feat(restart): audit lifecycle events and durable incidents (#665)

Add restart_audit with mcp.restart.* event schema, redacted emission via
gitea_audit, correlation ids, and incident materialization for failed drain
and break-glass. Wire gitea_request_mcp_restart to always audit impact
previews and fail closed on privileged apply when the audit sink is enabled
but write fails.

Closes #665

allocator_service.py

@@ -133,10 +133,15 @@ ROLE_ACTIONS: dict[str, tuple[tuple[str, ...], tuple[str, ...]]] = {
 
 
 # Body phrases that prove an issue is an implementation container, not a
-# unit of direct author work (#844). Matched case-insensitively against the
-# issue body. Title alone is never sufficient (ordinary issues may mention
-# "epic" incidentally).
+# unit of direct author work (#844 / #854). Matched case-insensitively against
+# the issue body. Title alone is never sufficient (ordinary issues may mention
+# "epic", "roadmap", "vision", or "umbrella" incidentally).
+#
+# #854 extends the #844 marker set so product-vision (#652), phased-roadmap
+# (#653), and umbrella (#655) coordination records — which do not use the word
+# "epic" — are classified with the same semantic exclusion as epic containers.
 _CHILD_ONLY_BODY_MARKERS: tuple[str, ...] = (
+    # Epic / child-only (#844, live #631)
     "implementation is delivered via child issues only",
     "implementation is delivered through child issues only",
     "implementation is delivered via child issues",
@@ -150,9 +155,26 @@ _CHILD_ONLY_BODY_MARKERS: tuple[str, ...] = (
     "coordination container",
     "child-only container",
     "implementation is delegated to child",
+    # Vision / roadmap / umbrella coordination (#854, live #652/#653/#655).
+    # Prefer authoritative non-implementation / child-only scope language over
+    # bare words like "roadmap" so ordinary implementable issues that mention
+    # a parent vision or roadmap stay eligible.
+    "do not implement features on this issue",
+    "implementing features on this roadmap issue",
+    "implementation is via linked children only",
+    "no product feature claimed complete on this issue alone",
+    "phased delivery roadmap and epic sequencing",
+    "this issue is the enduring source of truth",
+    "enduring source of truth for the",
+    "canonical product vision — enduring source of truth",
+    "canonical product vision - enduring source of truth",
+    "state: vision-active",
+    "state: roadmap-active",
 )
 
-# Explicit epic / umbrella labels (structured evidence preferred over title).
+# Explicit epic / umbrella / vision / roadmap labels (structured evidence
+# preferred over title). Tracker alone is *not* included — ordinary issues
+# may carry a tracker label without being non-implementable containers.
 _EPIC_LABELS: frozenset[str] = frozenset(
     {
         "type:epic",
@@ -161,6 +183,16 @@ _EPIC_LABELS: frozenset[str] = frozenset(
         "scope:epic",
         "type:umbrella",
         "umbrella",
+        "kind:umbrella",
+        "scope:umbrella",
+        "type:vision",
+        "vision",
+        "kind:vision",
+        "scope:vision",
+        "type:roadmap",
+        "roadmap",
+        "kind:roadmap",
+        "scope:roadmap",
     }
 )
 
@@ -222,16 +254,45 @@ class WorkCandidate:
         }
 
 
+def _title_container_prefix(title_l: str) -> str | None:
+    """Return a coordination-title prefix token if *title_l* uses one (#854).
+
+    Title prefixes alone never exclude; they only corroborate body/label
+    evidence. Ordinary issues may say "roadmap" or "vision" mid-title.
+    """
+    for prefix, token in (
+        ("epic:", "title_epic_prefix"),
+        ("epic ", "title_epic_prefix"),
+        ("umbrella:", "title_umbrella_prefix"),
+        ("umbrella ", "title_umbrella_prefix"),
+        ("roadmap:", "title_roadmap_prefix"),
+        ("roadmap ", "title_roadmap_prefix"),
+        ("product vision:", "title_vision_prefix"),
+        ("product vision ", "title_vision_prefix"),
+        ("vision:", "title_vision_prefix"),
+        ("vision ", "title_vision_prefix"),
+    ):
+        if title_l.startswith(prefix):
+            return token
+    return None
+
+
 def classify_epic_or_child_only_container(
     c: WorkCandidate,
 ) -> tuple[bool, str | None]:
-    """Return whether *c* is an epic / child-only implementation container (#844).
+    """Return whether *c* is a non-implementable coordination container (#844/#854).
 
     Exclusion uses structured evidence first (labels, body scope language).
-    A bare title containing the word "epic" is **not** enough — ordinary
-    implementable issues may mention epics incidentally. A title that is
-    explicitly prefixed ``Epic:`` only counts when the body also proves
-    child-only / no-direct-implementation scope (or an epic label is present).
+    A bare title containing the words "epic", "roadmap", "vision", or
+    "umbrella" is **not** enough — ordinary implementable issues may mention
+    those terms incidentally. Explicit title prefixes (``Epic:``, ``Roadmap:``,
+    ``Product vision:``, ``Umbrella:``) only count when the body also proves
+    child-only / no-direct-implementation scope (or a container label is
+    present).
+
+    Covers epic, product-vision, phased-roadmap, umbrella, and child-only
+    records so the allocator never assigns coordination containers as direct
+    author work.
 
     PRs are never classified as containers here (they already have a head).
     """
@@ -245,25 +306,28 @@ def classify_epic_or_child_only_container(
     title_l = title.lower()
 
     body_hits = [m for m in _CHILD_ONLY_BODY_MARKERS if m in body_l]
-    title_epic_prefix = title_l.startswith("epic:") or title_l.startswith("epic ")
+    title_prefix = _title_container_prefix(title_l)
 
     if epic_label:
         detail = f"label={epic_label[0]}"
         if body_hits:
             detail = f"{detail}; body_marker={body_hits[0]!r}"
+        if title_prefix:
+            detail = f"{title_prefix}; {detail}"
         return True, detail
 
     if body_hits:
-        # Body proves child-only / umbrella scope. Title "Epic:" is corroborating
-        # but not required — containers without the word still exclude.
+        # Body proves child-only / vision / roadmap / umbrella scope. Title
+        # prefixes are corroborating but not required — containers without the
+        # title word still exclude.
         detail = f"body_marker={body_hits[0]!r}"
-        if title_epic_prefix:
-            detail = f"title_epic_prefix; {detail}"
+        if title_prefix:
+            detail = f"{title_prefix}; {detail}"
         return True, detail
 
-    # Title-only "Epic:" without body scope evidence is insufficient (#844 AC:
-    # eligibility does not rely solely on the word "Epic" in a title).
-    # Similarly, incidental "epic" mid-title without markers stays eligible.
+    # Title-only coordination prefix without body scope evidence is
+    # insufficient (#844/#854 AC: eligibility does not rely solely on a title
+    # word). Incidental mid-title mentions without markers stay eligible.
     return False, None
 
 

docs/mcp-restart-audit.md

@@ -0,0 +1,93 @@
+# MCP restart audit events and incidents (#665)
+
+Restarts and recovery attempts leave a forensic trail. Failed drains and
+break-glass paths also raise durable Gitea incident issues so unsafe restarts
+cannot be silently repeated.
+
+Parent umbrella: **#655**. Related: impact coordinator **#658**, drain proof
+**#661**, restart classes **#663**, break-glass **#664**, post-restart reconcile
+**#662**, vision **#652**, roadmap **#653**, console recovery **#642**.
+
+## Components
+
+| Piece | Where | Responsibility |
+|-------|-------|----------------|
+| Event schema + emission | `restart_audit.py` | `mcp.restart.*` vocabulary, redacted payload builder, append-only sink via `gitea_audit` |
+| Fail-closed privileged gate | `restart_audit.require_audit_or_deny` | When `GITEA_AUDIT_LOG` is set and the write fails, privileged apply is denied |
+| Incident descriptors | `restart_audit.build_incident_descriptor` | Durable follow-up issues (failed drain, break-glass, reconcile unresolved, unguarded) |
+| Materializer | `restart_audit.materialize_incident` | Injected `create_issue_fn` (network kept out of pure tests) |
+| Wiring | `gitea_request_mcp_restart` | Correlation id, impact-preview audit, apply-gate / break-glass audit + incident creation |
+
+## Event vocabulary
+
+| Event type | When |
+|------------|------|
+| `mcp.restart.impact_preview` | Every `gitea_request_mcp_restart` evaluation |
+| `mcp.restart.drain_enter` | Drain window starts (schema reserved; emit from drain path) |
+| `mcp.restart.drain_exit` | Drain window ends |
+| `mcp.restart.drain_proof` | Drain-proof verification result |
+| `mcp.restart.apply_gate` | Apply hard gate (`dry_run=False`) |
+| `mcp.restart.break_glass` | Authorized break-glass bypass |
+| `mcp.restart.post_restart_reconcile` | Post-restart reconcile outcome |
+| `mcp.restart.narrower_recovery` | Narrower recovery attempt recorded |
+| `mcp.restart.unguarded_detected` | Unguarded restart path detected |
+
+All free text is redacted before sink write or issue body assembly. Emission
+never raises; callers decide fail-closed policy.
+
+## Correlation
+
+Each restart lifecycle mints a short `correlation_id` (`rst-` + 16 hex) shared
+across impact preview → apply gate → incident descriptors so operators can join
+the trail.
+
+## Privileged deny-on-audit-fail
+
+Rollout policy (issue #665):
+
+1. Configure `GITEA_AUDIT_LOG` so writes land.
+2. Only then enforce deny when a privileged restart path cannot audit.
+
+When audit is **not** configured, privileged apply still proceeds (no false
+denials during rollout). When audit **is** configured and the write fails,
+`apply_authorized` is cleared.
+
+## Incidents
+
+| Kind | Trigger |
+|------|---------|
+| `restart_failed_drain` | Apply denied by drain hard gate / failed proof |
+| `restart_break_glass` | Any authorized break-glass apply |
+| `restart_reconcile_unresolved` | Post-restart reconcile left work unresolved |
+| `restart_unguarded_detected` | Unguarded restart attempt detected |
+
+Break-glass **always** creates an incident descriptor (and a Gitea issue when
+the create path is available). Failed drain does the same. Incident bodies
+include correlation id, session, class, scope, proof id, and redacted reasons.
+
+Default labels: `mcp-health`, `safety`, `observability`, `status:ready`,
+`type:bug`, `workflow-hardening`.
+
+## Tool payload surface
+
+`gitea_request_mcp_restart` returns:
+
+* `correlation_id` — lifecycle join key
+* `restart_audit.impact_preview_written` — sink success for the preview event
+* `restart_audit.apply_gate_written` — sink success for apply/break-glass (apply only)
+* `restart_audit.incident_result` — materialization outcome when an incident was required
+* `incident` — durable descriptor (when gate requires follow-up)
+
+## Security
+
+* No secrets in audit payloads or issue bodies.
+* This module never restarts a process.
+* Drain proof verification remains #661; audit only records the decision.
+* Incident creation failures are recorded in `incident_result.reasons` and never
+  crash the restart evaluation path (audit write failure still fails closed for
+  privileged apply when the sink is enabled).
+
+## Tests
+
+See `tests/test_restart_audit.py`: schema, redaction, emission, deny policy,
+incident materialization mocks, break-glass / failed-drain selection.

docs/mcp-restart-classes.md

@@ -0,0 +1,53 @@
+# MCP restart classes and blast-radius permissions (#663)
+
+This is the machine-enforced class matrix used by
+`restart_coordinator.RESTART_CLASS_POLICIES`. It implements the narrower-first
+recovery ladder from #655 and the authorization policy from #656, using the
+path inventory from #657 and the impact coordinator from #658. Product and
+delivery lineage: vision #652 and roadmap #653.
+
+Unknown class names are denied. The coordinator requires both the class
+permission and an eligible request role. Approval gates are additional: a
+caller cannot turn a request permission into execution authority.
+
+| Restart class | Required permission | Expected blast radius | Drain requirement | Approval requirement | Audit requirement | Recovery behavior |
+|---|---|---|---|---|---|---|
+| `client_reconnect` | `mcp.reconnect.client` | none | none | self service | class, actor, client namespace, reason, outcome | Reconnect only the caller's client transport. No daemon or peer work changes. |
+| `session_reconnect` | `mcp.reconnect.session` | low | requesting-session safe point | self service | class, actor, session, reason, outcome | Rebind identity, capability, and workspace state for one session. |
+| `worker_restart` | `mcp.restart.worker.request` | low | target worker | controller approval + automated gates | class, actor, worker, approval, scoped drain, outcome | Restart one worker after its own leases and mutations drain. |
+| `role_runtime_restart` | `mcp.restart.role_runtime.request` | medium | target role runtime | controller approval + automated gates | class, actor, role namespace, approval, scoped drain, outcome | Restart and re-probe one role runtime; unrelated roles remain available. |
+| `connector_restart` | `mcp.restart.connector.request` | medium | target connector | controller approval + automated gates | class, actor, connector, approval, scoped drain, outcome | Restart one connector while unrelated runtimes remain available. |
+| `configuration_reload` | `mcp.reload.configuration.request` | low | mutation quiesce | controller approval + automated gates | class, actor, configuration revision, approval, outcome | Gracefully reload configuration without replacing the daemon. |
+| `rolling_mcp_restart` | `mcp.restart.rolling.request` | medium | one instance at a time | controller approval + automated gates | class, actor, instance order, approval, per-instance drains, outcome | Drain, restart, verify, and restore each instance before advancing. |
+| `full_mcp_restart` | `mcp.restart.full.request` | high | all sessions and mutations | controller approval + automated gates | class, actor, full impact, approval, full drain proof, outcome | Replace the complete MCP runtime only after a verified full drain. |
+| `host_restart` | `mcp.restart.host.request` | high | all host work | controller approval + infrastructure operator | class, actor, host/change or incident id, approval, full drain proof, outcome | Hand off to infrastructure ownership and reconcile every runtime afterward. |
+
+## Drain boundary
+
+Only `full_mcp_restart` and `host_restart` set `full_drain_required=true`.
+Reconnects and configuration reloads do not disrupt peer sessions. Worker,
+role-runtime, and connector restarts evaluate only their explicitly named
+target. Rolling restart drains one instance at a time. Missing required target
+scope denies the request rather than silently widening it to a full restart.
+
+## Permission and approval boundary
+
+Author, reviewer, merger, and reconciler roles may self-request reconnects and
+request scoped worker/role/connector/reload recovery. They cannot request
+rolling, full, or host restart classes. Controller/operator/admin roles may
+request the broader classes, while execution remains operator/admin-owned.
+Controller approval is independently required for every class above a session
+reconnect. Host restart additionally requires infrastructure-operator proof.
+
+The MCP request tool derives class permissions from its authenticated runtime
+role. It does not accept caller-supplied permissions. Controller and operator
+authorization are read from the already-running daemon environment, never
+from a request argument.
+
+## Audit and failure behavior
+
+Every impact audit and every console restart/reload audit includes a
+`restart_class` field. The impact audit also includes the exact
+`required_permission`. Unknown classes, missing permissions, ineligible roles,
+missing approval, missing scoped targets, and incomplete inventory all deny
+fail closed. Manual process kills remain forbidden and contaminating (#630).

docs/mcp-restart-coordinator.md

@@ -6,11 +6,23 @@ console (#642 / #652) can see the blast radius *before* concurrent LLM work is
 disrupted. Uncoordinated restarts destroy in-flight author/reviewer/merger work
 and give operators no way to see what they are about to break.
 
-This lands the coordinator + impact DTO + a dry-run MCP tool. It is the single
+This lands the coordinator + impact DTO + the MCP tool. It is the single
 sanctioned entry point for restart evaluation post-#657 (which inventoried the
-restart/reload/kill paths). The **mutative apply** path — actually performing a
-restart — is a later child gated by a drain proof and is explicitly out of
-scope here.
+restart/reload/kill paths).
+
+The **drain-proof hard gate now executes inside this tool** (#661, via PR #882):
+an apply request (`dry_run=False`) is evaluated against a drain proof here and
+denied when that proof is missing, expired, unclean, tampered with, or stale.
+It is no longer a separate child operation. What remains a later child is only
+the **execution** step — actually stopping and restoring a process. This tool
+still never restarts anything: `apply_supported` is always `false` and
+`restart_performed` is always `false`.
+
+The coordinator now routes every request through the restart-class policy
+matrix defined for #663. See
+[`mcp-restart-classes.md`](./mcp-restart-classes.md) for permissions, expected
+blast radius, scoped drain and approval requirements, audit fields, and
+recovery behavior for all nine classes.
 
 ## Components
 
@@ -19,7 +31,9 @@ scope here.
 | `restart_coordinator.evaluate_restart_impact` | `restart_coordinator.py` | Pure classification: inventory → impact report DTO. No I/O, no restart. |
 | `RestartImpactReport` / `SessionImpact` / `LeaseImpact` | `restart_coordinator.py` | Console-facing DTO (`.as_dict()` is JSON-serializable). |
 | `ControlPlaneDB.list_sessions` | `control_plane_db.py` | Read-only session inventory (the process-level unit a restart kills). |
-| `gitea_request_mcp_restart` | `gitea_mcp_server.py` | MCP tool: gathers inventory from the #613 DB, calls the coordinator, returns the report. Dry-run only. |
+| `gitea_request_mcp_restart` | `gitea_mcp_server.py` | MCP tool: gathers inventory from the #613 DB, calls the coordinator, returns the report, and on `dry_run=False` runs the #661 drain-proof hard gate. Never restarts a process. |
+| `drain_proof.gate_apply_restart` | `drain_proof.py` | The #661 hard gate: verifies a drain proof against the current impact fingerprint, or records an authorized break-glass bypass. |
+| `restart_audit` | `restart_audit.py` | #665 forensic trail: `mcp.restart.*` events via `gitea_audit`, correlation ids, durable incidents for failed drain / break-glass. See [`mcp-restart-audit.md`](./mcp-restart-audit.md). |
 
 ## Dimensions evaluated
 
@@ -77,17 +91,61 @@ authorization is present.
 ```text
 gitea_request_mcp_restart(remote, host, org, repo,
                           dry_run=True, request_override=False,
-                          session_id=None, limit=200)
+                          session_id=None, limit=200,
+                          restart_class="full_mcp_restart",
+                          target_session_id=None, target_role=None,
+                          target_connector=None,
+                          drain_proof_json=None,
+                          request_break_glass=False)
 ```
 
-Read-only, dry-run, and it **never restarts anything**. `apply_supported` is
-always `false`; passing `dry_run=False` performs no restart and reports that
-apply is gated by a drain proof (a separate child).
+It **never restarts anything**: `apply_supported` is always `false` and
+`restart_performed` is always `false`.
+
+### Dry-run versus apply
+
+| Call | Behavior |
+|------|----------|
+| `dry_run=True` (default) | Read-only impact preview. No drain proof is required or consulted. |
+| `dry_run=False` | The #661 drain-proof hard gate runs **in this tool**. The outcome is reported under `apply_gate` / `apply_authorized`; a denial also returns a durable `incident` descriptor. Still no restart. |
+
+### Authorization ordering
+
+An apply requires **both** authorizations, and they are independent:
+
+1. **Restart-class authorization** (#663) — the requester's role and permissions
+   must allow the requested class, the class's approval requirement must be
+   satisfied, and any target-scoped class must name its target. Failing any of
+   these makes `allow_restart` `false`.
+2. **Drain-proof gate** (#661) — a valid, unexpired, clean proof bound to the
+   current impact fingerprint, or an authorized break-glass.
+
+`apply_authorized` is the conjunction: `gate.allow and allow_restart`. A clean
+drain proof therefore cannot override a class or requester-role denial, and a
+denied class never reports an authorized apply. `apply_gate` carries
+`drain_gate_allow` and `restart_class_authorized` so a denial is attributable to
+the authorization that produced it.
+
+### Break-glass
+
+Break-glass bypasses the **drain proof only** — never the restart-class matrix.
+It is honoured solely when `request_break_glass` is set *and* the environment
+carries `GITEA_BREAKGLASS_RESTART_AUTHORIZATION`; like operator override, the
+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.
+
+### Fail closed on apply
+
+A missing, malformed, expired, unclean, tampered, or fingerprint-stale drain
+proof denies the apply and returns an `incident` descriptor. An unknown restart
+class denies before any of this. Ambiguity always denies.
 
 ## Audit
 
 Every evaluation carries an `audit_record` (event, coordinator version, verdict,
-allow decision, blast radius, counts, timestamp) so restart decisions are
+restart class, required permission, allow decision, blast radius, counts,
+timestamp) so restart decisions are
 auditable. No secrets flow through the coordinator — session ids, pids, and
 profiles are operational metadata only.
 

docs/webui-local-dev.md

@@ -57,6 +57,8 @@ status, onboarding checklist state, and the fail-closed error payloads (#635).
 | `/system-health` | System-health dashboard — readiness, version/uptime, dependencies, MCP namespaces, stale-runtime parity (#639) |
 | `/queue` | Live PR and issue queue dashboard (#429) |
 | `/api/queue` | JSON queue export with pagination metadata |
+| `/traffic` | Workflow traffic-control view — runnable, leased, blocked, needs-controller, terminal-complete (#640) |
+| `/api/traffic` | JSON traffic-control export with state classifications and next safe role actions |
 | `/projects` | Project registry list with status and onboarding progress (#427, #635) |
 | `/projects/{id}` | Project detail + onboarding checklist |
 | `/api/v1/projects` | Versioned JSON registry export (#635) |
@@ -75,7 +77,9 @@ status, onboarding checklist state, and the fail-closed error payloads (#635).
 | `/api/actions/{id}/preview` | Mutation ledger preview (GET, read-only) |
 | `/leases` | Lease and collision visibility (#433) |
 | `/api/leases` | JSON lease/collision export |
-| `/sessions` | Phase 1 shell stub — session inventory (backed by #636) |
+| `/sessions` | Runtime and session view (#641) — health + inventory sessions/namespaces/worktrees |
+| `/api/sessions` | JSON export for the runtime/session view |
+| `/api/v1/sessions` | Versioned alias of `/api/sessions` |
 | `/inventory` | Phase 1 shell stub — unified inventory (backed by #636) |
 | `/timeline` | Phase 1 shell stub — workflow event timeline |
 | `/policy` | Phase 1 shell stub — capability/role policy placeholder |
@@ -85,6 +89,35 @@ Most routes are GET-only. POST/PUT/PATCH/DELETE return `405` with
 `read-only-mvp`, except `/audit` and `/api/audit` which accept POST for
 local validator preview only (no Gitea mutations, no server-side storage).
 
