assertEqual(health["worker_identity"], direct["worker_identity"])
self.assertEqual(health["session_id"], "sess-agree")
self.assertEqual(health["client_name"], "gemini")
def test_namespace_health_can_report_client_managed_at_all(self):
"""The old derivation was structurally incapable of this."""
env = {"GITEA_CLIENT_MANAGED": "1", "GITEA_MCP_PROFILE": "prgs-author"}
health = mcp_namespace_health.classify_namespace_probe(
"gitea-author",
configured=True,
registered_tools=["gitea_whoami"],
probe_result={"success": True},
probe_source="client_namespace",
process={"pid": 1234, "profile": "prgs-author", "env": env},
)
self.assertTrue(
health["is_client_managed"],
"a client-managed launch must be reportable as client-managed",
)
def test_namespace_health_without_attachment_fails_closed(self):
health = mcp_namespace_health.classify_namespace_probe(
"gitea-author",
configured=True,
registered_tools=["gitea_whoami"],
probe_result={"success": True},
probe_source="client_namespace",
process={"pid": 1234, "profile": "prgs-author", "env": {}},
)
self.assertTrue(health["provenance_fail_closed"])
self.assertEqual(health["provenance"], mwi.PROVENANCE_UNPROVEN)
self.assertIsNone(health["session_id"])
def test_no_false_reconnect_loop_for_an_owned_session(self):
"""A proven owner must not be told to reconnect."""
registry = _registry()
worker = _attach(registry, client="claude", session="s", generation="g")
verdict = mwi.assess_provenance(
registry=registry, worker_identity=worker["identity"], env={}, now=NOW
)
self.assertFalse(verdict["fail_closed"])
self.assertEqual(verdict["blocker_kind"], mwi.BLOCKER_NONE)
self.assertEqual(verdict["reasons"], [])
if __name__ == "__main__":
unittest.main()
@@ -1,243 +0,0 @@
"""Tests for issue #970: safely resolve worktree bindings whose paths are missing.
Covers all acceptance criteria for #970:
- missing path reporting with repo/branch/issue/session/lease/host correlation
- fail closed on unavailable host or mount failure
- fail closed on moved worktree paths
- fail closed on live lease or active session/process
- atomic pre-mutation re-validation preventing recreation races
- targeted retirement preserving unrelated worktrees & metadata
- safe and idempotent repeated execution
- worktrees dimension resolution post-cleanup
- resolution of the two historical post-PR #968 records (#795 / #635)
"""
from __future__ import annotations
import json
import os
import shutil
import tempfile
import unittest
import control_plane_db as cpd
import lease_lifecycle
import missing_worktree_reconcile as mwr
class TestMissingWorktreeReconcile(unittest.TestCase):
def setUp(self):
self.temp_dir = tempfile.mkdtemp()
self.db_path = os.path.join(self.temp_dir, "control_plane.sqlite3")
self.db = cpd.ControlPlaneDB(self.db_path)
self.project_root = os.path.join(self.temp_dir, "repo")
os.makedirs(os.path.join(self.project_root, "branches"), exist_ok=True)
# Initialize git repo in project_root for git rev-parse checks
os.system(f"git -C '{self.project_root}' init -q")
def tearDown(self):
shutil.rmtree(self.temp_dir, ignore_errors=True)
def test_audit_missing_worktree_bindings_reports_associations(self):
res = self.db.assign_and_lease(
remote="prgs",
org="Scaled-Tech-Consulting",
repo="Gitea-Tools",
kind="pr",
number=795,
expected_head_sha="head795",
role="reviewer",
session_id="reviewer-pr795-test-session",
worktree_path="/tmp/nonexistent_wt_path_795",
)
lease_id = res.lease_id
self.assertIsNotNone(lease_id)
# Release the lease (dead/released)
self.db.release_lease_recorded(lease_id, session_id="reviewer-pr795-test-session")
audit = mwr.audit_missing_worktree_bindings(
self.db,
project_root=self.project_root,
remote="prgs",
org="Scaled-Tech-Consulting",