+### Traffic-control state vocabulary (#640)
+
+The traffic view classifies each open issue/PR into exactly one bucket:
+
+| Bucket | Meaning | Operator implication |
+|--------|---------|----------------------|
+| **runnable** | No active lease, no block reason, safe for its expected role | Next role may start work |
+| **leased** | Active author claim or reviewer PR lease | Do not stomp; wait or adopt via role tools |
+| **blocked** | Dependency, missing head pin, conflict, or unmet dependency | Author remediation first |
+| **needs_controller** | Contaminated, controller-only diagnosis, or `status:blocked` | Controller only |
+| **terminal_complete** | Reconciler / terminal-lock territory | Reconciler cleanup path |
+
+`status:blocked` items route to **needs_controller**, not **blocked**:
+`expected_role_for_candidate` sends them to the controller, and the blocker
+reason renders in either bucket.
+
+**Live path contracts (do not invent):**
+
+- PR head pins come from `QueueItem.signals["head_sha"]` (full SHA). Display
+  `extra["head_sha"]` is truncated and must never be used for routing.
+- Reviewer leases are keyed as `(pr, pr_number)` only — never via a linked
+  `issue_number` on the same lease marker.
+- Issue claims come from `claim_inventory["entries"]`
+  (`issue_claim_heartbeat.build_claim_inventory`). There is no `active_claims`
+  key.
+- Queue display badges are only: `blocked`, `claimed`, `duplicate`, `stale`,
+  `in-review`, `open`. Review verdicts (`request-changes`, `approved`) are
+  **not** queue badges; traffic does not invent them from the queue loader.
+
 ## System health API (#634)
 
 `GET /api/v1/system/health` is the structured, read-only health surface for
@@ -253,11 +286,46 @@ The header carries two read-only status badges — an **environment** badge
 a **mode: read-only** badge — plus a **Docs** link to this document. No
 privileged action controls are present in the Phase 1 shell.
 
-Not-yet-implemented surfaces (`/sessions`, `/inventory`, `/timeline`,
-`/policy`, `/insights`) resolve to graceful read-only stub pages instead of
-404s; their backing views land in later child issues of #631 (the inventory
-surfaces are backed by #636). Mutating methods on stub routes still fail closed
-with `read-only-mvp`.
+Not-yet-implemented surfaces (`/inventory`, `/timeline`, `/policy`,
+`/insights`) resolve to graceful read-only stub pages instead of 404s; their
+backing views land in later child issues of #631 (the inventory surfaces are
+backed by #636). Mutating methods on stub routes still fail closed with
+`read-only-mvp`.
+
+### Runtime and sessions (#641)
+
+`/sessions` is a live Phase 1 read-only view that composes:
+
+* runtime health from `#430` (profile, role, identity, master parity, stale warning)
+* control-plane sessions / leases and filesystem locks / worktrees / namespaces from `#636`
+* durable contamination markers when detectable (`#630` runtime recovery, `#671` stable-branch push)
+
+It surfaces stale indicators (dead PID, expired lease) and never silences an
+active contamination marker. Recovery links point only at sanctioned
+reconnect/operator restart docs (`docs/mcp-namespace-eof-recovery.md`,
+`docs/mcp-namespace-health.md`, `docs/mcp-restart-path-inventory.md`, this
+document). The page does **not** restart, kill, or take over sessions; manual
+`pkill` of MCP daemons is contamination, not recovery.
+
+Honesty rules specific to this view:
+
+* **Ownership columns never assert absence they cannot prove.** When the
+  `leases` or `locks` section is degraded or unavailable, the Leases and
+  Worktree-binding cells render `unknown (inventory <status>)` with an
+  *authority unproven* badge instead of `none` / `unbound`, and a caveat names
+  the unreadable sections. A worktree binding is correlated through lease work
+  numbers, so it is unproven when *either* section fails to read.
+  `/api/sessions` carries the same facts as `ownership_authority_complete`,
+  `ownership_section_status`, and per-row `lease_authority` /
+  `worktree_authority`, so a JSON consumer can tell "holds none" from "could
+  not be read".
+* **Contamination text is redacted at the display boundary.** Marker payloads
+  (`command_summary`, `reason_class`, `session_id`, `role`) are
+  operator-supplied free text that does not arrive through inventory scrubbing,
+  so they pass through `webui.inventory.scrub_text`, which collapses `$HOME` and
+  redacts credential-shaped tokens and URL userinfo *anywhere* in the string.
+  The write-time redactor is a narrow denylist and is not relied on. The field
+  itself is kept — it is the `#630` evidence naming which daemon was killed.
 
 ## System-health dashboard (#639)
 

gitea_mcp_server.py

@@ -2068,6 +2068,8 @@ import lease_lifecycle  # noqa: E402
 import lease_policy  # noqa: E402
 import workflow_dashboard  # noqa: E402  # #605 live queue/lease dashboard
 import restart_coordinator  # noqa: E402  # #658 MCP restart coordinator/impact
+import drain_proof  # noqa: E402  # #661 pre-restart drain proof and hard gate
+import restart_audit  # noqa: E402  # #665 restart audit events + incidents
 import incident_bridge  # noqa: E402
 import sentry_observability  # noqa: E402  (#606 optional Sentry observability)
 import sentry_incident_bridge  # noqa: E402  (#607 Sentry→Gitea incident bridge)
@@ -9551,15 +9553,13 @@ def gitea_edit_pr(
     if closing:
         gate_reasons = _profile_operation_gate("gitea.pr.close")
         if gate_reasons:
-            return {
-                "success": False,
-                "performed": False,
-                "pr_number": pr_number,
-                "requested_state": "closed",
-                "required_permission": "gitea.pr.close",
-                "reasons": gate_reasons,
-                "permission_report": _permission_block_report("gitea.pr.close"),
-            }
+            return _build_operation_gate_refusal(
+                "gitea.pr.close",
+                gate_reasons,
+                pr_number=pr_number,
+                requested_state="closed",
+                required_permission="gitea.pr.close",
+            )
 
     h, o, r = _resolve(remote, host, org, repo)
     auth = _auth(h)
@@ -13822,13 +13822,17 @@ def gitea_view_issue(
 
 def _permission_block_report(required_operation: str,
                              identity: str | None = None) -> dict:
-    """Structured, LLM-safe explanation of a permission denial (#142).
+    """Structured, LLM-safe explanation of a permission denial (#142, #897).
 
     Built only after a gate has already refused; it adds guidance to the
     refusal and never widens any permission, performs network I/O, or
     raises (fail-soft: degrades to a minimal fail-closed report). Names
     configured profiles only — never auth references, tokens, endpoint
     URLs, or keychain IDs.
+
+    #897: never fabricate a missing permission when the active profile
+    already allows the operation. That path is a diagnostic defect (the
+    refusal was not a permission denial), not a cue to switch profiles.
     """
     report = {
         "requested_operation": required_operation,
@@ -13840,6 +13844,7 @@ def _permission_block_report(required_operation: str,
         "matching_configured_profiles": [],
         "runtime_switching_supported": False,
         "different_mcp_namespace_required": True,
+        "diagnostic_defect": False,
         "exact_safe_next_action": (
             "Ask the operator to fix GITEA_MCP_CONFIG/GITEA_MCP_PROFILE; "
             "the active profile could not be resolved (fail closed)."),
@@ -13854,6 +13859,32 @@ def _permission_block_report(required_operation: str,
     report["active_allowed_operations"] = (
         profile.get("allowed_operations") or [])
 
+    # #897: fail closed as a diagnostic defect when the active profile
+    # already holds the operation — callers must not invent "missing".
+    try:
+        holds, _hold_reason = gitea_config.check_operation(
+            required_operation,
+            profile.get("allowed_operations") or [],
+            profile.get("forbidden_operations") or [],
+        )
+    except Exception:
+        holds = False
+    if holds:
+        report["missing_permission"] = None
+        report["required_permission"] = required_operation
+        report["diagnostic_defect"] = True
+        report["different_mcp_namespace_required"] = False
+        report["exact_safe_next_action"] = (
+            "Diagnostic defect: the active profile already allows "
+            f"{required_operation}. This is not a permission denial — "
+            "inspect blocker_kind / reasons (stale-runtime or runtime-mode). "
+            "Do not call gitea_activate_profile or switch MCP sessions."
+        )
+        report["matching_configured_profiles"] = [
+            p for p in [profile.get("profile_name")] if p
+        ]
+        return report
+
     matching = []
     try:
         config = gitea_config.load_config() or {}
@@ -13902,6 +13933,205 @@ def _permission_block_report(required_operation: str,
     return report
 
 
+def _reason_is_stale_runtime(reason: str) -> bool:
+    """True when *reason* is a master-parity / stale-daemon refusal (#897)."""
+    r = (reason or "").lower()
+    if not r:
+        return False
+    if "stale relative to live master" in r:
+        return True
+    if "server code is stale" in r:
+        return True
+    if "daemon is stale" in r:
+        return True
+    if "started at commit" in r and "workspace master is now" in r:
+        return True
+    if "mcp server started at" in r and "stale" in r:
+        return True
+    if "restart the server to load the current capability gates" in r:
+        return True
+    if "restart/reconnect before mutating" in r:
+        return True
+    return False
+
+
+def _reason_is_runtime_mode(reason: str) -> bool:
+    """True when *reason* is a stable-control / runtime-mode refusal (#897)."""
+    r = (reason or "").lower()
+    if not r:
+        return False
+    if _reason_is_stale_runtime(reason):
+        return False
+    if "runtime mode could not be assessed" in r:
+        return True
+    if "runtime mode is" in r:
+        return True
+    if "stable control runtime" in r:
+        return True
+    if "dev-test" in r and ("runtime" in r or "production" in r):
+        return True
+    if "development worktree" in r or "dev worktree" in r:
+        return True
+    if "launched from a 'branches/" in r or "launched from a \"branches/" in r:
+        return True
+    if "process-root / active-workspace alignment" in r:
+        return True
+    if "namespace" in r and "reproof" in r:
+        return True
+    return False
+
+
+def _reason_is_permission(reason: str) -> bool:
+    """True when *reason* is a genuine profile-permission denial (#897)."""
+    r = (reason or "").lower()
+    if not r:
+        return False
+    if _reason_is_stale_runtime(reason) or _reason_is_runtime_mode(reason):
+        return False
+    if "profile could not be resolved" in r:
+        return True
+    if "profile has no configured allowed operations" in r:
+        return True
+    if "profile forbids" in r:
+        return True
+    if "profile is not allowed to" in r:
+        return True
+    if "unrecognized forbidden operation" in r:
+        return True
+    return False
+
+
+def _classify_operation_gate_reasons(reasons: list[str]) -> dict:
+    """Partition gate reasons into stale / runtime-mode / permission (#897)."""
+    stale: list[str] = []
+    runtime_mode: list[str] = []
+    permission: list[str] = []
+    other: list[str] = []
+    for reason in reasons or []:
+        if _reason_is_stale_runtime(reason):
+            stale.append(reason)
+        elif _reason_is_runtime_mode(reason):
+            runtime_mode.append(reason)
+        elif _reason_is_permission(reason):
+            permission.append(reason)
+        else:
+            other.append(reason)
+    return {
+        "stale_runtime": stale,
+        "runtime_mode": runtime_mode,
+        "permission": permission,
+        "other": other,
+    }
+
+
+def _stale_runtime_reconnect_action() -> str:
+    """Sanctioned recovery for a stale daemon — reconnect only (#685/#897)."""
+    return (
+        "Reconnect the IDE/client MCP session so the server reloads at the "
+        "current master head. Do not call gitea_activate_profile or switch "
+        "MCP role sessions — profile switching does not clear a stale daemon."
+    )
+
+
+def _build_operation_gate_refusal(
+    required_operation: str,
+    reasons: list[str],
+    **extra_fields,
+) -> dict:
+    """Structured gate refusal with typed blockers (#897).
+
+    Stale-runtime and runtime-mode refusals never attach a
+    ``permission_report`` and never recommend profile switching. True
+    permission denials still get ``permission_report``. When both apply,
+    causes are reported separately under distinct fields.
+    """
+    classified = _classify_operation_gate_reasons(reasons)
+    stale = classified["stale_runtime"]
+    runtime_mode = classified["runtime_mode"]
+    permission = classified["permission"]
+    other = classified["other"]
+
+    blocked: dict = {
+        "success": False,
+        "performed": False,
+        "reasons": list(reasons),
+        "mutation_performed": False,
+        "session_context_audit": session_ctx.mutation_context_audit_fields(),
+        "gate_reason_classes": {
+            "stale_runtime": list(stale),
+            "runtime_mode": list(runtime_mode),
+            "permission": list(permission),
+            "other": list(other),
+        },
+    }
+
+    if stale:
+        parity = _current_master_parity()
+        blocked["blocker_kind"] = "runtime_reconnect_required"
+        blocked["restart_required"] = True
+        blocked["stop_required"] = True
+        blocked["startup_head"] = parity.get("startup_head")
+        blocked["current_head"] = parity.get("current_head")
+        blocked["daemon_start_head"] = (
+            parity.get("daemon_start_head") or parity.get("startup_head")
+        )
+        blocked["local_head"] = (
+            parity.get("local_head") or parity.get("current_head")
+        )
+        blocked["live_remote_head"] = parity.get("live_remote_head")
+        blocked["live_stale"] = bool(parity.get("live_stale"))
+        blocked["live_known"] = bool(parity.get("live_known"))
+        blocked["exact_safe_next_action"] = _stale_runtime_reconnect_action()
+        if permission or other:
+            blocked["permission_block_reasons"] = list(permission) + list(other)
+            blocked["stale_runtime_reasons"] = list(stale)
+        # Never attach permission_report for a staleness refusal.
+        blocked.update(extra_fields)
+        return blocked
+
+    if runtime_mode:
+        blocked["blocker_kind"] = "runtime_mode_blocked"
+        blocked["restart_required"] = False
+        blocked["stop_required"] = True
+        blocked["exact_safe_next_action"] = (
+            "Real workflow mutations run only on the promoted stable control "
+            "runtime. Promote/reload the stable runtime; do not call "
+            "gitea_activate_profile or switch MCP role sessions to clear a "
+            "runtime-mode block."
+        )
+        if permission or other:
+            blocked["permission_block_reasons"] = list(permission) + list(other)
+            blocked["runtime_mode_reasons"] = list(runtime_mode)
+        blocked.update(extra_fields)
+        return blocked
+
+    # Pure permission (or unclassified-as-permission) denial.
+    blocked["blocker_kind"] = "permission_denied"
+    blocked["permission_report"] = _permission_block_report(required_operation)
+    blocked.update(extra_fields)
+    return blocked
+
+
+def _permission_report_for_gate_reasons(
+    required_operation: str,
+    reasons: list[str] | None,
+) -> dict | None:
+    """Attach ``permission_report`` only for true permission denials (#897).
+
+    Call sites that historically always attached a permission report after
+    ``_profile_operation_gate`` should use this so stale/runtime refusals
+    do not emit a fabricated missing-permission payload.
+    """
+    if not reasons:
+        return None
+    classified = _classify_operation_gate_reasons(reasons)
+    if classified["stale_runtime"] or classified["runtime_mode"]:
+        return None
+    if not (classified["permission"] or classified["other"]):
+        return None
+    return _permission_block_report(required_operation)
+
+
 def _role_for_operation(op: str) -> str | None:
     # Normalize op first
     try:
@@ -14078,7 +14308,7 @@ def _master_parity_block(op: str) -> list[str]:
 
 
 def _profile_operation_gate(op: str) -> list[str]:
-    """Profile permission check for a single gated operation (#126, #216, #420).
+    """Profile permission check for a single gated operation (#126, #216, #420, #897).
 
     Issue discussion comments are gated separately from the gitea.pr.*
     review/merge family: listing requires ``gitea.read``, creating requires
@@ -14091,21 +14321,26 @@ def _profile_operation_gate(op: str) -> list[str]:
     capability gate that has since been merged, and when the runtime itself is
     not the promoted stable control runtime (#615) -- a dev/test or unknown
     runtime holds production credentials but has not been promoted.
+
+    #897: collect *all* independent refusal classes (stale, runtime-mode,
+    permission) rather than short-circuiting after the first. Callers that
+    only need a boolean still treat any non-empty list as blocked; typed
+    consumers (``_build_operation_gate_refusal``) can separate causes.
     """
-    stale_reasons = _master_parity_block(op)
-    if stale_reasons:
-        return stale_reasons
-    runtime_reasons = _runtime_mode_block(op)
-    if runtime_reasons:
-        return runtime_reasons
+    reasons: list[str] = []
+    reasons.extend(_master_parity_block(op))
+    reasons.extend(_runtime_mode_block(op))
     try:
         profile = get_profile()
     except Exception as exc:
-        return [f"profile could not be resolved (fail closed): {_redact(str(exc))}"]
+        reasons.append(
+            f"profile could not be resolved (fail closed): {_redact(str(exc))}"
+        )
+        return reasons
     op_ok, op_reason = gitea_config.check_operation(
         op, profile["allowed_operations"], profile["forbidden_operations"])
     if op_ok:
-        return []
+        return reasons
 
     if _try_auto_switch_for_operation(op):
         try:
@@ -14113,17 +14348,26 @@ def _profile_operation_gate(op: str) -> list[str]:
             op_ok, op_reason = gitea_config.check_operation(
                 op, profile["allowed_operations"], profile["forbidden_operations"])
             if op_ok:
-                return []
+                return reasons
         except Exception as exc:
-            return [f"profile could not be resolved (fail closed): {_redact(str(exc))}"]
+            reasons.append(
+                f"profile could not be resolved (fail closed): {_redact(str(exc))}"
+            )
+            return reasons
 
     if op_reason == "no-allowed-operations":
-        return ["profile has no configured allowed operations (fail closed)"]
-    if op_reason == "forbidden":
-        return [f"profile forbids '{op}'"]
-    if op_reason == "invalid-forbidden-entry":
-        return ["profile has an unrecognized forbidden operation entry (fail closed)"]
-    return [f"profile is not allowed to {op}"]
+        reasons.append(
+            "profile has no configured allowed operations (fail closed)"
+        )
+    elif op_reason == "forbidden":
+        reasons.append(f"profile forbids '{op}'")
+    elif op_reason == "invalid-forbidden-entry":
+        reasons.append(
+            "profile has an unrecognized forbidden operation entry (fail closed)"
+        )
+    else:
+        reasons.append(f"profile is not allowed to {op}")
+    return reasons
 
 
 def _mutation_config_authority_block(required_operation: str) -> dict | None:
@@ -14343,10 +14587,14 @@ def _session_context_mutation_block(
 
 
 def _profile_permission_block(required_operation: str, **extra_fields) -> dict | None:
-    """Structured permission denial for gated tools (#69, #142).
+    """Structured operation-gate denial for gated tools (#69, #142, #897).
 
     Returns a block dict when the active profile forbids *required_operation*,
-    or ``None`` when the gate passes. Never performs network I/O.
+    the daemon is stale, or the runtime mode is not mutation-safe — or
+    ``None`` when the gate passes. Never performs network I/O.
+
+    #897: stale-runtime and runtime-mode refusals are typed
+    (``blocker_kind``) and never carry a ``permission_report``.
     """
     req_role = "reviewer" if any(required_operation.startswith(p) for p in (
         "gitea.pr.approve", "gitea.pr.merge", "gitea.pr.request_changes", "gitea.pr.review"
@@ -14356,15 +14604,9 @@ def _profile_permission_block(required_operation: str, **extra_fields) -> dict |
 
     reasons = _profile_operation_gate(required_operation)
     if reasons:
-        blocked = {
-            "success": False,
-            "performed": False,
-            "reasons": reasons,
-            "permission_report": _permission_block_report(required_operation),
-            "session_context_audit": session_ctx.mutation_context_audit_fields(),
-        }
-        blocked.update(extra_fields)
-        return blocked
+        return _build_operation_gate_refusal(
+            required_operation, reasons, **extra_fields
+        )
 
     auth_block = _mutation_config_authority_block(required_operation)
     if auth_block is not None:
@@ -14493,20 +14735,14 @@ def gitea_acquire_reviewer_pr_lease(
     """Acquire a per-PR reviewer lease before review/merge mutations (#407)."""
     read_block = _profile_operation_gate("gitea.read")
     if read_block:
-        return {
-            "success": False,
-            "acquired": False,
-            "reasons": read_block,
-            "permission_report": _permission_block_report("gitea.read"),
-        }
+        return _build_operation_gate_refusal(
+            "gitea.read", read_block, acquired=False
+        )
     comment_block = _profile_operation_gate("gitea.pr.comment")
     if comment_block:
-        return {
-            "success": False,
-            "acquired": False,
-            "reasons": comment_block,
-            "permission_report": _permission_block_report("gitea.pr.comment"),
-        }
+        return _build_operation_gate_refusal(
+            "gitea.pr.comment", comment_block, acquired=False
+        )
 
     # task=acquire_reviewer_pr_lease so verify_preflight_purity runs shared #604
     # anti-stomp for the declared lease-acquire mutation inventory entry.
@@ -14622,20 +14858,14 @@ def gitea_acquire_merger_pr_lease(
     """
     read_block = _profile_operation_gate("gitea.read")
     if read_block:
-        return {
-            "success": False,
-            "acquired": False,
-            "reasons": read_block,
-            "permission_report": _permission_block_report("gitea.read"),
-        }
+        return _build_operation_gate_refusal(
+            "gitea.read", read_block, acquired=False
+        )
     comment_block = _profile_operation_gate("gitea.pr.comment")
     if comment_block:
-        return {
-            "success": False,
-            "acquired": False,
-            "reasons": comment_block,
-            "permission_report": _permission_block_report("gitea.pr.comment"),
-        }
+        return _build_operation_gate_refusal(
+            "gitea.pr.comment", comment_block, acquired=False
+        )
     merge_block = _profile_operation_gate("gitea.pr.merge")
     if merge_block:
         return {
@@ -19374,14 +19604,12 @@ def gitea_update_pr_branch_by_merge(
     # Permission: author branch push / PR mutation surface.
     push_block = _profile_operation_gate("gitea.branch.push")
     if push_block:
-        return {
-            "success": False,
-            "performed": False,
-            "mutation_allowed": False,
-            "reasons": push_block,
-            "permission_report": _permission_block_report("gitea.branch.push"),
-            "role_kind": role,
-        }
+        return _build_operation_gate_refusal(
+            "gitea.branch.push",
+            push_block,
+            mutation_allowed=False,
+            role_kind=role,
+        )
 
     if role != "author":
         pre = pr_sync_status.assess_update_pr_branch_preflight(
@@ -22342,19 +22570,40 @@ def gitea_request_mcp_restart(
     request_override: bool = False,
     session_id: str | None = None,
     limit: int = 200,
+    restart_class: str = "full_mcp_restart",
+    target_session_id: str | None = None,
+    target_role: str | None = None,
+    target_connector: str | None = None,
+    drain_proof_json: str | None = None,
+    request_break_glass: bool = False,
 ) -> dict:
     """Evaluate a proposed MCP restart and return an impact preview (#658).
 
-    Central restart coordinator: gathers live control-plane state (sessions,
+    Central restart coordinator: resolves the requested restart class, gathers
+    live control-plane state (sessions,
     leases/locks, in-flight issue/PR work, mutations, worktrees) and returns a
     blast-radius impact report with a ``safe`` / ``unsafe`` / ``override``
     verdict, so the console (#642/#652) and operators can see what a restart
     would disrupt *before* any concurrent LLM work is destroyed.
 
-    This tool is **dry-run and never restarts anything.** The mutative apply
-    path is a separate child gated by a drain proof (non-goal here); calling
-    with ``dry_run=False`` still performs no restart and reports that apply is
-    not yet available.
+    This tool **never restarts a process.** In dry-run (the default) it returns
+    only the impact preview. With ``dry_run=False`` it enforces the #661 hard
+    gate: the apply request must present a valid, unexpired, clean drain proof
+    (``drain_proof_json``) or it is denied and a durable incident descriptor is
+    returned under ``incident``. Break-glass is the only bypass and is honoured
+    only when ``request_break_glass`` is set *and* the environment carries
+    ``GITEA_BREAKGLASS_RESTART_AUTHORIZATION``. Even an authorized gate performs
+    no restart here; actual execution is a further child. The gate outcome is
+    reported under ``apply_gate``.
+
+    ``apply_authorized`` requires **both** authorizations to pass: the #661 drain
+    gate *and* the #663 restart-class matrix (``allow_restart``). They are
+    independent — the drain gate proves the blast radius was drained and knows
+    nothing about whether this requester may request this class — so a class the
+    matrix denied never reports an authorized apply. Break-glass bypasses the
+    drain proof only; it never bypasses the class matrix. ``apply_gate`` carries
+    ``drain_gate_allow`` and ``restart_class_authorized`` so a denial is
+    attributable to the authorization that produced it.
 
     Operator override authority is read from the process environment
     (``GITEA_OPERATOR_RESTART_OVERRIDE_AUTHORIZATION``), never self-asserted by
@@ -22439,6 +22688,9 @@ def gitea_request_mcp_restart(
 
     profile = get_profile()
     profile_name = (profile.get("profile_name") or "").strip() or "session"
+    requester_role = (
+        profile.get("role_kind") or profile.get("role") or ""
+    ).strip().lower()
     sid = (session_id or "").strip() or f"{profile_name}-{os.getpid()}"
 
     # Override authority is read from the environment only — a worker session
@@ -22448,6 +22700,15 @@ def gitea_request_mcp_restart(
         (os.environ.get("GITEA_OPERATOR_RESTART_OVERRIDE_AUTHORIZATION") or "").strip()
     )
     operator_override = bool(request_override and operator_authorized)
+    controller_approved = bool(
+        (
+            os.environ.get("GITEA_CONTROLLER_RESTART_APPROVAL_AUTHORIZATION")
+            or ""
+        ).strip()
+    )
+    requester_permissions = restart_coordinator.permissions_for_role(
+        requester_role
+    )
 
     inventory = {
         "sessions": sessions,
@@ -22462,6 +22723,14 @@ def gitea_request_mcp_restart(
         operator_override=operator_override,
         requesting_session_id=sid,
         dry_run=True,  # coordinator is always analysis-only (#658)
+        restart_class=restart_class,
+        requester_role=requester_role,
+        requester_permissions=requester_permissions,
+        controller_approved=controller_approved,
+        operator_authorized=operator_authorized,
+        target_session_id=target_session_id,
+        target_role=target_role,
+        target_connector=target_connector,
     )
 
     payload = report.as_dict()
@@ -22473,12 +22742,186 @@ def gitea_request_mcp_restart(
     payload["requesting_session_id"] = sid
     payload["operator_override_requested"] = bool(request_override)
     payload["operator_override_authorized"] = operator_authorized
+    payload["controller_approval_authorized"] = controller_approved
+    payload["requester_role"] = requester_role
+    payload["requester_permissions"] = list(requester_permissions)
+    # Actual restart execution remains a further child; this tool never restarts
+    # a process. What #661 adds is the *hard gate*: an apply request (dry_run
+    # False) must present a valid, unexpired, clean drain proof, or it is denied
+    # and a durable incident is raised. Break-glass is the only bypass and its
+    # authorization is read from the environment, never self-asserted.
     payload["apply_supported"] = False
+    # #665: correlation id threads impact preview → apply gate → incidents.
+    correlation_id = restart_audit.new_correlation_id()
+    payload["correlation_id"] = correlation_id
+    auth_user = None
+    try:
+        # remote-only: host override is for operator diagnostics, not required here
+        auth_user = (gitea_whoami(remote=remote) or {}).get("username")
+    except Exception:  # noqa: BLE001 — identity is best-effort for audit
+        auth_user = None
+    preview_audit = restart_audit.record_restart_lifecycle(
+        event_type=restart_audit.EVENT_IMPACT_PREVIEW,
+        outcome=str(report.verdict or "unknown"),
+        correlation_id=correlation_id,
+        remote=remote,
+        org=o,
+        repo=r,
+        requesting_session_id=sid,
+        restart_class=restart_class,
+        profile_name=profile_name,
+        authenticated_username=auth_user,
+        reasons=list(report.reasons or []),
+        details={
+            "dry_run": True,
+            "allow_restart": bool(report.allow_restart),
+            "inventory_complete": inventory_complete,
+        },
+        privileged=False,
+    )
+    payload["restart_audit"] = {
+        "correlation_id": correlation_id,
+        "impact_preview_written": preview_audit["audit_written"],
+    }
     if not dry_run:
-        payload["reasons"] = list(payload.get("reasons") or []) + [
-            "apply requested but not supported: sanctioned restart apply is "
-            "gated by a drain proof (separate child); no restart performed (#658)"
-        ]
+        proof_obj: dict | None = None
+        proof_parse_error: str | None = None
+        if drain_proof_json:
+            try:
+                parsed = json.loads(drain_proof_json)
+                proof_obj = parsed if isinstance(parsed, dict) else None
+                if proof_obj is None:
+                    proof_parse_error = "drain_proof_json is not a JSON object"
+            except (ValueError, TypeError) as exc:
+                proof_parse_error = f"invalid drain_proof_json: {_redact(str(exc))}"
+
+        break_glass_authorized = bool(
+            (
+                os.environ.get("GITEA_BREAKGLASS_RESTART_AUTHORIZATION") or ""
+            ).strip()
+        )
+        break_glass = bool(request_break_glass and break_glass_authorized)
+
+        expected_fp = drain_proof.impact_fingerprint(report.as_dict())
+        gate = drain_proof.gate_apply_restart(
+            proof=proof_obj,
+            break_glass=break_glass,
+            expected_impact_fingerprint=expected_fp,
+            requesting_session_id=sid,
+        )
+        gate_payload = gate.as_dict()
+        if proof_parse_error and not break_glass:
+            gate_payload["reasons"] = [proof_parse_error] + list(
+                gate_payload.get("reasons") or []
+            )
+        # The #663 restart-class matrix and the #661 drain gate are two
+        # independent authorizations, and an apply requires BOTH. ``gate.allow``
+        # proves only that the blast radius was drained — or that break-glass
+        # was authorized — and knows nothing about whether this requester may
+        # request this class at all. Conjoining them keeps a class the matrix
+        # denied from ever reporting an authorized apply, and keeps break-glass
+        # scoped to what it is for: bypassing the drain proof, never the
+        # least-privilege class matrix.
+        restart_class_authorized = bool(report.allow_restart)
+        gate_payload["drain_gate_allow"] = bool(gate.allow)
+        gate_payload["restart_class_authorized"] = restart_class_authorized
+        if not restart_class_authorized:
+            gate_payload["reasons"] = list(gate_payload.get("reasons") or []) + [
+                "restart class authorization denied; apply denied regardless of "
+                "drain proof or break-glass (fail closed, #663)",
+                *(report.authorization_reasons or []),
+            ]
+        payload["apply_gate"] = gate_payload
+        payload["apply_authorized"] = bool(gate.allow and restart_class_authorized)
+        payload["break_glass_requested"] = bool(request_break_glass)
+        payload["break_glass_authorized"] = break_glass_authorized
+        # Even an authorized gate performs no restart here: execution is a later
+        # child. The gate proves the apply path *would* be permitted.
+        payload["reasons"] = list(payload.get("reasons") or []) + list(
+            gate_payload.get("reasons") or []
+        )
+        if not gate.allow and gate.incident is not None:
+            payload["incident"] = gate.incident
+
+        # #665: audit apply-gate + materialize durable incidents for failed
+        # drain and break-glass. Privileged apply denies if audit is enabled
+        # and the sink write fails.
+        incident_desc = restart_audit.incident_from_apply_gate(
+            gate_payload={
+                **gate_payload,
+                "incident": gate.incident,
+                "allow": gate.allow,
+            },
+            break_glass=break_glass,
+            correlation_id=correlation_id,
+            requesting_session_id=sid,
+            restart_class=restart_class,
+            remote=remote,
+            org=o,
+            repo=r,
+        )
+        if incident_desc is not None:
+            payload["incident"] = incident_desc
+
+        def _create_restart_incident_issue(
+            *, title, body, labels, org=None, repo=None, **_kw
+        ):
+            return gitea_create_issue(
+                title=title,
+                body=body,
+                labels=labels,
+                remote=remote,
+                host=h,
+                org=org or o,
+                repo=repo or r,
+            )
+
+        apply_audit = restart_audit.record_restart_lifecycle(
+            event_type=(
+                restart_audit.EVENT_BREAK_GLASS
+                if break_glass
+                else restart_audit.EVENT_APPLY_GATE
+            ),
+            outcome=(
+                "break_glass"
+                if break_glass
+                else ("allow" if payload["apply_authorized"] else "deny")
+            ),
+            correlation_id=correlation_id,
+            remote=remote,
+            org=o,
+            repo=r,
+            requesting_session_id=sid,
+            restart_class=restart_class,
+            profile_name=profile_name,
+            authenticated_username=auth_user,
+            reasons=list(gate_payload.get("reasons") or []),
+            details={
+                "apply_authorized": payload["apply_authorized"],
+                "drain_gate_allow": gate_payload.get("drain_gate_allow"),
+                "restart_class_authorized": restart_class_authorized,
+                "break_glass": break_glass,
+                "proof_id": gate_payload.get("proof_id"),
+            },
+            privileged=True,
+            create_incident=incident_desc,
+            create_issue_fn=_create_restart_incident_issue
+            if incident_desc is not None
+            else None,
+            dry_run_incident=False,
+        )
+        payload["restart_audit"] = {
+            "correlation_id": correlation_id,
+            "impact_preview_written": preview_audit["audit_written"],
+            "apply_gate_written": apply_audit["audit_written"],
+            "incident_result": apply_audit.get("incident_result"),
+        }
+        if apply_audit["deny_reasons"]:
+            payload["apply_authorized"] = False
+            payload["reasons"] = list(payload.get("reasons") or []) + list(
+                apply_audit["deny_reasons"]
+            )
+            payload["success"] = True
     return payload
 
 

restart_audit.py

@@ -0,0 +1,426 @@
+"""MCP restart lifecycle audit events and incident materialization (#665).
+
+Restarts and recovery attempts must leave a forensic trail: impact previews,
+drain enter/exit, drain-proof results, apply gate verdicts, break-glass, and
+post-restart reconcile outcomes. Failed drains and break-glass must also raise
+durable Gitea incident issues so they cannot be silently repeated.
+
+This module is the pure + sink layer for that trail:
+
+* **Schema** — ``mcp.restart.*`` event names and a redacted payload builder.
+* **Emission** — append-only via :mod:`gitea_audit` (off when ``GITEA_AUDIT_LOG``
+  is unset; privileged apply can still *require* a successful write).
+* **Incidents** — descriptors for failed drain / break-glass / unguarded restart,
+  plus an optional materializer that creates a Gitea issue through an injected
+  ``create_issue_fn`` (keeps this module free of network I/O in tests).
+
+Design rules:
+
+* **No secrets.** All free text is redacted before write or issue body assembly.
+* **Never raises from emission.** ``emit_restart_event`` returns False on sink
+  failure so callers can decide fail-closed policy for privileged restarts.
+* **Does not restart.** Audit never executes a process restart.
+* **Drain proof stays #661.** This module records what the gate decided; it
+  does not re-verify proofs.
+"""
+
+from __future__ import annotations
+
+import uuid
+from datetime import datetime, timezone
+from typing import Any, Callable, Mapping, Sequence
+
+import gitea_audit
+
+# ── Event vocabulary (stable identifiers for operators + tests) ───────────────
+
+EVENT_IMPACT_PREVIEW = "mcp.restart.impact_preview"
+EVENT_DRAIN_ENTER = "mcp.restart.drain_enter"
+EVENT_DRAIN_EXIT = "mcp.restart.drain_exit"
+EVENT_DRAIN_PROOF = "mcp.restart.drain_proof"
+EVENT_APPLY_GATE = "mcp.restart.apply_gate"
+EVENT_BREAK_GLASS = "mcp.restart.break_glass"
+EVENT_POST_RESTART_RECONCILE = "mcp.restart.post_restart_reconcile"
+EVENT_NARROWER_RECOVERY = "mcp.restart.narrower_recovery"
+EVENT_UNGUARDED_DETECTED = "mcp.restart.unguarded_detected"
+
+RESTART_EVENT_TYPES: frozenset[str] = frozenset(
+    {
+        EVENT_IMPACT_PREVIEW,
+        EVENT_DRAIN_ENTER,
+        EVENT_DRAIN_EXIT,
+        EVENT_DRAIN_PROOF,
+        EVENT_APPLY_GATE,
+        EVENT_BREAK_GLASS,
+        EVENT_POST_RESTART_RECONCILE,
+        EVENT_NARROWER_RECOVERY,
+        EVENT_UNGUARDED_DETECTED,
+    }
+)
+
+# Incident kinds (durable Gitea issues).
+INCIDENT_FAILED_DRAIN = "restart_failed_drain"
+INCIDENT_BREAK_GLASS = "restart_break_glass"
+INCIDENT_RECONCILE_UNRESOLVED = "restart_reconcile_unresolved"
+INCIDENT_UNGUARDED = "restart_unguarded_detected"
+
+DEFAULT_INCIDENT_LABELS: tuple[str, ...] = (
+    "mcp-health",
+    "safety",
+    "observability",
+    "status:ready",
+    "type:bug",
+    "workflow-hardening",
+)
+
+CreateIssueFn = Callable[..., dict[str, Any]]
+
+
+def _utc_now_iso() -> str:
+    return datetime.now(timezone.utc).isoformat()
+
+
+def new_correlation_id() -> str:
+    """Mint a short correlation id shared across a restart lifecycle."""
+    return f"rst-{uuid.uuid4().hex[:16]}"
+
+
+def build_restart_event(
+    *,
+    event_type: str,
+    outcome: str,
+    correlation_id: str | None = None,
+    remote: str | None = None,
+    org: str | None = None,
+    repo: str | None = None,
+    requesting_session_id: str | None = None,
+    restart_class: str | None = None,
+    profile_name: str | None = None,
+    authenticated_username: str | None = None,
+    reasons: Sequence[str] | None = None,
+    details: Mapping[str, Any] | None = None,
+    now: str | None = None,
+) -> dict[str, Any]:
+    """Build a redacted ``mcp.restart.*`` audit event.
+
+    Raises ``ValueError`` on unknown event types so a typo cannot silently land
+    under a free-form action name.
+    """
+    name = str(event_type or "").strip()
+    if name not in RESTART_EVENT_TYPES:
+        raise ValueError(
+            f"unknown restart audit event_type {name!r}; expected one of "
+            f"{sorted(RESTART_EVENT_TYPES)}"
+        )
+    redacted_reasons = [
+        gitea_audit.redact(str(r)) for r in (reasons or []) if str(r).strip()
+    ]
+    redacted_details = gitea_audit.redact(dict(details or {}))
+    if not isinstance(redacted_details, dict):
+        redacted_details = {"value": redacted_details}
+
+    event = gitea_audit.build_event(
+        action=name,
+        result=str(outcome or "unknown"),
+        remote=remote,
+        repository=f"{org}/{repo}" if org and repo else None,
+        profile_name=profile_name,
+        authenticated_username=authenticated_username,
+        reason="; ".join(redacted_reasons) if redacted_reasons else None,
+        request_metadata={
+            "event_family": "mcp.restart",
+            "correlation_id": correlation_id or new_correlation_id(),
+            "restart_class": restart_class,
+            "requesting_session_id": requesting_session_id,
+            "org": org,
+            "repo": repo,
+            "details": redacted_details,
+            "reasons": redacted_reasons,
+        },
+        now=now or _utc_now_iso(),
+        operation=name,
+    )
+    event["action_type"] = "restart_lifecycle"
+    event["event_type"] = name
+    event["correlation_id"] = (event.get("request_metadata") or {}).get(
+        "correlation_id"
+    )
+    return event
+
+
+def emit_restart_event(event: Mapping[str, Any], *, path: str | None = None) -> bool:
+    """Append *event* to the audit sink. Never raises. Returns write success."""
+    try:
+        return bool(gitea_audit.write_event(dict(event), path=path))
+    except Exception:
+        return False
+
+
+def require_audit_or_deny(
+    *,
+    privileged: bool,
+    written: bool,
+    audit_enabled: bool | None = None,
+) -> list[str]:
+    """Return deny reasons when a privileged restart path fails to audit.
+
+    When audit is not configured (``GITEA_AUDIT_LOG`` unset), privileged apply
+    still proceeds under the rollout policy "enable audit before enforcing
+    deny-on-audit-fail" — but *if* audit is enabled and the write fails,
+    privileged apply is denied (fail closed).
+    """
+    enabled = (
+        gitea_audit.audit_enabled() if audit_enabled is None else bool(audit_enabled)
+    )
+    if not privileged:
+        return []
+    if not enabled:
+        return []
+    if written:
+        return []
+    return [
+        "privileged restart path requires a successful audit write; "
+        "audit sink failed (fail closed, #665)"
+    ]
+
+
+# ── Incident descriptors ──────────────────────────────────────────────────────
+
+
+def build_incident_descriptor(
+    *,
+    kind: str,
+    reasons: Sequence[str],
+    correlation_id: str | None = None,
+    requesting_session_id: str | None = None,
+    restart_class: str | None = None,
+    remote: str | None = None,
+    org: str | None = None,
+    repo: str | None = None,
+    proof_id: str | None = None,
+    at: str | None = None,
+) -> dict[str, Any]:
+    """Build a durable incident descriptor (no network)."""
+    titles = {
+        INCIDENT_FAILED_DRAIN: "Restart denied: drain proof failed the hard gate",
+        INCIDENT_BREAK_GLASS: "Break-glass MCP restart authorized",
+        INCIDENT_RECONCILE_UNRESOLVED: "Post-restart reconcile left unresolved work",
+        INCIDENT_UNGUARDED: "Unguarded MCP restart attempt detected",
+    }
+    title = titles.get(kind, f"MCP restart incident ({kind})")
+    redacted_reasons = [
+        gitea_audit.redact(str(r)) for r in reasons if str(r).strip()
+    ]
+    return {
+        "kind": kind,
+        "title": title,
+        "labels": list(DEFAULT_INCIDENT_LABELS),
+        "reasons": redacted_reasons,
+        "correlation_id": correlation_id,
+        "requesting_session_id": requesting_session_id,
+        "restart_class": restart_class,
+        "remote": remote,
+        "org": org,
+        "repo": repo,
+        "proof_id": proof_id,
+        "at": at or _utc_now_iso(),
+        "source": "restart_audit#665",
+    }
+
+
+def incident_body(descriptor: Mapping[str, Any]) -> str:
+    """Render a redacted markdown body for a Gitea incident issue."""
+    reasons = descriptor.get("reasons") or []
+    reason_lines = "\n".join(f"- {gitea_audit.redact(str(r))}" for r in reasons) or (
+        "- (no reasons recorded)"
+    )
+    return "\n".join(
+        [
+            "<!-- mcp-restart-incident:v1 -->",
+            f"## MCP restart incident (`{descriptor.get('kind')}`)",
+            "",
+            f"**Correlation:** `{descriptor.get('correlation_id') or 'none'}`",
+            f"**Session:** `{descriptor.get('requesting_session_id') or 'none'}`",
+            f"**Class:** `{descriptor.get('restart_class') or 'none'}`",
+            f"**Scope:** `{descriptor.get('remote')}/{descriptor.get('org')}/"
+            f"{descriptor.get('repo')}`",
+            f"**At:** `{descriptor.get('at')}`",
+            f"**Proof id:** `{descriptor.get('proof_id') or 'none'}`",
+            "",
+            "### Reasons",
+            reason_lines,
+            "",
+            "### Operator next steps",
+            "- Treat this as durable follow-up work under the restart-governance umbrella (#655).",
+            "- Do not invent a second restart path; use sanctioned coordinator tools only.",
+            "- Raw secrets must never appear in this issue (already redacted).",
+            "",
+            f"_Source: {descriptor.get('source')}_",
+        ]
+    )
+
+
+def materialize_incident(
+    descriptor: Mapping[str, Any],
+    *,
+    create_issue_fn: CreateIssueFn | None,
+    dry_run: bool = False,
+) -> dict[str, Any]:
+    """Create a Gitea issue from *descriptor* when *create_issue_fn* is provided.
+
+    Returns a result dict with ``created`` / ``issue_number`` / ``dry_run`` /
+    ``reasons``. Never raises.
+    """
+    base: dict[str, Any] = {
+        "created": False,
+        "dry_run": bool(dry_run),
+        "issue_number": None,
+        "kind": descriptor.get("kind"),
+        "reasons": [],
+        "descriptor": dict(descriptor),
+    }
+    if dry_run:
+        base["reasons"] = ["dry-run only; no Gitea issue created"]
+        return base
+    if create_issue_fn is None:
+        base["reasons"] = [
+            "create_issue_fn not provided; incident descriptor retained only"
+        ]
+        return base
+    try:
+        result = create_issue_fn(
+            title=str(descriptor.get("title") or "MCP restart incident"),
+            body=incident_body(descriptor),
+            labels=list(descriptor.get("labels") or DEFAULT_INCIDENT_LABELS),
+            org=descriptor.get("org"),
+            repo=descriptor.get("repo"),
+        )
+        number = None
+        if isinstance(result, dict):
+            number = result.get("number") or result.get("issue_number")
+        if number is not None:
+            base["created"] = True
+            base["issue_number"] = int(number)
+            base["reasons"] = [f"created incident issue #{int(number)}"]
+        else:
+            base["reasons"] = ["create_issue_fn returned no issue number"]
+    except Exception as exc:  # noqa: BLE001 — never break restart path here
+        base["reasons"] = [
+            f"incident issue creation failed: {gitea_audit.redact(str(exc))}"
+        ]
+    return base
+
+
+def record_restart_lifecycle(
+    *,
+    event_type: str,
+    outcome: str,
+    correlation_id: str,
+    remote: str | None = None,
+    org: str | None = None,
+    repo: str | None = None,
+    requesting_session_id: str | None = None,
+    restart_class: str | None = None,
+    profile_name: str | None = None,
+    authenticated_username: str | None = None,
+    reasons: Sequence[str] | None = None,
+    details: Mapping[str, Any] | None = None,
+    privileged: bool = False,
+    create_incident: Mapping[str, Any] | None = None,
+    create_issue_fn: CreateIssueFn | None = None,
+    dry_run_incident: bool = False,
+    audit_path: str | None = None,
+) -> dict[str, Any]:
+    """Emit one restart audit event and optionally materialize an incident.
+
+    Returns ``{event, audit_written, deny_reasons, incident_result}``.
+    """
+    event = build_restart_event(
+        event_type=event_type,
+        outcome=outcome,
+        correlation_id=correlation_id,
+        remote=remote,
+        org=org,
+        repo=repo,
+        requesting_session_id=requesting_session_id,
+        restart_class=restart_class,
+        profile_name=profile_name,
+        authenticated_username=authenticated_username,
+        reasons=reasons,
+        details=details,
+    )
+    written = emit_restart_event(event, path=audit_path)
+    deny = require_audit_or_deny(privileged=privileged, written=written)
+    incident_result = None
+    if create_incident is not None:
+        incident_result = materialize_incident(
+            create_incident,
+            create_issue_fn=create_issue_fn,
+            dry_run=dry_run_incident,
+        )
+    return {
+        "event": event,
+        "audit_written": written,
+        "deny_reasons": deny,
+        "incident_result": incident_result,
+        "correlation_id": correlation_id,
+    }
+
+
+def incident_from_apply_gate(
+    *,
+    gate_payload: Mapping[str, Any],
+    break_glass: bool,
+    correlation_id: str,
+    requesting_session_id: str | None,
+    restart_class: str | None,
+    remote: str | None,
+    org: str | None,
+    repo: str | None,
+) -> dict[str, Any] | None:
+    """Choose an incident descriptor from an apply-gate payload, if required."""
+    reasons = list(gate_payload.get("reasons") or [])
+    proof_id = gate_payload.get("proof_id")
+    if break_glass:
+        return build_incident_descriptor(
+            kind=INCIDENT_BREAK_GLASS,
+            reasons=reasons
+            or ["break-glass restart path used; durable incident required (#665)"],
+            correlation_id=correlation_id,
+            requesting_session_id=requesting_session_id,
+            restart_class=restart_class,
+            remote=remote,
+            org=org,
+            repo=repo,
+            proof_id=proof_id if isinstance(proof_id, str) else None,
+        )
+    # Failed drain / deny path.
+    incident = gate_payload.get("incident")
+    if isinstance(incident, Mapping) and incident:
+        # Normalize gate-provided descriptor into our schema.
+        return build_incident_descriptor(
+            kind=INCIDENT_FAILED_DRAIN,
+            reasons=list(incident.get("reasons") or reasons),
+            correlation_id=correlation_id,
+            requesting_session_id=requesting_session_id
+            or incident.get("requesting_session_id"),
+            restart_class=restart_class,
+            remote=remote,
+            org=org,
+            repo=repo,
+            proof_id=incident.get("proof_id") or proof_id,
+            at=incident.get("at"),
+        )
+    if not gate_payload.get("allow") and not gate_payload.get("drain_gate_allow", True):
+        return build_incident_descriptor(
+            kind=INCIDENT_FAILED_DRAIN,
+            reasons=reasons or ["restart apply denied"],
+            correlation_id=correlation_id,
+            requesting_session_id=requesting_session_id,
+            restart_class=restart_class,
+            remote=remote,
+            org=org,
+            repo=repo,
+            proof_id=proof_id if isinstance(proof_id, str) else None,
+        )
+    return None