repo="Gitea-Tools",
)
self.assertTrue(audit["host_mount_healthy"])
self.assertEqual(audit["missing_count"], 1)
self.assertEqual(audit["confirmed_stale_count"], 1)
candidate = audit["confirmed_stale_candidates"][0]
self.assertEqual(candidate["lease_id"], lease_id)
self.assertEqual(candidate["work_number"], 795)
self.assertEqual(candidate["session_id"], "reviewer-pr795-test-session")
self.assertEqual(candidate["classification"], mwr.CLASS_CONFIRMED_STALE_DELETED)
self.assertTrue(candidate["retire_eligible"])
def test_unavailable_host_mount_blocks_retirement(self):
# Point to a non-existent project_root
bad_root = os.path.join(self.temp_dir, "nonexistent_repo_dir")
audit = mwr.audit_missing_worktree_bindings(
self.db,
project_root=bad_root,
remote="prgs",
)
self.assertFalse(audit["host_mount_healthy"])
res = self.db.assign_and_lease(
remote="prgs", org="org", repo="repo", kind="issue", number=100,
role="author", session_id="sess-100", worktree_path="/tmp/missing_100",
)
self.db.release_lease_recorded(res.lease_id, session_id="sess-100")
audit2 = mwr.audit_missing_worktree_bindings(self.db, project_root=bad_root, remote="prgs")
self.assertEqual(audit2["unavailable_host_count"], 1)
self.assertEqual(audit2["confirmed_stale_count"], 0)
self.assertFalse(audit2["missing_bindings"][0]["retire_eligible"])
def test_live_lease_and_live_session_block_retirement(self):
res = self.db.assign_and_lease(
remote="prgs", org="org", repo="repo", kind="issue", number=200,
role="author", session_id="sess-live-200", worktree_path="/tmp/missing_live_200",
owner_pid=os.getpid(),
)
audit = mwr.audit_missing_worktree_bindings(self.db, project_root=self.project_root, remote="prgs")
self.assertEqual(audit["live_protected_count"], 1)
self.assertEqual(audit["confirmed_stale_count"], 0)
# Attempting resolution on live protected binding is refused
c = audit["missing_bindings"][0]
res_mut = mwr.resolve_missing_worktree_binding(self.db, binding=c, dry_run=False, operator_authorized=True, project_root=self.project_root)
self.assertFalse(res_mut["success"])
self.assertEqual(res_mut["reason"], "revalidation_failed")
def test_moved_worktree_path_blocks_retirement(self):
# Create a worktree under branches/ to simulate moved path
moved_dir = os.path.join(self.project_root, "branches", "issue-635-project-registry-api")
os.makedirs(moved_dir, exist_ok=True)
res = self.db.assign_and_lease(
remote="prgs", org="org", repo="repo", kind="issue", number=635,
role="author", session_id="sess-635", worktree_path="/tmp/old_missing_path_635",
)
self.db.release_lease_recorded(res.lease_id, session_id="sess-635")
cls_info = mwr.classify_worktree_binding(
recorded_path="/tmp/old_missing_path_635",
branch="issue-635-project-registry-api",
project_root=self.project_root,
)
self.assertEqual(cls_info["classification"], mwr.CLASS_MOVED_WORKTREE)
self.assertFalse(cls_info["retire_eligible"])
self.assertEqual(os.path.realpath(cls_info["moved_to_path"]), os.path.realpath(moved_dir))
def test_recreation_race_revalidation_blocks_mutation(self):
path = os.path.join(self.temp_dir, "recreated_wt")
res = self.db.assign_and_lease(
remote="prgs", org="org", repo="repo", kind="issue", number=300,
role="author", session_id="sess-300", worktree_path=path,
)
self.db.release_lease_recorded(res.lease_id, session_id="sess-300")
# Initial audit when path is missing
audit = mwr.audit_missing_worktree_bindings(self.db, project_root=self.project_root, remote="prgs")
candidate = audit["confirmed_stale_candidates"][0]
# Simulate concurrent recreation of worktree path
os.makedirs(path, exist_ok=True)
# Pre-mutation revalidation detects recreation and blocks
res_mut = mwr.resolve_missing_worktree_binding(
self.db,