restart_coordinator.py

@@ -28,11 +28,12 @@ from __future__ import annotations
 
 from dataclasses import dataclass, field
 from datetime import datetime, timezone
+from enum import Enum
 from typing import Any, Mapping, Sequence
 
 import lease_lifecycle
 
-COORDINATOR_VERSION = "1.0.0-issue-658"
+COORDINATOR_VERSION = "1.1.0-issue-663"
 
 # Restart verdicts. Exactly the three the acceptance criteria name.
 VERDICT_SAFE = "safe"
@@ -54,6 +55,194 @@ LEASE_FRESHNESS_LIVE = "active"
 DEFAULT_SESSION_HEARTBEAT_STALE_SECONDS = 900
 
 
+class RestartClass(str, Enum):
+    """The only restart/recovery classes accepted by the coordinator."""
+
+    CLIENT_RECONNECT = "client_reconnect"
+    SESSION_RECONNECT = "session_reconnect"
+    WORKER_RESTART = "worker_restart"
+    ROLE_RUNTIME_RESTART = "role_runtime_restart"
+    CONNECTOR_RESTART = "connector_restart"
+    CONFIGURATION_RELOAD = "configuration_reload"
+    ROLLING_MCP_RESTART = "rolling_mcp_restart"
+    FULL_MCP_RESTART = "full_mcp_restart"
+    HOST_RESTART = "host_restart"
+
+
+@dataclass(frozen=True)
+class RestartClassPolicy:
+    """Least-privilege policy for one :class:`RestartClass`."""
+
+    restart_class: RestartClass
+    required_permission: str
+    expected_blast_radius: str
+    drain_requirement: str
+    full_drain_required: bool
+    approval_requirement: str
+    audit_requirement: str
+    recovery_behavior: str
+    request_roles: tuple[str, ...]
+    execution_roles: tuple[str, ...]
+
+    def as_dict(self) -> dict[str, Any]:
+        return {
+            "restart_class": self.restart_class.value,
+            "required_permission": self.required_permission,
+            "expected_blast_radius": self.expected_blast_radius,
+            "drain_requirement": self.drain_requirement,
+            "full_drain_required": self.full_drain_required,
+            "approval_requirement": self.approval_requirement,
+            "audit_requirement": self.audit_requirement,
+            "recovery_behavior": self.recovery_behavior,
+            "request_roles": list(self.request_roles),
+            "execution_roles": list(self.execution_roles),
+        }
+
+
+WORKER_ROLES = ("author", "reviewer", "merger", "reconciler")
+CONTROL_ROLES = ("controller", "operator", "admin")
+ALL_REQUEST_ROLES = WORKER_ROLES + CONTROL_ROLES
+
+RESTART_CLASS_POLICIES: dict[RestartClass, RestartClassPolicy] = {
+    RestartClass.CLIENT_RECONNECT: RestartClassPolicy(
+        RestartClass.CLIENT_RECONNECT,
+        "mcp.reconnect.client",
+        BLAST_NONE,
+        "none",
+        False,
+        "self_service",
+        "record class, actor, client namespace, reason, and outcome",
+        "Reconnect only the caller's client transport; no daemon or peer session changes.",
+        ALL_REQUEST_ROLES,
+        ALL_REQUEST_ROLES,
+    ),
+    RestartClass.SESSION_RECONNECT: RestartClassPolicy(
+        RestartClass.SESSION_RECONNECT,
+        "mcp.reconnect.session",
+        BLAST_LOW,
+        "requesting_session_safe_point",
+        False,
+        "self_service",
+        "record class, actor, session id, reason, and outcome",
+        "Rebind identity, capability, and workspace state for one session.",
+        ALL_REQUEST_ROLES,
+        ALL_REQUEST_ROLES,
+    ),
+    RestartClass.WORKER_RESTART: RestartClassPolicy(
+        RestartClass.WORKER_RESTART,
+        "mcp.restart.worker.request",
+        BLAST_LOW,
+        "target_worker",
+        False,
+        "controller_approval_and_automated_gates",
+        "record class, actor, target worker, approval, drain proof, and outcome",
+        "Restart one worker after its own lease and mutation scope is drained.",
+        ALL_REQUEST_ROLES,
+        ("operator", "admin"),
+    ),
+    RestartClass.ROLE_RUNTIME_RESTART: RestartClassPolicy(
+        RestartClass.ROLE_RUNTIME_RESTART,
+        "mcp.restart.role_runtime.request",
+        BLAST_MEDIUM,
+        "target_role_runtime",
+        False,
+        "controller_approval_and_automated_gates",
+        "record class, actor, role namespace, approval, drain proof, and outcome",
+        "Restart only the selected role runtime and then re-probe that namespace.",
+        ALL_REQUEST_ROLES,
+        ("operator", "admin"),
+    ),
+    RestartClass.CONNECTOR_RESTART: RestartClassPolicy(
+        RestartClass.CONNECTOR_RESTART,
+        "mcp.restart.connector.request",
+        BLAST_MEDIUM,
+        "target_connector",
+        False,
+        "controller_approval_and_automated_gates",
+        "record class, actor, connector id, approval, drain proof, and outcome",
+        "Restart one connector while unrelated role runtimes remain available.",
+        ALL_REQUEST_ROLES,
+        ("operator", "admin"),
+    ),
+    RestartClass.CONFIGURATION_RELOAD: RestartClassPolicy(
+        RestartClass.CONFIGURATION_RELOAD,
+        "mcp.reload.configuration.request",
+        BLAST_LOW,
+        "mutation_quiesce",
+        False,
+        "controller_approval_and_automated_gates",
+        "record class, actor, configuration revision, approval, and outcome",
+        "Gracefully reload configuration without replacing the daemon process.",
+        ALL_REQUEST_ROLES,
+        ("operator", "admin"),
+    ),
+    RestartClass.ROLLING_MCP_RESTART: RestartClassPolicy(
+        RestartClass.ROLLING_MCP_RESTART,
+        "mcp.restart.rolling.request",
+        BLAST_MEDIUM,
+        "one_instance_at_a_time",
+        False,
+        "controller_approval_and_automated_gates",
+        "record class, actor, instance order, approval, per-instance drains, and outcome",
+        "Drain, restart, verify, and restore one instance before advancing to the next.",
+        CONTROL_ROLES,
+        ("operator", "admin"),
+    ),
+    RestartClass.FULL_MCP_RESTART: RestartClassPolicy(
+        RestartClass.FULL_MCP_RESTART,
+        "mcp.restart.full.request",
+        BLAST_HIGH,
+        "all_sessions_and_mutations",
+        True,
+        "controller_approval_and_automated_gates",
+        "record class, actor, full impact report, approval, drain proof, and outcome",
+        "Stop and restore the complete MCP runtime only after a verified full drain.",
+        CONTROL_ROLES,
+        ("operator", "admin"),
+    ),
+    RestartClass.HOST_RESTART: RestartClassPolicy(
+        RestartClass.HOST_RESTART,
+        "mcp.restart.host.request",
+        BLAST_HIGH,
+        "all_host_work",
+        True,
+        "controller_approval_plus_infrastructure_operator",
+        "record class, actor, host, incident or change id, approval, drain proof, and outcome",
+        "Hand off to infrastructure ownership; reconcile every runtime after the host returns.",
+        ("controller", "operator", "admin"),
+        ("operator", "admin"),
+    ),
+}
+
+
+def resolve_restart_class(value: RestartClass | str) -> RestartClass:
+    """Resolve a restart class or fail closed for an unknown value."""
+
+    if isinstance(value, RestartClass):
+        return value
+    try:
+        return RestartClass(str(value).strip())
+    except ValueError as exc:
+        raise ValueError(f"unknown restart class {value!r}; deny (fail closed)") from exc
+
+
+def restart_class_policy(value: RestartClass | str) -> RestartClassPolicy:
+    """Return the canonical policy for *value*."""
+
+    return RESTART_CLASS_POLICIES[resolve_restart_class(value)]
+
+
+def permissions_for_role(role: str | None) -> tuple[str, ...]:
+    """Return request permissions granted to a workflow role by this policy."""
+
+    normalized = str(role or "").strip().lower()
+    return tuple(
+        policy.required_permission
+        for policy in RESTART_CLASS_POLICIES.values()
+        if normalized in policy.request_roles
+    )
+
+
 def _utc_now() -> datetime:
     return datetime.now(timezone.utc)
 
@@ -75,6 +264,7 @@ class SessionImpact:
     heartbeat_stale: bool
     is_requester: bool
     live: bool
+    connector: str | None = None
 
     def as_dict(self) -> dict[str, Any]:
         return {
@@ -87,6 +277,7 @@ class SessionImpact:
             "heartbeat_stale": self.heartbeat_stale,
             "is_requester": self.is_requester,
             "live": self.live,
+            "connector": self.connector,
         }
 
 
@@ -105,6 +296,7 @@ class LeaseImpact:
     disruptive: bool
     is_mutation: bool
     is_critical_section: bool
+    connector: str | None = None
 
     def as_dict(self) -> dict[str, Any]:
         return {
@@ -119,6 +311,7 @@ class LeaseImpact:
             "disruptive": self.disruptive,
             "is_mutation": self.is_mutation,
             "is_critical_section": self.is_critical_section,
+            "connector": self.connector,
         }
 
 
@@ -127,6 +320,13 @@ class RestartImpactReport:
     """Impact preview DTO returned to the console / operator (#642/#652)."""
 
     coordinator_version: str
+    restart_class: str
+    restart_policy: dict[str, Any]
+    policy_enforced: bool
+    permission_authorized: bool
+    role_authorized: bool
+    approval_satisfied: bool
+    authorization_reasons: list[str]
     evaluated_at: str
     dry_run: bool
     restart_performed: bool
@@ -153,6 +353,13 @@ class RestartImpactReport:
     def as_dict(self) -> dict[str, Any]:
         return {
             "coordinator_version": self.coordinator_version,
+            "restart_class": self.restart_class,
+            "restart_policy": dict(self.restart_policy),
+            "policy_enforced": self.policy_enforced,
+            "permission_authorized": self.permission_authorized,
+            "role_authorized": self.role_authorized,
+            "approval_satisfied": self.approval_satisfied,
+            "authorization_reasons": list(self.authorization_reasons),
             "evaluated_at": self.evaluated_at,
             "dry_run": self.dry_run,
             "restart_performed": self.restart_performed,
@@ -206,6 +413,7 @@ def _classify_session(
             requesting_session_id and session_id == requesting_session_id
         ),
         live=live,
+        connector=(str(row.get("connector") or "").strip() or None),
     )
 
 
@@ -258,6 +466,7 @@ def _classify_lease(row: Mapping[str, Any]) -> LeaseImpact:
         disruptive=disruptive,
         is_mutation=is_mutation,
         is_critical_section=disruptive,
+        connector=(str(row.get("connector") or "").strip() or None),
     )
 
 
@@ -279,6 +488,14 @@ def evaluate_restart_impact(
     requesting_session_id: str | None = None,
     dry_run: bool = True,
     session_heartbeat_stale_seconds: int = DEFAULT_SESSION_HEARTBEAT_STALE_SECONDS,
+    restart_class: RestartClass | str | None = None,
+    requester_role: str | None = None,
+    requester_permissions: Sequence[str] | None = None,
+    controller_approved: bool = False,
+    operator_authorized: bool = False,
+    target_session_id: str | None = None,
+    target_role: str | None = None,
+    target_connector: str | None = None,
 ) -> RestartImpactReport:
     """Evaluate a proposed MCP restart and return an impact preview.
 
@@ -301,6 +518,61 @@ def evaluate_restart_impact(
     """
     moment = now or _utc_now()
     reasons: list[str] = []
+    authorization_reasons: list[str] = []
+
+    # ``None`` preserves the pre-#663 impact-only API for callers that have not
+    # yet been migrated. All MCP requests pass an explicit class and therefore
+    # take the fail-closed policy path.
+    policy_enforced = restart_class is not None
+    try:
+        resolved_class = resolve_restart_class(
+            restart_class or RestartClass.FULL_MCP_RESTART
+        )
+        policy = RESTART_CLASS_POLICIES[resolved_class]
+        unknown_class = False
+    except ValueError as exc:
+        resolved_class = None
+        policy = None
+        unknown_class = True
+        authorization_reasons.append(str(exc))
+
+    normalized_role = str(requester_role or "").strip().lower()
+    granted = {str(p).strip() for p in (requester_permissions or ())}
+    if policy_enforced and policy is not None:
+        permission_authorized = policy.required_permission in granted
+        role_authorized = normalized_role in policy.request_roles
+        if not permission_authorized:
+            authorization_reasons.append(
+                f"missing required permission {policy.required_permission!r}"
+            )
+        if not role_authorized:
+            authorization_reasons.append(
+                f"role {normalized_role or 'unknown'!r} may not request "
+                f"{policy.restart_class.value}"
+            )
+    elif unknown_class:
+        permission_authorized = False
+        role_authorized = False
+    else:
+        permission_authorized = True
+        role_authorized = True
+
+    if policy_enforced and policy is not None:
+        approval = policy.approval_requirement
+        if approval == "self_service":
+            approval_satisfied = True
+        elif approval == "controller_approval_plus_infrastructure_operator":
+            approval_satisfied = bool(controller_approved and operator_authorized)
+        else:
+            approval_satisfied = bool(controller_approved)
+        if not approval_satisfied:
+            authorization_reasons.append(
+                f"approval requirement not satisfied: {approval}"
+            )
+    elif unknown_class:
+        approval_satisfied = False
+    else:
+        approval_satisfied = True
 
     inventory_complete = bool(inventory.get("inventory_complete", False))
     incomplete_reasons = [str(r) for r in (inventory.get("incomplete_reasons") or [])]
@@ -323,15 +595,67 @@ def evaluate_restart_impact(
     ]
     lease_impacts = [_classify_lease(l) for l in leases_raw]
 
-    # Only *other* live sessions and live leases constitute blast radius: a
-    # restart that would kill only the requesting session with no other work in
-    # flight is safe.
+    # Route impact through the selected class. Narrow classes never inherit a
+    # full-runtime drain merely because unrelated work exists.
+    target_complete = True
+    if resolved_class in {
+        RestartClass.CLIENT_RECONNECT,
+        RestartClass.SESSION_RECONNECT,
+        RestartClass.CONFIGURATION_RELOAD,
+    }:
+        scoped_sessions: list[SessionImpact] = []
+        scoped_leases: list[LeaseImpact] = []
+    elif resolved_class == RestartClass.WORKER_RESTART:
+        selected_session = (target_session_id or "").strip()
+        target_complete = bool(selected_session)
+        scoped_sessions = [
+            s for s in session_impacts if s.session_id == selected_session
+        ]
+        scoped_leases = [
+            l for l in lease_impacts if l.session_id == selected_session
+        ]
+    elif resolved_class == RestartClass.ROLE_RUNTIME_RESTART:
+        selected_role = (target_role or "").strip().lower()
+        target_complete = bool(selected_role)
+        scoped_sessions = [
+            s for s in session_impacts if str(s.role or "").lower() == selected_role
+        ]
+        scoped_leases = [
+            l for l in lease_impacts if str(l.role or "").lower() == selected_role
+        ]
+    elif resolved_class == RestartClass.CONNECTOR_RESTART:
+        selected_connector = (target_connector or "").strip()
+        target_complete = bool(selected_connector)
+        scoped_sessions = [
+            s for s in session_impacts if s.connector == selected_connector
+        ]
+        scoped_leases = [
+            l for l in lease_impacts if l.connector == selected_connector
+        ]
+    else:
+        scoped_sessions = list(session_impacts)
+        scoped_leases = list(lease_impacts)
+
+    if policy_enforced and not target_complete:
+        authorization_reasons.append(
+            f"target required for {resolved_class.value if resolved_class else 'unknown class'}"
+        )
+
     other_live_sessions = [
-        s for s in session_impacts if s.live and not s.is_requester
+        s for s in scoped_sessions if s.live and not s.is_requester
     ]
-    disruptive_leases = [l for l in lease_impacts if l.disruptive]
-    critical_sections = [l for l in lease_impacts if l.is_critical_section]
-    mutations = [l for l in lease_impacts if l.is_mutation]
+    disruptive_leases = [l for l in scoped_leases if l.disruptive]
+    critical_sections = [l for l in scoped_leases if l.is_critical_section]
+    mutations = [l for l in scoped_leases if l.is_mutation]
+    terminal_lock_in_scope = (
+        terminal_lock
+        if resolved_class
+        not in {
+            RestartClass.CLIENT_RECONNECT,
+            RestartClass.SESSION_RECONNECT,
+        }
+        else None
+    )
 
     affected_issues = sorted(
         {
@@ -348,9 +672,24 @@ def evaluate_restart_impact(
         }
     )
 