binding=candidate,
dry_run=False,
operator_authorized=True,
project_root=self.project_root,
)
self.assertFalse(res_mut["success"])
self.assertEqual(res_mut["reason"], "revalidation_failed")
def test_authorized_cleanup_retires_stale_binding_and_is_idempotent(self):
# Create 2 leases: 1 stale missing, 1 valid existing worktree
valid_path = os.path.join(self.project_root, "branches", "valid_wt")
os.makedirs(valid_path, exist_ok=True)
stale_path = "/tmp/stale_missing_400"
res1 = self.db.assign_and_lease(
remote="prgs", org="org", repo="repo", kind="issue", number=401,
role="author", session_id="sess-401", worktree_path=valid_path,
)
res2 = self.db.assign_and_lease(
remote="prgs", org="org", repo="repo", kind="issue", number=402,
role="author", session_id="sess-402", worktree_path=stale_path,
)
self.db.release_lease_recorded(res2.lease_id, session_id="sess-402")
# Dry-run reconciliation
dry = mwr.reconcile_missing_worktree_bindings(
self.db, project_root=self.project_root, remote="prgs", dry_run=True, operator_authorized=True
)
self.assertTrue(dry["success"])
self.assertTrue(dry["dry_run"])
self.assertEqual(dry["stale_candidates_count"], 1)
# Apply reconciliation
applied = mwr.reconcile_missing_worktree_bindings(
self.db, project_root=self.project_root, remote="prgs", dry_run=False, operator_authorized=True
)
self.assertTrue(applied["success"])
self.assertFalse(applied["dry_run"])
self.assertTrue(applied["worktrees_dimension_resolved"])
self.assertEqual(applied["after_audit"]["missing_count"], 0)
# Verify DB: lease 2 worktree_path is cleared, lease 1 worktree_path remains intact
l1 = self.db.get_lease_workflow_state(res1.lease_id)
self.assertEqual(l1["lease"]["worktree_path"], valid_path)
l2 = self.db.get_lease_workflow_state(res2.lease_id)
self.assertEqual(l2["lease"]["worktree_path"], "")
prov = json.loads(l2["lease"]["provenance_json"])
self.assertTrue(prov.get("worktree_path_retired"))
self.assertEqual(prov.get("retired_worktree_path"), stale_path)
# Re-run (idempotency check)
rerun = mwr.reconcile_missing_worktree_bindings(
self.db, project_root=self.project_root, remote="prgs", dry_run=False, operator_authorized=True
)
self.assertTrue(rerun["success"])
self.assertTrue(rerun["worktrees_dimension_resolved"])
self.assertEqual(rerun["stale_candidates_count"], 0)
def test_historical_post_pr968_records_resolution(self):
# Create historical records matching #970 description
res_795 = self.db.assign_and_lease(
remote="prgs", org="Scaled-Tech-Consulting", repo="Gitea-Tools", kind="pr", number=795,
expected_head_sha="head795hist", role="reviewer", session_id="reviewer-pr795-hist",
worktree_path="/Users/jasonwalker/Development/Gitea-Tools/branches/review-feat-issue-628-autonomous-handoffs-orchestration",
)
self.db.release_lease_recorded(res_795.lease_id, session_id="reviewer-pr795-hist")
res_635 = self.db.assign_and_lease(
remote="prgs", org="Scaled-Tech-Consulting", repo="Gitea-Tools", kind="issue", number=635,
role="author", session_id="author-issue635-hist",
worktree_path="/Users/jasonwalker/Development/Gitea-Tools/branches/issue-635-project-registry-api",
)
self.db.release_lease_recorded(res_635.lease_id, session_id="author-issue635-hist")
# Audit finds both historical records
audit = mwr.audit_missing_worktree_bindings(self.db, project_root=self.project_root, remote="prgs")
self.assertEqual(audit["confirmed_stale_count"], 2)
# Reconcile safely resolves both
rec = mwr.reconcile_missing_worktree_bindings(
self.db, project_root=self.project_root, remote="prgs", dry_run=False, operator_authorized=True
)
self.assertTrue(rec["worktrees_dimension_resolved"])
self.assertEqual(rec["after_audit"]["missing_count"], 0)
if __name__ == "__main__":
unittest.main()