-    disruptive = bool(disruptive_leases or other_live_sessions or terminal_lock)
+    disruptive = bool(
+        disruptive_leases or other_live_sessions or terminal_lock_in_scope
+    )
 
-    if not inventory_complete:
+    authorization_ok = bool(
+        not unknown_class
+        and permission_authorized
+        and role_authorized
+        and approval_satisfied
+        and target_complete
+    )
+
+    if policy_enforced and not authorization_ok:
+        verdict = VERDICT_UNSAFE
+        allow_restart = False
+        reasons.append("restart class authorization denied (fail closed)")
+        reasons.extend(authorization_reasons)
+    elif not inventory_complete:
         verdict = VERDICT_UNSAFE
         allow_restart = False
         reasons.append(
@@ -381,7 +720,7 @@ def evaluate_restart_impact(
             f"{len(critical_sections)} critical section(s) in flight "
             "(active lease with a live owner)"
         )
-    if terminal_lock:
+    if terminal_lock_in_scope:
         reasons.append("active terminal (merge) lock present")
 
     override_would_allow = bool(inventory_complete and disruptive)
@@ -411,6 +750,12 @@ def evaluate_restart_impact(
     audit_record = {
         "event": "restart_impact_evaluated",
         "coordinator_version": COORDINATOR_VERSION,
+        "restart_class": (
+            resolved_class.value if resolved_class else str(restart_class or "")
+        ),
+        "required_permission": (
+            policy.required_permission if policy is not None else None
+        ),
         "evaluated_at": moment.isoformat(),
         "dry_run": dry_run,
         "operator_override": bool(operator_override),
@@ -424,6 +769,15 @@ def evaluate_restart_impact(
 
     return RestartImpactReport(
         coordinator_version=COORDINATOR_VERSION,
+        restart_class=(
+            resolved_class.value if resolved_class else str(restart_class or "")
+        ),
+        restart_policy=policy.as_dict() if policy is not None else {},
+        policy_enforced=policy_enforced,
+        permission_authorized=permission_authorized,
+        role_authorized=role_authorized,
+        approval_satisfied=approval_satisfied,
+        authorization_reasons=authorization_reasons,
         evaluated_at=moment.isoformat(),
         dry_run=dry_run,
         restart_performed=False,
@@ -440,9 +794,11 @@ def evaluate_restart_impact(
         affected_issues=affected_issues,
         affected_prs=affected_prs,
         mutations=mutations,
-        terminal_lock=dict(terminal_lock)
-        if isinstance(terminal_lock, Mapping)
-        else terminal_lock,
+        terminal_lock=(
+            dict(terminal_lock_in_scope)
+            if isinstance(terminal_lock_in_scope, Mapping)
+            else terminal_lock_in_scope
+        ),
         ack_state=ack_state,
         prior_recovery_attempts=prior_recovery_attempts,
         counts=counts,

tests/test_drain_proof.py

@@ -0,0 +1,907 @@
+"""Tests for the pre-restart drain proof and hard gate (#661).
+
+Covers the acceptance criteria:
+
+1. Restart apply without a proof fails closed.
+2. A successful drain produces a verifiable proof.
+3. An open unsafe mutation makes the proof fail (multi-session fixture).
+4. Pass / fail / expired verification paths.
+
+Plus the security posture: forged/tampered proofs are rejected, break-glass is
+the only bypass and is never silent, a stale blast-radius fingerprint rejects a
+proof, and no per-process secret ever leaks into a serialized artifact.
+"""
+
+from __future__ import annotations
+
+import os
+import unittest
+from datetime import datetime, timedelta, timezone
+
+import drain_proof as dp
+import restart_coordinator as rc
+
+
+NOW = datetime(2026, 7, 24, 6, 0, 0, tzinfo=timezone.utc)
+SECRET = b"unit-test-drain-proof-secret-0123456789abcdef"
+
+
+def _live_pid() -> int:
+    return os.getpid()
+
+
+def _clean_drain_state() -> dict:
+    """Every drain action succeeded, no sessions outstanding."""
+
+    return {
+        "assignments_stopped": True,
+        "checkpoints_complete": True,
+        "handoffs_verified": True,
+        "leases_handled": True,
+        "acks": {},  # no other live sessions to acknowledge
+        "ack_timeout_policy_applied": False,
+    }
+
+
+def _safe_report() -> dict:
+    """Impact report with no other live work: a restart here is safe."""
+
+    report = rc.evaluate_restart_impact(
+        {"sessions": [], "leases": [], "inventory_complete": True},
+        now=NOW,
+        requesting_session_id="prgs-controller-1-req",
+    )
+    return report.as_dict()
+
+
+def _unsafe_mutation_report() -> dict:
+    """Multi-session report: a second session holds a live author mutation."""
+
+    sessions = [
+        {
+            "session_id": "prgs-controller-1-req",
+            "role": "controller",
+            "profile": "prgs-controller",
+            "pid": _live_pid(),
+            "status": "active",
+            "last_heartbeat_at": NOW.isoformat(),
+        },
+        {
+            "session_id": "prgs-author-99",
+            "role": "author",
+            "profile": "prgs-author",
+            "pid": _live_pid(),
+            "status": "active",
+            "last_heartbeat_at": NOW.isoformat(),
+        },
+    ]
+    leases = [
+        {
+            "lease_id": "lease-mut",
+            "session_id": "prgs-author-99",
+            "role": "author",
+            "phase": "implementing",
+            "work_kind": "issue",
+            "work_number": 661,
+            "worktree_path": "branches/issue-661",
+            "freshness": {"freshness": "active"},
+        }
+    ]
+    report = rc.evaluate_restart_impact(
+        {"sessions": sessions, "leases": leases, "inventory_complete": True},
+        now=NOW,
+        requesting_session_id="prgs-controller-1-req",
+    )
+    return report.as_dict()
+
+
+class BuildDrainProofTests(unittest.TestCase):
+    def test_clean_drain_produces_verifiable_clean_proof(self):
+        """AC#2: a successful drain produces a verifiable proof."""
+
+        proof = dp.build_drain_proof(
+            impact_report=_safe_report(),
+            drain_state=_clean_drain_state(),
+            requesting_session_id="prgs-controller-1-req",
+            now=NOW,
+            secret=SECRET,
+        )
+        self.assertTrue(proof.clean)
+        self.assertEqual(proof.failed_checks, [])
+        self.assertEqual(
+            {c.name for c in proof.checks}, set(dp.REQUIRED_CHECKS)
+        )
+        result = dp.verify_drain_proof(
+            proof.as_dict(), now=NOW, secret=SECRET
+        )
+        self.assertTrue(result.valid, result.reasons)
+        self.assertFalse(result.expired)
+        self.assertFalse(result.tampered)
+
+    def test_open_mutation_makes_proof_unclean(self):
+        """AC#3: an unsafe mutation still in flight fails the proof."""
+
+        proof = dp.build_drain_proof(
+            impact_report=_unsafe_mutation_report(),
+            drain_state=_clean_drain_state(),
+            now=NOW,
+            secret=SECRET,
+        )
+        self.assertFalse(proof.clean)
+        self.assertIn(dp.CHECK_NO_INFLIGHT_MUTATIONS, proof.failed_checks)
+        # Leases-handled also fails: the report still shows a disruptive lease.
+        self.assertIn(dp.CHECK_LEASES_HANDLED, proof.failed_checks)
+        result = dp.verify_drain_proof(proof.as_dict(), now=NOW, secret=SECRET)
+        self.assertFalse(result.valid)
+
+    def test_incomplete_inventory_fails_no_mutations_check(self):
+        proof = dp.build_drain_proof(
+            impact_report={"inventory_complete": False},
+            drain_state=_clean_drain_state(),
+            now=NOW,
+            secret=SECRET,
+        )
+        self.assertFalse(proof.clean)
+        self.assertIn(dp.CHECK_NO_INFLIGHT_MUTATIONS, proof.failed_checks)
+
+    def test_missing_checkpoint_flag_fails_closed(self):
+        state = _clean_drain_state()
+        del state["checkpoints_complete"]
+        proof = dp.build_drain_proof(
+            impact_report=_safe_report(), drain_state=state, now=NOW, secret=SECRET
+        )
+        self.assertFalse(proof.clean)
+        self.assertIn(dp.CHECK_CHECKPOINTS_COMPLETE, proof.failed_checks)
+
+    def test_non_true_flags_fail_closed(self):
+        """A truthy-but-not-True value (e.g. the string 'yes') must not pass."""
+
+        state = _clean_drain_state()
+        state["assignments_stopped"] = "yes"
+        proof = dp.build_drain_proof(
+            impact_report=_safe_report(), drain_state=state, now=NOW, secret=SECRET
+        )
+        self.assertIn(dp.CHECK_ASSIGNMENTS_STOPPED, proof.failed_checks)
+
+    def test_ack_timeout_policy_satisfies_ack_check(self):
+        state = _clean_drain_state()
+        state["acks"] = {"prgs-author-99": "pending"}
+        state["ack_timeout_policy_applied"] = True
+        proof = dp.build_drain_proof(
+            impact_report=_safe_report(), drain_state=state, now=NOW, secret=SECRET
+        )
+        names = {c.name: c.passed for c in proof.checks}
+        self.assertTrue(names[dp.CHECK_ACKS_OR_TIMEOUT])
+
+    def test_outstanding_acks_without_timeout_fail(self):
+        state = _clean_drain_state()
+        state["acks"] = {"prgs-author-99": "pending"}
+        state["ack_timeout_policy_applied"] = False
+        proof = dp.build_drain_proof(
+            impact_report=_safe_report(), drain_state=state, now=NOW, secret=SECRET
+        )
+        self.assertIn(dp.CHECK_ACKS_OR_TIMEOUT, proof.failed_checks)
+
+    def test_all_acked_satisfies_ack_check(self):
+        state = _clean_drain_state()
+        state["acks"] = {"prgs-author-99": "acked", "prgs-author-2": "acknowledged"}
+        proof = dp.build_drain_proof(
+            impact_report=_safe_report(), drain_state=state, now=NOW, secret=SECRET
+        )
+        names = {c.name: c.passed for c in proof.checks}
+        self.assertTrue(names[dp.CHECK_ACKS_OR_TIMEOUT])
+
+
+class VerifyDrainProofTests(unittest.TestCase):
+    def _clean_proof_dict(self) -> dict:
+        return dp.build_drain_proof(
+            impact_report=_safe_report(),
+            drain_state=_clean_drain_state(),
+            now=NOW,
+            secret=SECRET,
+        ).as_dict()
+
+    def test_missing_proof_is_invalid(self):
+        result = dp.verify_drain_proof(None, now=NOW, secret=SECRET)
+        self.assertFalse(result.valid)
+        self.assertIsNone(result.proof_id)
+
+    def test_expired_proof_is_invalid(self):
+        """AC#4: an expired proof fails verification."""
+
+        proof = self._clean_proof_dict()
+        later = NOW + timedelta(seconds=dp.DEFAULT_PROOF_TTL_SECONDS + 1)
+        result = dp.verify_drain_proof(proof, now=later, secret=SECRET)
+        self.assertFalse(result.valid)
+        self.assertTrue(result.expired)
+
+    def test_proof_valid_just_before_expiry(self):
+        proof = self._clean_proof_dict()
+        almost = NOW + timedelta(seconds=dp.DEFAULT_PROOF_TTL_SECONDS - 1)
+        result = dp.verify_drain_proof(proof, now=almost, secret=SECRET)
+        self.assertTrue(result.valid, result.reasons)
+
+    def test_wrong_secret_rejected(self):
+        """A proof minted in a prior process (different secret) will not verify."""
+
+        proof = self._clean_proof_dict()
+        result = dp.verify_drain_proof(proof, now=NOW, secret=b"other-secret")
+        self.assertFalse(result.valid)
+        self.assertTrue(result.tampered)
+
+    def test_flipping_clean_flag_is_detected(self):
+        """Forging clean=True on an unclean proof breaks the signature."""
+
+        unclean = dp.build_drain_proof(
+            impact_report=_unsafe_mutation_report(),
+            drain_state=_clean_drain_state(),
+            now=NOW,
+            secret=SECRET,
+        ).as_dict()
+        self.assertFalse(unclean["clean"])
+        unclean["clean"] = True  # forge
+        result = dp.verify_drain_proof(unclean, now=NOW, secret=SECRET)
+        self.assertFalse(result.valid)
+        self.assertTrue(result.tampered)
+
+    def test_tampering_a_check_is_detected(self):
+        unclean = dp.build_drain_proof(
+            impact_report=_unsafe_mutation_report(),
+            drain_state=_clean_drain_state(),
+            now=NOW,
+            secret=SECRET,
+        ).as_dict()
+        for c in unclean["checks"]:
+            if c["name"] == dp.CHECK_NO_INFLIGHT_MUTATIONS:
+                c["passed"] = True  # forge the failing check to pass
+        result = dp.verify_drain_proof(unclean, now=NOW, secret=SECRET)
+        self.assertFalse(result.valid)
+        self.assertTrue(result.tampered)
+
+    def test_missing_required_check_rejected(self):
+        proof = self._clean_proof_dict()
+        proof["checks"] = [
+            c for c in proof["checks"] if c["name"] != dp.CHECK_HANDOFFS_OK
+        ]
+        result = dp.verify_drain_proof(proof, now=NOW, secret=SECRET)
+        self.assertFalse(result.valid)
+
+    def test_stale_fingerprint_rejected(self):
+        proof = self._clean_proof_dict()
+        result = dp.verify_drain_proof(
+            proof,
+            now=NOW,
+            secret=SECRET,
+            expected_impact_fingerprint="deadbeef",
+        )
+        self.assertFalse(result.valid)
+
+    def test_matching_fingerprint_accepted(self):
+        report = _safe_report()
+        proof = dp.build_drain_proof(
+            impact_report=report,
+            drain_state=_clean_drain_state(),
+            now=NOW,
+            secret=SECRET,
+        ).as_dict()
+        fp = dp.impact_fingerprint(report)
+        result = dp.verify_drain_proof(
+            proof, now=NOW, secret=SECRET, expected_impact_fingerprint=fp
+        )
+        self.assertTrue(result.valid, result.reasons)
+
+
+class GateApplyRestartTests(unittest.TestCase):
+    def _clean_proof_dict(self) -> dict:
+        return dp.build_drain_proof(
+            impact_report=_safe_report(),
+            drain_state=_clean_drain_state(),
+            now=NOW,
+            secret=SECRET,
+        ).as_dict()
+
+    def test_apply_without_proof_denied(self):
+        """AC#1: restart apply without a proof fails closed + raises incident."""
+
+        decision = dp.gate_apply_restart(proof=None, now=NOW, secret=SECRET)
+        self.assertFalse(decision.allow)
+        self.assertEqual(decision.verdict, dp.GATE_DENY)
+        self.assertIsNotNone(decision.incident)
+        self.assertEqual(
+            decision.incident["kind"], "restart_drain_gate_denied"
+        )
+
+    def test_apply_with_valid_proof_allowed(self):
+        decision = dp.gate_apply_restart(
+            proof=self._clean_proof_dict(), now=NOW, secret=SECRET
+        )
+        self.assertTrue(decision.allow)
+        self.assertEqual(decision.verdict, dp.GATE_ALLOW)
+        self.assertIsNone(decision.incident)
+
+    def test_apply_with_expired_proof_denied_with_incident(self):
+        later = NOW + timedelta(seconds=dp.DEFAULT_PROOF_TTL_SECONDS + 5)
+        decision = dp.gate_apply_restart(
+            proof=self._clean_proof_dict(), now=later, secret=SECRET
+        )
+        self.assertFalse(decision.allow)
+        self.assertIsNotNone(decision.incident)
+
+    def test_apply_with_unclean_proof_denied(self):
+        """AC#3 at the gate: an unsafe-mutation proof is denied."""
+
+        unclean = dp.build_drain_proof(
+            impact_report=_unsafe_mutation_report(),
+            drain_state=_clean_drain_state(),
+            now=NOW,
+            secret=SECRET,
+        ).as_dict()
+        decision = dp.gate_apply_restart(proof=unclean, now=NOW, secret=SECRET)
+        self.assertFalse(decision.allow)
+        self.assertIsNotNone(decision.incident)
+
+    def test_break_glass_allows_without_proof_but_records_bypass(self):
+        decision = dp.gate_apply_restart(
+            proof=None, now=NOW, secret=SECRET, break_glass=True
+        )
+        self.assertTrue(decision.allow)
+        self.assertEqual(decision.verdict, dp.GATE_BREAK_GLASS)
+        self.assertTrue(decision.break_glass)
+        self.assertIsNone(decision.incident)
+        self.assertTrue(decision.audit_record["break_glass"])
+
+    def test_denied_gate_carries_stale_fingerprint_reason(self):
+        decision = dp.gate_apply_restart(
+            proof=self._clean_proof_dict(),
+            now=NOW,
+            secret=SECRET,
+            expected_impact_fingerprint="not-the-fingerprint",
+        )
+        self.assertFalse(decision.allow)
+
+
+class SecretHygieneTests(unittest.TestCase):
+    def test_secret_never_serialized(self):
+        proof = dp.build_drain_proof(
+            impact_report=_safe_report(),
+            drain_state=_clean_drain_state(),
+            now=NOW,
+            secret=SECRET,
+        )
+        blob = dp._canonical(proof.as_dict())
+        self.assertNotIn(SECRET.decode(), blob)
+        # The signature is a hex digest, not the raw secret.
+        self.assertNotIn(SECRET.hex(), blob)
+
+    def test_incident_descriptor_has_no_secret(self):
+        decision = dp.gate_apply_restart(proof=None, now=NOW, secret=SECRET)
+        blob = dp._canonical(decision.incident)
+        self.assertNotIn(SECRET.decode(), blob)
+
+
+def _drained_report_with_live_sessions(count: int) -> dict:
+    """Report with ``count`` other live sessions but nothing in flight.
+
+    Every other checklist item passes against this report, so a failure
+    isolates the acknowledgement check rather than tripping on mutations.
+    """
+
+    sessions = [
+        {
+            "session_id": "prgs-controller-1-req",
+            "role": "controller",
+            "profile": "prgs-controller",
+            "pid": _live_pid(),
+            "status": "active",
+            "last_heartbeat_at": NOW.isoformat(),
+        }
+    ]
+    for index in range(count):
+        sessions.append(
+            {
+                "session_id": f"prgs-author-{index}",
+                "role": "author",
+                "profile": "prgs-author",
+                "pid": _live_pid(),
+                "status": "active",
+                "last_heartbeat_at": NOW.isoformat(),
+            }
+        )
+    report = rc.evaluate_restart_impact(
+        {"sessions": sessions, "leases": [], "inventory_complete": True},
+        now=NOW,
+        requesting_session_id="prgs-controller-1-req",
+    )
+    return report.as_dict()
+
+
+class AcknowledgementFailClosedTests(unittest.TestCase):
+    """Acknowledgement evidence must fail closed unless explicitly verified.
+
+    Regression cover for the reviewed fail-open on PR #882: an absent ``acks``
+    key collapsed to ``{}`` and was read as "no other live sessions required to
+    acknowledge", so a proof minted clean and the restart gate allowed while the
+    impact report still showed other live sessions.
+    """
+
+    def _state(self, **overrides) -> dict:
+        state = _clean_drain_state()
+        state.pop("acks", None)
+        state["ack_timeout_policy_applied"] = False
+        state.update(overrides)
+        return state
+
+    def _acks_check(self, proof) -> dp.DrainCheck:
+        return next(c for c in proof.checks if c.name == dp.CHECK_ACKS_OR_TIMEOUT)
+
+    def _build(self, report: dict, state: dict):
+        return dp.build_drain_proof(
+            impact_report=report, drain_state=state, now=NOW, secret=SECRET
+        )
+
+    def assertAcksFailClosed(self, report: dict, state: dict) -> None:
+        proof = self._build(report, state)
+        self.assertFalse(self._acks_check(proof).passed)
+        self.assertIn(dp.CHECK_ACKS_OR_TIMEOUT, proof.failed_checks)
+        self.assertFalse(proof.clean)
+
+    # --- missing / null / empty / malformed ------------------------------
+
+    def test_missing_acks_key_with_live_sessions_fails_closed(self):
+        """The exact reviewed defect: absent key, three other live sessions."""
+        report = _drained_report_with_live_sessions(3)
+        self.assertEqual(report["counts"]["sessions_live_other"], 3)
+        state = self._state()
+        self.assertNotIn("acks", state)
+        proof = self._build(report, state)
+        check = self._acks_check(proof)
+        self.assertFalse(check.passed)
+        self.assertNotIn("no other live sessions", check.detail)
+        self.assertIn("fail closed", check.detail)
+        self.assertFalse(proof.clean)
+        self.assertEqual(proof.failed_checks, [dp.CHECK_ACKS_OR_TIMEOUT])
+
+    def test_none_acks_with_live_sessions_fails_closed(self):
+        self.assertAcksFailClosed(
+            _drained_report_with_live_sessions(2), self._state(acks=None)
+        )
+
+    def test_empty_acks_with_live_sessions_fails_closed(self):
+        self.assertAcksFailClosed(
+            _drained_report_with_live_sessions(1), self._state(acks={})
+        )
+
+    def test_malformed_acks_fail_closed(self):
+        for malformed in ([], "ack", 7, ("ack",), True):
+            with self.subTest(malformed=malformed):
+                self.assertAcksFailClosed(
+                    _drained_report_with_live_sessions(1),
+                    self._state(acks=malformed),
+                )
+
+    # --- stale / unproven values -----------------------------------------
+
+    def test_stale_or_unproven_ack_values_fail_closed(self):
+        for value in ("pending", "stale", "unknown", "", None, True, 1, NOW):
+            with self.subTest(value=value):
+                self.assertAcksFailClosed(
+                    _drained_report_with_live_sessions(1),
+                    self._state(acks={"prgs-author-0": value}),
+                )
+
+    def test_partial_coverage_fails_closed(self):
+        """Fewer acknowledgements than the report's live-session count."""
+        self.assertAcksFailClosed(
+            _drained_report_with_live_sessions(3),
+            self._state(acks={"prgs-author-0": "ack"}),
+        )
+
+    def test_one_unacked_entry_among_many_fails_closed(self):
+        self.assertAcksFailClosed(
+            _drained_report_with_live_sessions(2),
+            self._state(acks={"prgs-author-0": "ack", "prgs-author-1": "pending"}),
+        )
+
+    def test_unproven_live_session_count_fails_closed(self):
+        """A missing or malformed count cannot prove nobody had to acknowledge."""
+        malformed_counts = (
+            None,
+            {},
+            {"sessions_live_other": None},
+            {"sessions_live_other": "3"},
+            {"sessions_live_other": -1},
+            {"sessions_live_other": True},
+        )
+        for counts in malformed_counts:
+            with self.subTest(counts=counts):
+                report = _drained_report_with_live_sessions(0)
+                if counts is None:
+                    report.pop("counts", None)
+                else:
+                    report["counts"] = counts
+                self.assertAcksFailClosed(report, self._state())
+
+    # --- valid evidence still passes -------------------------------------
+
+    def test_complete_valid_acks_pass(self):
+        report = _drained_report_with_live_sessions(2)
+        state = self._state(
+            acks={"prgs-author-0": "ack", "prgs-author-1": "acknowledged"}
+        )
+        proof = self._build(report, state)
+        self.assertTrue(self._acks_check(proof).passed)
+        self.assertTrue(proof.clean)
+        self.assertEqual(proof.failed_checks, [])
+
+    def test_no_other_live_sessions_still_passes(self):
+        """Intended behavior retained: zero live sessions needs no acks."""
+        report = _drained_report_with_live_sessions(0)
+        self.assertEqual(report["counts"]["sessions_live_other"], 0)
+        proof = self._build(report, self._state())
+        check = self._acks_check(proof)
+        self.assertTrue(check.passed)
+        self.assertIn("sessions_live_other=0", check.detail)
+        self.assertTrue(proof.clean)
+
+    # --- timeout policy cannot become a second fail-open ------------------
+
+    def test_unproven_timeout_policy_cannot_open_the_gate(self):
+        for value in (None, "true", "yes", 1, "True", [], {}):
+            with self.subTest(value=value):
+                self.assertAcksFailClosed(
+                    _drained_report_with_live_sessions(2),
+                    self._state(ack_timeout_policy_applied=value),
+                )
+
+    def test_explicit_timeout_policy_permits(self):
+        proof = self._build(
+            _drained_report_with_live_sessions(2),
+            self._state(ack_timeout_policy_applied=True),
+        )
+        check = self._acks_check(proof)
+        self.assertTrue(check.passed)
+        self.assertIn("timeout policy", check.detail)
+        self.assertTrue(proof.clean)
+
+    # --- the gate itself must deny ---------------------------------------
+
+    def test_failed_ack_check_denies_the_restart_gate(self):
+        report = _drained_report_with_live_sessions(3)
+        proof = self._build(report, self._state())
+        self.assertFalse(proof.clean)
+        decision = dp.gate_apply_restart(
+            proof=proof.as_dict(),
+            now=NOW,
+            secret=SECRET,
+            expected_impact_fingerprint=dp.impact_fingerprint(report),
+        )
+        self.assertFalse(decision.allow)
+        self.assertEqual(decision.verdict, dp.GATE_DENY)
+        self.assertIsNotNone(decision.incident)
+
+    def test_unclean_ack_proof_fails_verification(self):
+        report = _drained_report_with_live_sessions(3)
+        proof = self._build(report, self._state())
+        result = dp.verify_drain_proof(
+            proof.as_dict(),
+            now=NOW,
+            secret=SECRET,
+            expected_impact_fingerprint=dp.impact_fingerprint(report),
+        )
+        self.assertFalse(result.valid)
+        self.assertFalse(result.clean)
+
+
+def _identity_report(*, requester: str, others: tuple[str, ...]) -> dict:
+    """Report with explicitly named requester and other live sessions.
+
+    Unlike :func:`_drained_report_with_live_sessions`, the session ids are
+    chosen by the caller so a test can supply acknowledgements for the *wrong*
+    identities while keeping the count correct.
+    """
+
+    sessions = [
+        {
+            "session_id": requester,
+            "role": "controller",
+            "profile": "prgs-controller",
+            "pid": _live_pid(),
+            "status": "active",
+            "last_heartbeat_at": NOW.isoformat(),
+        }
+    ]
+    for session_id in others:
+        sessions.append(
+            {
+                "session_id": session_id,
+                "role": "author",
+                "profile": "prgs-author",
+                "pid": _live_pid(),
+                "status": "active",
+                "last_heartbeat_at": NOW.isoformat(),
+            }
+        )
+    report = rc.evaluate_restart_impact(
+        {"sessions": sessions, "leases": [], "inventory_complete": True},
+        now=NOW,
+        requesting_session_id=requester,
+    )
+    return report.as_dict()
+
+
+class AcknowledgementIdentityBindingTests(unittest.TestCase):
+    """Acknowledgement coverage must be bound to session identity, not counted.
+
+    Regression cover for the second reviewed fail-open on PR #882 (review 582,
+    blocker B1): coverage compared ``acked_count`` against
+    ``counts.sessions_live_other``, so acknowledgements supplied for the
+    requesting session and for ids that do not exist satisfied the obligations
+    of the live sessions that never answered. The required identities are
+    carried by the report itself — ``ack_state`` keys and ``affected_sessions``
+    filtered on ``live and not is_requester`` — and only an acknowledgement
+    keyed by one of those ids may count for it.
+    """
+
+    def _state(self, **overrides) -> dict:
+        state = _clean_drain_state()
+        state.pop("acks", None)
+        state["ack_timeout_policy_applied"] = False
+        state.update(overrides)
+        return state
+
+    def _acks_check(self, proof) -> dp.DrainCheck:
+        return next(c for c in proof.checks if c.name == dp.CHECK_ACKS_OR_TIMEOUT)
+
+    def _build(self, report: dict, state: dict):
+        return dp.build_drain_proof(
+            impact_report=report, drain_state=state, now=NOW, secret=SECRET
+        )
+
+    def assertAcksFailClosed(self, report: dict, state: dict) -> dp.DrainCheck:
+        """Failure must propagate through the check, the proof, and the gate."""
+
+        proof = self._build(report, state)
+        check = self._acks_check(proof)
+        self.assertFalse(check.passed)
+        self.assertFalse(proof.clean)
+        self.assertIn(dp.CHECK_ACKS_OR_TIMEOUT, proof.failed_checks)
+        decision = dp.gate_apply_restart(
+            proof=proof.as_dict(),
+            now=NOW,
+            secret=SECRET,
+            expected_impact_fingerprint=dp.impact_fingerprint(report),
+        )
+        self.assertEqual(decision.verdict, dp.GATE_DENY)
+        self.assertFalse(decision.allow)
+        return check
+
+    # --- the reviewer's exact reproduction --------------------------------
+
+    def test_requester_plus_unknown_id_cannot_satisfy_two_live_sessions(self):
+        """Review 582 B1 verbatim: requester + a nonexistent session.
+
+        ``sessions_live_other=2`` with ``ack_state`` naming ``other-0`` and
+        ``other-1``; the drain state supplies an acknowledgement from the
+        requesting session itself and from a session that does not exist. The
+        count matches, the identities do not.
+        """
+
+        report = _identity_report(requester="req", others=("other-0", "other-1"))
+        self.assertEqual(report["counts"]["sessions_live_other"], 2)
+        self.assertEqual(
+            report["ack_state"], {"other-0": "pending", "other-1": "pending"}
+        )
+        state = self._state(acks={"req": "ack", "totally-bogus-session": "ack"})
+        check = self.assertAcksFailClosed(report, state)
+        self.assertIn("other-0", check.detail)
+        self.assertIn("other-1", check.detail)
+        self.assertIn("fail closed", check.detail)
+
+    # --- wrong / unknown / requester identities ---------------------------
+
+    def test_sufficient_count_of_wrong_ids_fails_closed(self):
+        """Right cardinality, wrong identities: two acks, neither required."""
+
+        report = _identity_report(requester="req", others=("other-0", "other-1"))
+        state = self._state(acks={"ghost-a": "ack", "ghost-b": "ack"})
+        check = self.assertAcksFailClosed(report, state)
+        self.assertIn("do not count", check.detail)
+
+    def test_more_acks_than_required_still_fails_on_wrong_ids(self):
+        """Coverage cannot be bought with volume: five acks, none required."""
+
+        report = _identity_report(requester="req", others=("other-0", "other-1"))
+        state = self._state(acks={f"ghost-{i}": "acknowledged" for i in range(5)})
+        self.assertAcksFailClosed(report, state)
+
+    def test_partial_identity_match_fails_closed(self):
+        """One required id acknowledged, the rest padded with unknown ids."""
+
+        report = _identity_report(
+            requester="req", others=("other-0", "other-1", "other-2")
+        )
+        state = self._state(
+            acks={"other-0": "ack", "ghost-1": "ack", "ghost-2": "ack"}
+        )
+        check = self.assertAcksFailClosed(report, state)
+        self.assertIn("other-1", check.detail)
+        self.assertIn("other-2", check.detail)
+
+    def test_requester_ack_never_satisfies_another_sessions_obligation(self):
+        """The requester is excluded from the required set and stays excluded."""
+
+        report = _identity_report(requester="req", others=("other-0",))
+        requester_rows = [s for s in report["affected_sessions"] if s["is_requester"]]
+        self.assertEqual([s["session_id"] for s in requester_rows], ["req"])
+        self.assertNotIn("req", report["ack_state"])
+        check = self.assertAcksFailClosed(report, self._state(acks={"req": "ack"}))
+        self.assertIn("other-0", check.detail)
+
+    def test_fabricated_ids_do_not_count_toward_coverage(self):
+        report = _identity_report(requester="req", others=("other-0",))
+        for bogus in ("", " ", "other-0 extra", "OTHER-0", "other-01", "0"):
+            with self.subTest(bogus=bogus):
+                self.assertAcksFailClosed(report, self._state(acks={bogus: "ack"}))
+
+    # --- per-session state must be explicitly valid ------------------------
+
+    def test_unproven_per_session_states_fail_closed(self):
+        """A required id present but not explicitly acknowledged fails closed."""
+
+        report = _identity_report(requester="req", others=("other-0", "other-1"))
+        for value in ("pending", "stale", "unknown", "", None, True, 1, NOW):
+            with self.subTest(value=value):
+                self.assertAcksFailClosed(
+                    report,
+                    self._state(acks={"other-0": "ack", "other-1": value}),
+                )
+
+    def test_report_ack_state_placeholder_is_never_read_as_an_ack(self):
+        """``ack_state`` values are the report's own placeholders, not evidence."""
+
+        report = _identity_report(requester="req", others=("other-0",))
+        report["ack_state"] = {"other-0": "ack"}
+        self.assertAcksFailClosed(report, self._state())
+
+    # --- missing / malformed / contradictory identity evidence -------------
+
+    def test_missing_identity_evidence_fails_closed(self):
+        report = _identity_report(requester="req", others=("other-0",))
+        report.pop("ack_state", None)
+        report.pop("affected_sessions", None)
+        check = self.assertAcksFailClosed(report, self._state(acks={"other-0": "ack"}))
+        self.assertIn("no session-identity evidence", check.detail)
+
+    def test_malformed_ack_state_fails_closed(self):
+        for malformed in ([], "other-0", 7, None, ("other-0",)):
+            with self.subTest(malformed=malformed):
+                report = _identity_report(requester="req", others=("other-0",))
+                report["ack_state"] = malformed
+                self.assertAcksFailClosed(
+                    report, self._state(acks={"other-0": "ack"})
+                )
+
+    def test_non_string_ack_state_key_fails_closed(self):
+        report = _identity_report(requester="req", others=("other-0",))
+        report["ack_state"] = {7: "pending"}
+        self.assertAcksFailClosed(report, self._state(acks={"other-0": "ack"}))
+
+    def test_malformed_affected_sessions_fails_closed(self):
+        for malformed in ("sessions", 7, {"session_id": "other-0"}, [None], [7]):
+            with self.subTest(malformed=malformed):
+                report = _identity_report(requester="req", others=("other-0",))
+                report.pop("ack_state", None)
+                report["affected_sessions"] = malformed
+                self.assertAcksFailClosed(
+                    report, self._state(acks={"other-0": "ack"})
+                )
+
+    def test_affected_sessions_without_explicit_booleans_fails_closed(self):
+        """``live``/``is_requester`` must be real booleans, never inferred."""
+
+        report = _identity_report(requester="req", others=("other-0",))
+        report.pop("ack_state", None)
+        for row in report["affected_sessions"]:
+            if row["session_id"] == "other-0":
+                row["is_requester"] = "false"
+        self.assertAcksFailClosed(report, self._state(acks={"other-0": "ack"}))
+
+    def test_affected_sessions_missing_live_flag_fails_closed(self):
+        report = _identity_report(requester="req", others=("other-0",))
+        report.pop("ack_state", None)
+        for row in report["affected_sessions"]:
+            row.pop("live", None)
+        self.assertAcksFailClosed(report, self._state(acks={"other-0": "ack"}))
+
+    def test_contradictory_ack_state_and_affected_sessions_fails_closed(self):
+        """Both views present and disagreeing is unresolvable, not a tie-break."""
+
+        report = _identity_report(requester="req", others=("other-0", "other-1"))
+        report["ack_state"] = {"other-0": "pending", "other-9": "pending"}
+        check = self.assertAcksFailClosed(
+            report, self._state(acks={"other-0": "ack", "other-9": "ack"})
+        )
+        self.assertIn("contradicts itself", check.detail)
+
+    def test_identity_count_mismatch_fails_closed(self):
+        """Identity evidence that cannot be reconciled with the count denies."""
+
+        report = _identity_report(requester="req", others=("other-0", "other-1"))
+        report["counts"] = dict(report["counts"], sessions_live_other=1)
+        check = self.assertAcksFailClosed(
+            report, self._state(acks={"other-0": "ack", "other-1": "ack"})
+        )
+        self.assertIn("cannot be reconciled", check.detail)
+
+    def test_broken_identity_evidence_outranks_timeout_policy(self):
+        """The sanctioned timeout path cannot paper over an unreadable report."""
+
+        report = _identity_report(requester="req", others=("other-0",))
+        report["ack_state"] = "not-a-mapping"
+        self.assertAcksFailClosed(report, self._state(ack_timeout_policy_applied=True))
+
+    # --- legitimate success is preserved -----------------------------------
+
+    def test_every_required_session_acknowledged_passes(self):
+        report = _identity_report(
+            requester="req", others=("other-0", "other-1", "other-2")
+        )
+        state = self._state(
+            acks={
+                "other-0": "ack",
+                "other-1": "acked",
+                "other-2": "acknowledged",
+            }
+        )
+        proof = self._build(report, state)
+        check = self._acks_check(proof)
+        self.assertTrue(check.passed)
+        self.assertTrue(proof.clean)
+        self.assertEqual(proof.failed_checks, [])
+        self.assertIn("acknowledged by identity", check.detail)
+        decision = dp.gate_apply_restart(
+            proof=proof.as_dict(),
+            now=NOW,
+            secret=SECRET,
+            expected_impact_fingerprint=dp.impact_fingerprint(report),
+        )
+        self.assertEqual(decision.verdict, dp.GATE_ALLOW)
+        self.assertTrue(decision.allow)
+
+    def test_required_session_ack_tolerates_surrounding_whitespace(self):
+        report = _identity_report(requester="req", others=("other-0",))
+        proof = self._build(report, self._state(acks={" other-0 ": " ACK "}))
+        self.assertTrue(self._acks_check(proof).passed)
+        self.assertTrue(proof.clean)
+
+    def test_no_other_live_sessions_still_passes_with_identity_evidence(self):
+        report = _identity_report(requester="req", others=())
+        self.assertEqual(report["counts"]["sessions_live_other"], 0)
+        self.assertEqual(report["ack_state"], {})
+        proof = self._build(report, self._state())
+        check = self._acks_check(proof)
+        self.assertTrue(check.passed)
+        self.assertIn("sessions_live_other=0", check.detail)
+        self.assertTrue(proof.clean)
+
+    def test_explicit_timeout_policy_retains_intended_behavior(self):
+        """Valid, correctly typed timeout evidence still permits the check."""
+
+        report = _identity_report(requester="req", others=("other-0", "other-1"))
+        proof = self._build(report, self._state(ack_timeout_policy_applied=True))
+        check = self._acks_check(proof)
+        self.assertTrue(check.passed)
+        self.assertIn("timeout policy", check.detail)
+        self.assertTrue(proof.clean)
+
+    def test_timeout_policy_still_strictly_typed_under_identity_binding(self):
+        report = _identity_report(requester="req", others=("other-0",))
+        for value in (None, "true", "True", 1, [], {}):
+            with self.subTest(value=value):
+                self.assertAcksFailClosed(
+                    report, self._state(ack_timeout_policy_applied=value)
+                )
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_issue_854_semantic_container_exclusion.py

@@ -0,0 +1,378 @@
+"""Allocator semantic container exclusion for vision/roadmap/umbrella (#854).
+
+#844 excluded epic / child-only containers (live #631) but product-vision
+(#652), phased-roadmap (#653), and umbrella (#655) coordination records still
+ranked as implementable work. This module is the live-equivalent canary:
+
+* #631 / #652 / #653 / #655-shaped records are all excluded in one inventory.
+* Independently executable children remain eligible and can be selected.
+* Ordinary issues that merely mention vision / roadmap / umbrella stay eligible.
+* Excluded containers never receive assignments or workflow leases.
+* Structured skip reason ``epic_or_child_only_container`` is reported.
+* Candidate-set fingerprint remains stable after exclusions.
+"""
+
+from __future__ import annotations
+
+import os
+import tempfile
+import unittest
+
+from allocator_service import (
+    OUTCOME_ASSIGNED,
+    OUTCOME_PREVIEW,
+    SKIP_EPIC_OR_CHILD_ONLY_CONTAINER,
+    WorkCandidate,
+    allocate_next_work,
+    candidate_set_fingerprint,
+    classify_epic_or_child_only_container,
+)
+from control_plane_db import ControlPlaneDB
+
+REMOTE = "prgs"
+ORG = "Scaled-Tech-Consulting"
+REPO = "Gitea-Tools"
+
+# Minimal bodies mirroring live coordination records (not full issue text).
+_EPIC_631_BODY = """
+## Scope (umbrella)
+
+This epic owns the **product roadmap and linkage** for the Web Console.
+Implementation is delivered via child issues only.
+
+## Explicit non-goals
+
+* Do not implement product features in this epic issue itself.
+* No product feature implementation is claimed complete solely on this epic.
+"""
+
+_VISION_652_BODY = """
+## Canonical product vision — enduring source of truth
+
+**This issue is the enduring source of truth for the MCP Control Plane Web Console product vision.**
+
+## Implementation linkage
+
+* **Do not implement features on this issue.**
+* Sequencing: roadmap issue + #631 children.
+
+## Canonical issue state
+
+```text
+STATE: vision-active
+WHO_IS_NEXT: controller (triage/ordering) / author (implementation of linked children only)
+```
+"""
+
+_ROADMAP_653_BODY = """
+## Purpose
+
+This issue is the **phased delivery roadmap and epic sequencing** for the MCP Control Plane Web Console.
+
+## Non-goals
+
+* Implementing features on this roadmap issue.
+* Deleting vision items by omitting them from phases without #652 change log.
+
+## Canonical issue state
+
+```text
+STATE: roadmap-active
+WHO_IS_NEXT: author
+```
+"""
+
+_UMBRELLA_655_BODY = """
+## Scope (umbrella)
+
+This issue owns the **canonical restart-governance program**. Implementation is via linked children only.
+
+## Acceptance criteria (umbrella)
+
+6. No product feature claimed complete on this issue alone.
+"""
+
+_CHILD_BODY = """
+## Problem
+
+Operators need a workflow-event timeline model for Phase 1.
+
+## Acceptance criteria
+
+- [ ] Timeline model API exists
+"""
+
+
+def _issue(
+    number: int,
+    *,
+    title: str = "",
+    body: str = "",
+    labels: tuple[str, ...] = ("status:ready", "type:feature"),
+    priority: int = 20,
+) -> WorkCandidate:
+    return WorkCandidate(
+        kind="issue",
+        number=number,
+        state="open",
+        labels=labels,
+        title=title or f"issue {number}",
+        body=body,
+        priority=priority,
+    )
+
+
+def _live_shaped_containers() -> list[WorkCandidate]:
+    return [
+        _issue(
+            631,
+            title="Epic: MCP Control Plane Web Console",
+            body=_EPIC_631_BODY,
+        ),
+        _issue(
+            652,
+            title="Product vision: MCP Control Plane Web Console (canonical)",
+            body=_VISION_652_BODY,
+        ),
+        _issue(
+            653,
+            title="Roadmap: MCP Control Plane Web Console (phased delivery)",
+            body=_ROADMAP_653_BODY,
+        ),
+        _issue(
+            655,
+            title="Umbrella: Governed MCP restart coordination and zero-disruption recovery",
+            body=_UMBRELLA_655_BODY,
+        ),
+    ]
+
+
+class ClassifySemanticContainersTest(unittest.TestCase):
+    def test_652_vision_is_container(self) -> None:
+        c = _issue(
+            652,
+            title="Product vision: MCP Control Plane Web Console (canonical)",
+            body=_VISION_652_BODY,
+        )
+        is_c, detail = classify_epic_or_child_only_container(c)
+        self.assertTrue(is_c)
+        self.assertIsNotNone(detail)
+        self.assertIn("body_marker", detail or "")
+
+    def test_653_roadmap_is_container(self) -> None:
+        c = _issue(
+            653,
+            title="Roadmap: MCP Control Plane Web Console (phased delivery)",
+            body=_ROADMAP_653_BODY,
+        )
+        is_c, detail = classify_epic_or_child_only_container(c)
+        self.assertTrue(is_c)
+        self.assertIn("body_marker", detail or "")
+
+    def test_655_umbrella_is_container(self) -> None:
+        c = _issue(
+            655,
+            title="Umbrella: Governed MCP restart coordination and zero-disruption recovery",
+            body=_UMBRELLA_655_BODY,
+        )
+        is_c, detail = classify_epic_or_child_only_container(c)
+        self.assertTrue(is_c)
+        self.assertIn("body_marker", detail or "")
+
+    def test_631_still_container_after_854(self) -> None:
+        c = _issue(
+            631,
+            title="Epic: MCP Control Plane Web Console",
+            body=_EPIC_631_BODY,
+        )
+        is_c, detail = classify_epic_or_child_only_container(c)
+        self.assertTrue(is_c)
+        self.assertIn("body_marker", detail or "")
+
+    def test_incidental_vision_roadmap_umbrella_words_not_container(self) -> None:
+        cases = (
+            (
+                "Document vision handoff conventions",
+                "Update docs so implementable issues that mention a vision "
+                "remain independently executable.",
+            ),
+            (
+                "Clarify roadmap sequencing notes",
+                "Write a short note about how the roadmap issue relates to children.",
+            ),
+            (
+                "Umbrella recovery checklist for authors",
+                "Authors should still implement the concrete recovery fix here.",
+            ),
+            (
+                "Product vision wording in the help text",
+                "Fix a typo in the operator-facing help string that says product vision.",
+            ),
+        )
+        for title, body in cases:
+            with self.subTest(title=title):
+                c = _issue(900, title=title, body=body)
+                is_c, detail = classify_epic_or_child_only_container(c)
+                self.assertFalse(is_c)
+                self.assertIsNone(detail)
+
+    def test_title_prefix_alone_not_container(self) -> None:
+        for title in (
+            "Epic: something mentioned only in title",
+            "Roadmap: title only without body scope",
+            "Product vision: title only without body scope",
+            "Umbrella: title only without body scope",
+        ):
+            with self.subTest(title=title):
+                c = _issue(
+                    901,
+                    title=title,
+                    body="Implement a concrete fix for the allocator skip list.",
+                )
+                is_c, detail = classify_epic_or_child_only_container(c)
+                self.assertFalse(is_c)
+                self.assertIsNone(detail)
+
+    def test_roadmap_label_alone_is_container(self) -> None:
+        c = _issue(
+            902,
+            title="Console delivery sequencing",
+            body="Track phased delivery only.",
+            labels=("status:ready", "type:roadmap"),
+        )
+        is_c, detail = classify_epic_or_child_only_container(c)
+        self.assertTrue(is_c)
+        self.assertIn("type:roadmap", detail or "")
+
+    def test_child_referencing_parent_policy_stays_eligible(self) -> None:
+        """Children may quote parent policy without becoming containers."""
+        c = _issue(
+            637,
+            title="Web Console: Workflow-event timeline model (Phase 1)",
+            body=(
+                _CHILD_BODY
+                + "\n\nParent #652 says do not implement on the vision issue; "
+                "this child is the implementable unit."
+            ),
+        )
+        is_c, _ = classify_epic_or_child_only_container(c)
+        self.assertFalse(is_c)
+
+
+class AllocateSemanticContainerExclusionTest(unittest.TestCase):
+    def setUp(self) -> None:
+        self._tmp = tempfile.TemporaryDirectory()
+        self.addCleanup(self._tmp.cleanup)
+        self.db = ControlPlaneDB(os.path.join(self._tmp.name, "cp.sqlite3"))
+
+    def _alloc(self, candidates, **kwargs):
+        defaults = dict(
+            session_id="sess-854",
+            role="author",
+            remote=REMOTE,
+            org=ORG,
+            repo=REPO,
+            profile_name="prgs-author",
+            username="jcwalker3",
+            claims={},
+            apply=False,
+        )
+        defaults.update(kwargs)
+        return allocate_next_work(self.db, candidates=candidates, **defaults)
+
+    def test_live_equivalent_canary_excludes_all_containers_selects_child(self) -> None:
+        containers = _live_shaped_containers()
+        child = _issue(
+            637,
+            title="Web Console: Workflow-event timeline model (Phase 1)",
+            body=_CHILD_BODY,
+        )
+        inventory = containers + [child]
+        res = self._alloc(inventory, apply=False)
+        self.assertTrue(res["success"], res)
+        self.assertEqual(res["outcome"], OUTCOME_PREVIEW)
+        self.assertEqual(res["selected"]["number"], 637)
+
+        skipped = {s["number"]: s for s in res["skipped"]}
+        for number in (631, 652, 653, 655):
+            self.assertIn(number, skipped, res["skipped"])
+            self.assertEqual(
+                skipped[number]["reason_code"],
+                SKIP_EPIC_OR_CHILD_ONLY_CONTAINER,
+            )
+            self.assertIn(
+                SKIP_EPIC_OR_CHILD_ONLY_CONTAINER, skipped[number]["reason"]
+            )
+
+    def test_containers_cannot_receive_assignment_or_lease(self) -> None:
+        containers = _live_shaped_containers()
+        res = self._alloc(containers, apply=True)
+        self.assertTrue(res["success"], res)
+        self.assertNotEqual(res["outcome"], OUTCOME_ASSIGNED)
+        self.assertIsNone(res.get("assignment"))
+        self.assertIsNone(res.get("selected"))
+        skipped = {s["number"]: s for s in res["skipped"]}
+        for number in (631, 652, 653, 655):
+            self.assertEqual(
+                skipped[number]["reason_code"],
+                SKIP_EPIC_OR_CHILD_ONLY_CONTAINER,
+            )
+
+        leases = []
+        if hasattr(self.db, "list_active_leases"):
+            leases = self.db.list_active_leases(
+                remote=REMOTE, org=ORG, repo=REPO
+            )
+        if not leases and hasattr(self.db, "list_leases"):
+            leases = self.db.list_leases(remote=REMOTE, org=ORG, repo=REPO)
+        for lease in leases or []:
+            work_number = (
+                lease.get("work_number") if isinstance(lease, dict) else None
+            )
+            self.assertNotIn(work_number, {631, 652, 653, 655})
+
+    def test_apply_selects_child_not_container(self) -> None:
+        containers = _live_shaped_containers()
+        child = _issue(
+            637,
+            title="Web Console: Workflow-event timeline model (Phase 1)",
+            body=_CHILD_BODY,
+        )
+        res = self._alloc(containers + [child], apply=True)
+        self.assertTrue(res["success"], res)
+        self.assertEqual(res["outcome"], OUTCOME_ASSIGNED)
+        self.assertEqual(res["selected"]["number"], 637)
+        self.assertEqual(res["assignment"]["work_number"], 637)
+
+    def test_fingerprint_stable_with_containers_present(self) -> None:
+        containers = _live_shaped_containers()
+        child = _issue(
+            637,
+            title="Web Console: Workflow-event timeline model (Phase 1)",
+            body=_CHILD_BODY,
+        )
+        inventory = containers + [child]
+        fp_before = candidate_set_fingerprint(inventory)
+        res = self._alloc(inventory, apply=False)
+        self.assertTrue(res["success"], res)
+        self.assertEqual(res["selected"]["number"], 637)
+        # Allocator reports the same CAS fingerprint for the full candidate set.
+        reported = res.get("candidate_set_fingerprint")
+        self.assertEqual(reported, fp_before)
+        # Re-fingerprint of the same inventory is byte-stable.
+        self.assertEqual(candidate_set_fingerprint(inventory), fp_before)
+
+    def test_incidental_mentions_remain_eligible(self) -> None:
+        ordinary = _issue(
+            700,
+            title="Document roadmap handoff conventions",
+            body="Write runbook text about vision vs roadmap vs child issues.",
+        )
+        res = self._alloc([ordinary], apply=False)
+        self.assertTrue(res["success"], res)
+        self.assertEqual(res["selected"]["number"], 700)
+        self.assertEqual(res["skipped"], [])
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_issue_886_apply_authorization_conjunction.py

@@ -0,0 +1,381 @@
+"""``apply_authorized`` requires BOTH authorizations (#886 review blocker B1).
+
+The #663 restart-class matrix and the #661 drain-proof hard gate are independent
+authorizations that first coexisted when PR #882 landed on master and PR #886
+merged it into the restart-class branch. The union preserved both, but the apply
+decision consulted only the drain gate::
+
+    payload["apply_authorized"] = gate.allow          # pre-fix
+
+so a clean drain proof — or an authorized break-glass, which needs no proof at
+all — reported ``apply_authorized: True`` for a restart class the least-privilege
+matrix had just denied, in the same payload that carried
+``allow_restart: False`` and "role 'author' may not request full_mcp_restart".
+
+These tests pin the conjunction and the properties that must survive it. They
+exercise the real MCP tool, which previously had no test coverage at all — that
+absence is why the defect shipped.
+"""
+
+from __future__ import annotations
+
+import json
+import os
+import unittest
+from unittest.mock import patch
+
+import drain_proof
+import gitea_mcp_server as srv
+
+CONTROLLER_APPROVAL_ENV = "GITEA_CONTROLLER_RESTART_APPROVAL_AUTHORIZATION"
+BREAK_GLASS_ENV = "GITEA_BREAKGLASS_RESTART_AUTHORIZATION"
+
+# A quiet control plane: nothing live, so the blast radius never masks the
+# authorization outcome under test.
+QUIET_SESSIONS: list[dict] = []
+QUIET_LEASES: list[dict] = []
+
+
+class _FakeDB:
+    """Minimal control-plane DB stand-in for the restart inventory."""
+
+    def __init__(self, sessions=QUIET_SESSIONS, terminal=None):
+        self._sessions = list(sessions)
+        self._terminal = terminal
+
+    def list_sessions(self, statuses=None, limit=None):
+        return list(self._sessions)
+
+    def get_active_terminal_lock(self, remote=None, org=None, repo=None):
+        return self._terminal
+
+
+def _profile(role: str) -> dict:
+    return {"profile_name": f"prgs-{role}", "role_kind": role, "role": role}
+
+
+class _RestartToolHarness(unittest.TestCase):
+    """Drives the real ``gitea_request_mcp_restart`` with a stubbed inventory."""
+
+    def _call(self, *, role: str, env: dict | None = None, **kwargs) -> dict:
+        environ = {k: v for k, v in os.environ.items()
+                   if k not in (CONTROLLER_APPROVAL_ENV, BREAK_GLASS_ENV)}
+        environ.update(env or {})
+        with patch.object(srv, "_profile_operation_gate", return_value=None), \
+                patch.object(srv, "_resolve",
+                             return_value=("gitea.prgs.cc",
+                                           "Scaled-Tech-Consulting",
+                                           "Gitea-Tools")), \
+                patch.object(srv, "get_profile", return_value=_profile(role)), \
+                patch.object(srv, "_control_plane_db_or_error",
+                             return_value=(_FakeDB(), [])), \
+                patch.object(srv.lease_lifecycle, "list_active_leases",
+                             return_value={"leases": list(QUIET_LEASES)}), \
+                patch.dict(os.environ, environ, clear=True):
+            return srv.gitea_request_mcp_restart(
+                remote="prgs",
+                org="Scaled-Tech-Consulting",
+                repo="Gitea-Tools",
+                session_id="probe-session",
+                **kwargs,
+            )
+
+    def _clean_proof_for(self, preview: dict) -> str:
+        """Mint a genuinely clean, signature-valid proof bound to *preview*.
+
+        Built from the tool's own dry-run report, so the fingerprint matches and
+        the proof is rejected for authorization reasons only — never because it
+        was stale or forged.
+        """
+        proof = drain_proof.build_drain_proof(
+            impact_report=preview,
+            drain_state={
+                "assignments_stopped": True,
+                "checkpoints_complete": True,
+                "handoffs_verified": True,
+                "leases_handled": True,
+                "acks": {},
+            },
+            requesting_session_id="probe-session",
+        )
+        self.assertTrue(proof.clean, "harness must mint a clean proof")
+        return json.dumps(proof.as_dict())
+
+
+class TestConjunction(_RestartToolHarness):
+    """AC1/AC2 — the two authorizations are ANDed, in both directions."""
+
+    def test_gate_allow_with_class_denied_yields_apply_authorized_false(self):
+        # An author may not request full_mcp_restart (CONTROL_ROLES only).
+        preview = self._call(role="author", restart_class="full_mcp_restart")
+        self.assertFalse(preview["allow_restart"])
+
+        result = self._call(
+            role="author",
+            restart_class="full_mcp_restart",
+            dry_run=False,
+            drain_proof_json=self._clean_proof_for(preview),
+        )
+
+        self.assertTrue(result["apply_gate"]["drain_gate_allow"],
+                        "drain gate itself should have allowed this proof")
+        self.assertFalse(result["apply_gate"]["restart_class_authorized"])
+        self.assertFalse(result["apply_authorized"],
+                         "a clean proof must not authorize a denied class")
+        self.assertFalse(result["allow_restart"])
+
+    def test_gate_allow_with_class_allowed_can_yield_apply_authorized_true(self):
+        preview = self._call(
+            role="operator",
+            restart_class="full_mcp_restart",
+            env={CONTROLLER_APPROVAL_ENV: "operator-approved"},
+        )
+        self.assertTrue(preview["allow_restart"],
+                        "operator + controller approval must authorize the class")
+
+        result = self._call(
+            role="operator",
+            restart_class="full_mcp_restart",
+            dry_run=False,
+            drain_proof_json=self._clean_proof_for(preview),
+            env={CONTROLLER_APPROVAL_ENV: "operator-approved"},
+        )
+
+        self.assertTrue(result["apply_gate"]["drain_gate_allow"])
+        self.assertTrue(result["apply_gate"]["restart_class_authorized"])
+        self.assertTrue(result["apply_authorized"],
+                        "both authorizations pass; apply must be authorized")
+
+    def test_denial_is_attributable_to_the_authorization_that_caused_it(self):
+        preview = self._call(role="author", restart_class="full_mcp_restart")
+        result = self._call(
+            role="author",
+            restart_class="full_mcp_restart",
+            dry_run=False,
+            drain_proof_json=self._clean_proof_for(preview),
+        )
+        blob = " ".join(result["apply_gate"]["reasons"]).lower()
+        self.assertIn("restart class authorization denied", blob)
+        self.assertIn("full_mcp_restart", blob)
+
+
+class TestProofCannotOverrideAuthorization(_RestartToolHarness):
+    """AC3 — a clean proof never overrides a class or requester-role denial."""
+
+    def test_clean_proof_cannot_override_role_denial(self):
+        for role in ("author", "reviewer", "merger", "reconciler"):
+            with self.subTest(role=role):
+                preview = self._call(role=role, restart_class="full_mcp_restart")
+                result = self._call(
+                    role=role,
+                    restart_class="full_mcp_restart",
+                    dry_run=False,
+                    drain_proof_json=self._clean_proof_for(preview),
+                )
+                self.assertFalse(result["apply_authorized"])
+
+    def test_clean_proof_cannot_override_missing_controller_approval(self):
+        # Correct role, but the class demands controller approval and the
+        # environment carries none.
+        preview = self._call(role="operator", restart_class="full_mcp_restart")
+        self.assertFalse(preview["allow_restart"])
+        result = self._call(
+            role="operator",
+            restart_class="full_mcp_restart",
+            dry_run=False,
+            drain_proof_json=self._clean_proof_for(preview),
+        )
+        self.assertFalse(result["apply_authorized"])
+
+    def test_clean_proof_cannot_override_unknown_class(self):
+        preview = self._call(role="operator", restart_class="not_a_real_class",
+                             env={CONTROLLER_APPROVAL_ENV: "yes"})
+        self.assertFalse(preview["allow_restart"])
+        result = self._call(
+            role="operator",
+            restart_class="not_a_real_class",
+            dry_run=False,
+            drain_proof_json=self._clean_proof_for(preview),
+            env={CONTROLLER_APPROVAL_ENV: "yes"},
+        )
+        self.assertFalse(result["apply_authorized"])
+
+    def test_clean_proof_cannot_override_missing_scope_target(self):
+        # worker_restart without target_session_id fails closed on scoping.
+        preview = self._call(role="operator", restart_class="worker_restart",
+                             env={CONTROLLER_APPROVAL_ENV: "yes"})
+        self.assertFalse(preview["allow_restart"])
+        result = self._call(
+            role="operator",
+            restart_class="worker_restart",
+            dry_run=False,
+            drain_proof_json=self._clean_proof_for(preview),
+            env={CONTROLLER_APPROVAL_ENV: "yes"},
+        )
+        self.assertFalse(result["apply_authorized"])
+
+
+class TestBreakGlassDoesNotCollapseTheMatrix(_RestartToolHarness):
+    """AC4 — break-glass bypasses the drain proof only, never the class matrix."""
+
+    def test_break_glass_does_not_authorize_a_denied_class(self):
+        result = self._call(
+            role="author",
+            restart_class="host_restart",
+            dry_run=False,
+            request_break_glass=True,
+            env={BREAK_GLASS_ENV: "operator-issued"},
+        )
+        self.assertTrue(result["break_glass_authorized"])
+        self.assertTrue(result["apply_gate"]["drain_gate_allow"],
+                        "break-glass does satisfy the drain gate")
+        self.assertFalse(result["apply_gate"]["restart_class_authorized"])
+        self.assertFalse(result["apply_authorized"],
+                         "break-glass must not collapse the class matrix")
+
+    def test_break_glass_across_every_worker_role_and_restricted_class(self):
+        for role in ("author", "reviewer", "merger", "reconciler"):
+            for klass in ("rolling_mcp_restart", "full_mcp_restart",
+                          "host_restart"):
+                with self.subTest(role=role, restart_class=klass):
+                    result = self._call(
+                        role=role,
+                        restart_class=klass,
+                        dry_run=False,
+                        request_break_glass=True,
+                        env={BREAK_GLASS_ENV: "operator-issued"},
+                    )
+                    self.assertFalse(result["apply_authorized"])
+
+    def test_break_glass_still_works_when_the_class_is_authorized(self):
+        # Break-glass keeps its purpose: skipping the drain proof for a caller
+        # the matrix does allow.
+        result = self._call(
+            role="operator",
+            restart_class="full_mcp_restart",
+            dry_run=False,
+            request_break_glass=True,
+            env={BREAK_GLASS_ENV: "operator-issued",
+                 CONTROLLER_APPROVAL_ENV: "operator-approved"},
+        )
+        self.assertTrue(result["apply_authorized"])
+        self.assertEqual(result["apply_gate"]["verdict"], "break_glass")
+
+    def test_break_glass_is_not_self_assertable(self):
+        # Requested but no environment authorization -> no bypass, and the
+        # unproven apply is denied.
+        result = self._call(
+            role="operator",
+            restart_class="full_mcp_restart",
+            dry_run=False,
+            request_break_glass=True,
+            env={CONTROLLER_APPROVAL_ENV: "operator-approved"},
+        )
+        self.assertTrue(result["break_glass_requested"])
+        self.assertFalse(result["break_glass_authorized"])
+        self.assertFalse(result["apply_authorized"])
+        self.assertIn("incident", result)
+
+
+class TestRestrictedClassesStayDenied(_RestartToolHarness):
+    """AC5 — restricted classes remain denied to unauthorized requesters."""
+
+    def test_restricted_classes_denied_for_worker_roles(self):
+        for role in ("author", "reviewer", "merger", "reconciler"):
+            for klass in ("rolling_mcp_restart", "full_mcp_restart",
+                          "host_restart"):
+                with self.subTest(role=role, restart_class=klass):
+                    preview = self._call(
+                        role=role,
+                        restart_class=klass,
+                        env={CONTROLLER_APPROVAL_ENV: "yes"},
+                    )
+                    self.assertFalse(preview["allow_restart"])
+                    self.assertFalse(preview["permission_authorized"])
+                    self.assertFalse(preview["role_authorized"])
+
+    def test_host_restart_needs_controller_and_infrastructure_operator(self):
+        # controller approval alone is not enough for host_restart.
+        preview = self._call(role="controller", restart_class="host_restart",
+                             env={CONTROLLER_APPROVAL_ENV: "yes"})
+        self.assertFalse(preview["approval_satisfied"])
+        self.assertFalse(preview["allow_restart"])
+
+
+class TestExistingPathsStillWork(_RestartToolHarness):
+    """AC6 — valid scoped and unscoped restart paths are unaffected."""
+
+    def test_dry_run_never_reports_apply_authorization(self):
+        result = self._call(role="operator", restart_class="full_mcp_restart",
+                            env={CONTROLLER_APPROVAL_ENV: "yes"})
+        self.assertNotIn("apply_authorized", result)
+        self.assertNotIn("apply_gate", result)
+        self.assertFalse(result["apply_supported"])
+        self.assertFalse(result["restart_performed"])
+
+    def test_self_service_unscoped_classes_authorize_for_every_role(self):
+        for role in ("author", "reviewer", "merger", "reconciler",
+                     "controller", "operator", "admin"):
+            for klass in ("client_reconnect", "session_reconnect"):
+                with self.subTest(role=role, restart_class=klass):
+                    preview = self._call(role=role, restart_class=klass)
+                    self.assertTrue(preview["allow_restart"])
+
+    def test_scoped_class_with_target_authorizes_and_applies(self):
+        env = {CONTROLLER_APPROVAL_ENV: "operator-approved"}
+        preview = self._call(role="operator", restart_class="worker_restart",
+                             target_session_id="worker-1", env=env)
+        self.assertTrue(preview["allow_restart"])
+
+        result = self._call(
+            role="operator",
+            restart_class="worker_restart",
+            target_session_id="worker-1",
+            dry_run=False,
+            drain_proof_json=self._clean_proof_for(preview),
+            env=env,
+        )
+        self.assertTrue(result["apply_authorized"])
+
+    def test_apply_still_denies_without_any_proof(self):
+        # The #661 hard gate is untouched by the conjunction.
+        result = self._call(
+            role="operator",
+            restart_class="full_mcp_restart",
+            dry_run=False,
+            env={CONTROLLER_APPROVAL_ENV: "operator-approved"},
+        )
+        self.assertFalse(result["apply_gate"]["drain_gate_allow"])
+        self.assertTrue(result["apply_gate"]["restart_class_authorized"])
+        self.assertFalse(result["apply_authorized"])
+        self.assertEqual(result["incident"]["kind"], "restart_drain_gate_denied")
+
+    def test_apply_denies_on_malformed_proof(self):
+        result = self._call(
+            role="operator",
+            restart_class="full_mcp_restart",
+            dry_run=False,
+            drain_proof_json="{not valid json",
+            env={CONTROLLER_APPROVAL_ENV: "operator-approved"},
+        )
+        self.assertFalse(result["apply_authorized"])
+        self.assertTrue(any("invalid drain_proof_json" in reason
+                            for reason in result["apply_gate"]["reasons"]))
+
+    def test_tool_never_restarts_on_any_path(self):
+        for kwargs in (
+            {"restart_class": "client_reconnect"},
+            {"restart_class": "full_mcp_restart", "dry_run": False},
+            {"restart_class": "host_restart", "dry_run": False,
+             "request_break_glass": True},
+        ):
+            with self.subTest(**kwargs):
+                result = self._call(role="operator", env={
+                    CONTROLLER_APPROVAL_ENV: "yes", BREAK_GLASS_ENV: "yes"},
+                    **kwargs)
+                self.assertFalse(result["restart_performed"])
+                self.assertFalse(result["apply_supported"])
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_issue_897_permission_stale_runtime_classification.py

@@ -0,0 +1,453 @@
+"""#897: stale-runtime / runtime-mode refusals must not look like permission denials.
+
+Acceptance criteria (issue #897):
+
+* Stale-runtime and runtime-mode refusals are typed distinctly from
+  profile-permission refusals (distinct ``blocker_kind``).
+* A refusal caused by staleness or runtime mode never emits a
+  ``permission_report`` and never names a permission the active profile holds.
+* ``_permission_block_report`` verifies the active profile actually lacks the
+  operation before reporting it missing.
+* A stale-runtime refusal reports reconnect-only recovery and never recommends
+  ``gitea_activate_profile`` or an MCP session switch.
+* The blocker payload states the observed heads (parity fields).
+* Matrix across author / reviewer / merger / reconciler profiles.
+* Regression: ``gitea_create_issue`` on a stale daemon under ``prgs-author``
+  never returns ``missing_permission: gitea.issue.create``.
+"""
+from __future__ import annotations
+
+import os
+import sys
+import unittest
+from unittest.mock import patch
+
+sys.path.insert(0, str(__import__("pathlib").Path(__file__).resolve().parent.parent))
+
+import gitea_config  # noqa: E402
+import gitea_mcp_server as mcp_server  # noqa: E402
+
+SHA_START = "7af40fb5ff7debd5e9165fe97d9c7c279358e175"
+SHA_LIVE = "2f4dec832327513118f2fe92b74da25d124a01cb"
+
+ROLE_MATRIX = (
+    (
+        "prgs-author",
+        "author",
+        "gitea.issue.create",
+        [
+            "gitea.read",
+            "gitea.issue.create",
+            "gitea.issue.comment",
+            "gitea.issue.close",
+            "gitea.branch.create",
+            "gitea.branch.push",
+            "gitea.pr.create",
+            "gitea.pr.comment",
+            "gitea.repo.commit",
+        ],
+        ["gitea.pr.approve", "gitea.pr.merge", "gitea.pr.request_changes"],
+        "gitea.pr.merge",  # forbidden op for pure-permission case
+    ),
+    (
+        "prgs-reviewer",
+        "reviewer",
+        "gitea.pr.review",
+        [
+            "gitea.read",
+            "gitea.pr.review",
+            "gitea.pr.approve",
+            "gitea.pr.request_changes",
+            "gitea.pr.comment",
+            "gitea.issue.comment",
+        ],
+        ["gitea.branch.push", "gitea.pr.create"],
+        "gitea.branch.push",
+    ),
+    (
+        "prgs-merger",
+        "merger",
+        "gitea.pr.merge",
+        [
+            "gitea.read",
+            "gitea.pr.merge",
+            "gitea.pr.comment",
+            "gitea.issue.comment",
+        ],
+        ["gitea.pr.approve", "gitea.branch.push", "gitea.pr.create"],
+        "gitea.branch.push",
+    ),
+    (
+        "prgs-reconciler",
+        "reconciler",
+        "gitea.branch.delete",
+        [
+            "gitea.read",
+            "gitea.branch.delete",
+            "gitea.pr.comment",
+            "gitea.issue.comment",
+            "gitea.pr.close",
+            "gitea.issue.close",
+        ],
+        ["gitea.pr.approve", "gitea.pr.merge"],
+        "gitea.pr.merge",
+    ),
+)
+
+
+def _profile(name: str, role: str, allowed: list[str], forbidden: list[str]) -> dict:
+    return {
+        "profile_name": name,
+        "role": role,
+        "role_kind": role,
+        "allowed_operations": list(allowed),
+        "forbidden_operations": list(forbidden),
+        "identity": "test-user",
+    }
+
+
+def _config(profiles: dict) -> dict:
+    return {
+        "version": 2,
+        "profiles": {
+            name: {
+                "role": p["role"],
+                "allowed_operations": p["allowed_operations"],
+                "forbidden_operations": p["forbidden_operations"],
+            }
+            for name, p in profiles.items()
+        },
+        "rules": {"allow_runtime_switching": True},
+    }
+
+
+class Issue897Helpers(unittest.TestCase):
+    def test_classify_stale_reason_strings(self):
+        stale = (
+            f"live remote master is {SHA_LIVE[:12]} but the MCP server started "
+            f"at {SHA_START[:12]}; the daemon is stale relative to live master "
+            "-- restart/reconnect before mutating"
+        )
+        classified = mcp_server._classify_operation_gate_reasons([stale])
+        self.assertEqual(classified["stale_runtime"], [stale])
+        self.assertEqual(classified["permission"], [])
+        self.assertEqual(classified["runtime_mode"], [])
+
+    def test_classify_permission_reason(self):
+        reason = "profile is not allowed to gitea.pr.merge"
+        classified = mcp_server._classify_operation_gate_reasons([reason])
+        self.assertEqual(classified["permission"], [reason])
+        self.assertEqual(classified["stale_runtime"], [])
+
+    def test_classify_runtime_mode_reason(self):
+        reason = (
+            "runtime mode is 'dev-test' and the mutation targets the "
+            "production repository; dev/test runtimes must not mutate real "
+            "issues or PRs (ADR: stable control runtime vs dev runtime)"
+        )
+        classified = mcp_server._classify_operation_gate_reasons([reason])
+        self.assertEqual(classified["runtime_mode"], [reason])
+        self.assertEqual(classified["stale_runtime"], [])
+
+
+class Issue897PermissionBlockReport(unittest.TestCase):
+    def test_holds_op_is_diagnostic_defect_not_missing_permission(self):
+        profile = _profile(
+            "prgs-author",
+            "author",
+            ["gitea.read", "gitea.issue.create", "gitea.issue.comment"],
+            [],
+        )
+        with patch.object(mcp_server, "get_profile", return_value=profile), patch.object(
+            mcp_server.gitea_config, "load_config", return_value=_config({"prgs-author": profile})
+        ), patch.object(
+            mcp_server.gitea_config, "is_runtime_switching_enabled", return_value=True
+        ):
+            report = mcp_server._permission_block_report("gitea.issue.create")
+        self.assertTrue(report.get("diagnostic_defect"), report)
+        self.assertIsNone(report.get("missing_permission"), report)
+        action = (report.get("exact_safe_next_action") or "").lower()
+        # Must not *recommend* profile switching; mentioning the forbidden
+        # action in a "do not call" instruction is fine.
+        self.assertNotIn("call gitea_activate_profile with", action)
+        self.assertNotIn("switch to the author mcp session", action)
+        self.assertNotIn("switch to the reviewer mcp session", action)
+        self.assertIn("diagnostic defect", action)
+
+    def test_true_missing_permission_still_reports(self):
+        profile = _profile(
+            "prgs-author",
+            "author",
+            ["gitea.read", "gitea.issue.create"],
+            ["gitea.pr.merge"],
+        )
+        reviewer = _profile(
+            "prgs-reviewer",
+            "reviewer",
+            ["gitea.read", "gitea.pr.merge", "gitea.pr.approve"],
+            [],
+        )
+        with patch.object(mcp_server, "get_profile", return_value=profile), patch.object(
+            mcp_server.gitea_config,
+            "load_config",
+            return_value=_config({"prgs-author": profile, "prgs-reviewer": reviewer}),
+        ), patch.object(
+            mcp_server.gitea_config, "is_runtime_switching_enabled", return_value=True
+        ):
+            report = mcp_server._permission_block_report("gitea.pr.merge")
+        self.assertFalse(report.get("diagnostic_defect"), report)
+        self.assertEqual(report.get("missing_permission"), "gitea.pr.merge")
+        self.assertIn("prgs-reviewer", report.get("matching_configured_profiles") or [])
+
+
+class Issue897GateRefusalMatrix(unittest.TestCase):
+    def _stale_parity(self) -> dict:
+        return {
+            "in_parity": True,
+            "stale": False,
+            "restart_required": True,
+            "determinable": True,
+            "startup_head": SHA_START,
+            "current_head": SHA_START,
+            "daemon_start_head": SHA_START,
+            "local_head": SHA_START,
+            "live_remote_head": SHA_LIVE,
+            "live_known": True,
+            "live_stale": True,
+            "mutation_safe": False,
+            "reasons": [
+                f"live remote master is {SHA_LIVE[:12]} but the MCP server "
+                f"started at {SHA_START[:12]}; the daemon is stale relative "
+                "to live master -- restart/reconnect before mutating"
+            ],
+        }
+
+    def test_stale_plus_permitted_op_all_roles(self):
+        for name, role, permitted_op, allowed, forbidden, _forbidden_op in ROLE_MATRIX:
+            with self.subTest(profile=name, op=permitted_op):
+                profile = _profile(name, role, allowed, forbidden)
+                parity = self._stale_parity()
+                with patch.object(mcp_server, "get_profile", return_value=profile), patch.object(
+                    mcp_server, "_current_master_parity", return_value=parity
+                ), patch.object(
+                    mcp_server, "_master_parity_block", return_value=list(parity["reasons"])
+                ), patch.object(
+                    mcp_server, "_runtime_mode_block", return_value=[]
+                ), patch.object(
+                    mcp_server, "_ensure_matching_profile", return_value=None
+                ), patch.object(
+                    mcp_server.session_ctx,
+                    "mutation_context_audit_fields",
+                    return_value={"session_profile": name},
+                ):
+                    blocked = mcp_server._profile_permission_block(permitted_op)
+                self.assertIsNotNone(blocked, name)
+                assert blocked is not None
+                self.assertEqual(
+                    blocked.get("blocker_kind"),
+                    "runtime_reconnect_required",
+                    blocked,
+                )
+                self.assertNotIn("permission_report", blocked, blocked)
+                self.assertTrue(blocked.get("restart_required"), blocked)
+                self.assertEqual(blocked.get("startup_head"), SHA_START, blocked)
+                self.assertEqual(blocked.get("live_remote_head"), SHA_LIVE, blocked)
+                action = (blocked.get("exact_safe_next_action") or "").lower()
+                self.assertIn("reconnect", action)
+                self.assertNotIn("call gitea_activate_profile with", action)
+                self.assertNotIn("switch to the author mcp session", action)
+                self.assertNotIn("switch to the reviewer mcp session", action)
+
+    def test_fresh_plus_forbidden_op_all_roles(self):
+        for name, role, _permitted, allowed, forbidden, forbidden_op in ROLE_MATRIX:
+            with self.subTest(profile=name, op=forbidden_op):
+                profile = _profile(name, role, allowed, forbidden)
+                with patch.object(mcp_server, "get_profile", return_value=profile), patch.object(
+                    mcp_server, "_master_parity_block", return_value=[]
+                ), patch.object(
+                    mcp_server, "_runtime_mode_block", return_value=[]
+                ), patch.object(
+                    mcp_server, "_ensure_matching_profile", return_value=None
+                ), patch.object(
+                    mcp_server.session_ctx,
+                    "mutation_context_audit_fields",
+                    return_value={"session_profile": name},
+                ), patch.object(
+                    mcp_server.gitea_config,
+                    "load_config",
+                    return_value=_config({name: profile}),
+                ), patch.object(
+                    mcp_server.gitea_config, "is_runtime_switching_enabled", return_value=False
+                ):
+                    blocked = mcp_server._profile_permission_block(forbidden_op)
+                self.assertIsNotNone(blocked, name)
+                assert blocked is not None
+                self.assertEqual(blocked.get("blocker_kind"), "permission_denied", blocked)
+                self.assertIn("permission_report", blocked, blocked)
+                report = blocked["permission_report"]
+                self.assertEqual(report.get("missing_permission"), forbidden_op, report)
+                self.assertFalse(report.get("diagnostic_defect"), report)
+                # No runtime reconnect fields for pure permission denial
+                self.assertNotEqual(
+                    blocked.get("blocker_kind"), "runtime_reconnect_required"
+                )
+
+    def test_stale_plus_forbidden_op_both_causes_separated(self):
+        for name, role, _permitted, allowed, forbidden, forbidden_op in ROLE_MATRIX:
+            with self.subTest(profile=name, op=forbidden_op):
+                profile = _profile(name, role, allowed, forbidden)
+                parity = self._stale_parity()
+                stale_reason = parity["reasons"][0]
+                with patch.object(mcp_server, "get_profile", return_value=profile), patch.object(
+                    mcp_server, "_current_master_parity", return_value=parity
+                ), patch.object(
+                    mcp_server, "_master_parity_block", return_value=[stale_reason]
+                ), patch.object(
+                    mcp_server, "_runtime_mode_block", return_value=[]
+                ), patch.object(
+                    mcp_server, "_ensure_matching_profile", return_value=None
+                ), patch.object(
+                    mcp_server.session_ctx,
+                    "mutation_context_audit_fields",
+                    return_value={"session_profile": name},
+                ):
+                    # Gate collects both classes; force permission reason too.
+                    with patch.object(
+                        mcp_server,
+                        "_profile_operation_gate",
+                        return_value=[
+                            stale_reason,
+                            f"profile is not allowed to {forbidden_op}",
+                        ],
+                    ):
+                        blocked = mcp_server._profile_permission_block(forbidden_op)
+                self.assertIsNotNone(blocked)
+                assert blocked is not None
+                self.assertEqual(
+                    blocked.get("blocker_kind"), "runtime_reconnect_required", blocked
+                )
+                self.assertNotIn("permission_report", blocked, blocked)
+                self.assertIn("permission_block_reasons", blocked, blocked)
+                self.assertIn("stale_runtime_reasons", blocked, blocked)
+                classes = blocked.get("gate_reason_classes") or {}
+                self.assertTrue(classes.get("stale_runtime"), classes)
+                self.assertTrue(classes.get("permission"), classes)
+
+    def test_runtime_mode_block_no_permission_report(self):
+        profile = _profile(
+            "prgs-author",
+            "author",
+            ["gitea.read", "gitea.issue.create"],
+            [],
+        )
+        runtime_reason = (
+            "runtime mode is 'dev-test' and the mutation targets the "
+            "production repository; dev/test runtimes must not mutate real "
+            "issues or PRs (ADR: stable control runtime vs dev runtime)"
+        )
+        with patch.object(mcp_server, "get_profile", return_value=profile), patch.object(
+            mcp_server, "_master_parity_block", return_value=[]
+        ), patch.object(
+            mcp_server, "_runtime_mode_block", return_value=[runtime_reason]
+        ), patch.object(
+            mcp_server, "_ensure_matching_profile", return_value=None
+        ), patch.object(
+            mcp_server.session_ctx,
+            "mutation_context_audit_fields",
+            return_value={"session_profile": "prgs-author"},
+        ):
+            blocked = mcp_server._profile_permission_block("gitea.issue.create")
+        self.assertIsNotNone(blocked)
+        assert blocked is not None
+        self.assertEqual(blocked.get("blocker_kind"), "runtime_mode_blocked", blocked)
+        self.assertNotIn("permission_report", blocked, blocked)
+        action = (blocked.get("exact_safe_next_action") or "").lower()
+        self.assertNotIn("call gitea_activate_profile with", action)
+        self.assertIn("stable control runtime", action)
+
+
+class Issue897CreateIssueRegression(unittest.TestCase):
+    def test_create_issue_stale_daemon_never_missing_issue_create(self):
+        """Regression AC: stale prgs-author create_issue must not claim missing create."""
+        profile = _profile(
+            "prgs-author",
+            "author",
+            [
+                "gitea.read",
+                "gitea.issue.create",
+                "gitea.issue.comment",
+                "gitea.branch.create",
+                "gitea.branch.push",
+                "gitea.pr.create",
+                "gitea.pr.comment",
+                "gitea.repo.commit",
+            ],
+            [],
+        )
+        stale_reason = (
+            f"live remote master is {SHA_LIVE[:12]} but the MCP server started "
+            f"at {SHA_START[:12]}; the daemon is stale relative to live master "
+            "-- restart/reconnect before mutating"
+        )
+        parity = {
+            "in_parity": True,
+            "stale": False,
+            "restart_required": True,
+            "determinable": True,
+            "startup_head": SHA_START,
+            "current_head": SHA_START,
+            "daemon_start_head": SHA_START,
+            "local_head": SHA_START,
+            "live_remote_head": SHA_LIVE,
+            "live_known": True,
+            "live_stale": True,
+            "mutation_safe": False,
+            "reasons": [stale_reason],
+        }
+
+        with patch.object(mcp_server, "get_profile", return_value=profile), patch.object(
+            mcp_server, "_current_master_parity", return_value=parity
+        ), patch.object(
+            mcp_server, "_master_parity_block", return_value=[stale_reason]
+        ), patch.object(
+            mcp_server, "_runtime_mode_block", return_value=[]
+        ), patch.object(
+            mcp_server, "_ensure_matching_profile", return_value=None
+        ), patch.object(
+            mcp_server.session_ctx,
+            "mutation_context_audit_fields",
+            return_value={"session_profile": "prgs-author"},
+        ), patch.object(
+            mcp_server, "_mutation_config_authority_block", return_value=None
+        ), patch.object(
+            mcp_server, "_session_context_mutation_block", return_value=None
+        ):
+            blocked = mcp_server._profile_permission_block(
+                "gitea.issue.create", remote="prgs"
+            )
+
+        self.assertIsNotNone(blocked)
+        assert blocked is not None
+        self.assertEqual(blocked.get("blocker_kind"), "runtime_reconnect_required")
+        self.assertNotIn("permission_report", blocked)
+        # Even if a caller still built a raw report, holds-check must not claim missing.
+        with patch.object(mcp_server, "get_profile", return_value=profile):
+            raw = mcp_server._permission_block_report("gitea.issue.create")
+        self.assertIsNone(raw.get("missing_permission"), raw)
+        self.assertNotEqual(raw.get("missing_permission"), "gitea.issue.create")
+
+    def test_permission_report_for_gate_reasons_skips_stale(self):
+        stale = (
+            f"live remote master is {SHA_LIVE[:12]} but the MCP server started "
+            f"at {SHA_START[:12]}; the daemon is stale relative to live master "
+            "-- restart/reconnect before mutating"
+        )
+        self.assertIsNone(
+            mcp_server._permission_report_for_gate_reasons(
+                "gitea.issue.comment", [stale]
+            )
+        )
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_mcp_restart_governance_docs.py

@@ -105,3 +105,120 @@ def test_cross_links_do_not_embed_secrets():
         text = _read(path)
         for marker in ("ghp_", "BEGIN PRIVATE KEY", "Authorization: Bearer"):
             assert marker not in text, f"{path} contains {marker!r}"
+
+
+# --- Coordinator doc stays in lock-step with the tool (#886 review blocker B2) --
+#
+# PR #882 moved the #661 drain-proof hard gate *into* gitea_request_mcp_restart,
+# but the coordinator document still described the proof as "a separate child"
+# and omitted both new parameters. Nothing referenced that document, so nothing
+# caught the drift. These tests bind the prose to the real signature.
+
+COORDINATOR_DOC = REPO_ROOT / "docs" / "mcp-restart-coordinator.md"
+
+# Affirmative claims that were accurate before #661 landed and are now false.
+# Matched against whitespace-normalized text so re-wrapping cannot hide them.
+# Deliberately not the bare phrase "a separate child": the corrected prose uses
+# it in a negation ("no longer a separate child operation"), and a guard that
+# forbids naming the old behaviour would block explaining that it changed.
+STALE_PRE_661_PHRASES = (
+    "gated by a drain proof (a separate child)",
+    "is a later child gated by a drain proof",
+    "mutative apply path is explicitly out of scope",
+    "apply is gated by a drain proof (a separate child)",
+)
+
+
+def _documented_signature_block() -> str:
+    """The fenced signature block for the tool, as published in the doc."""
+    text = _read(COORDINATOR_DOC)
+    marker = "gitea_request_mcp_restart("
+    start = text.index(marker)
+    end = text.index("```", start)
+    return text[start:end]
+
+
+def test_documented_signature_matches_the_real_tool_signature():
+    import inspect
+
+    import gitea_mcp_server
+
+    block = _documented_signature_block()
+    real = inspect.signature(gitea_mcp_server.gitea_request_mcp_restart)
+    for name in real.parameters:
+        assert name in block, (
+            f"docs/mcp-restart-coordinator.md documents no {name!r} parameter; "
+            "the published signature has drifted from the tool"
+        )
+
+
+def test_drain_proof_and_break_glass_parameters_are_documented():
+    block = _documented_signature_block()
+    for name in ("drain_proof_json", "request_break_glass"):
+        assert name in block, f"signature block missing {name}"
+
+
+def test_restart_class_and_target_scoping_parameters_survive():
+    block = _documented_signature_block()
+    for name in ("restart_class", "target_session_id", "target_role",
+                 "target_connector"):
+        assert name in block, f"signature block lost #663 parameter {name}"
+
+
+def test_gate_is_documented_as_executing_inside_this_tool():
+    lower = _read(COORDINATOR_DOC).lower()
+    assert "inside this tool" in lower, (
+        "the coordinator doc must state that the drain-proof gate executes in "
+        "gitea_request_mcp_restart, not in a later child"
+    )
+    assert "no longer a separate child operation" in lower
+
+
+def test_stale_pre_661_wording_cannot_return():
+    normalized = " ".join(_read(COORDINATOR_DOC).split()).lower()
+    for phrase in STALE_PRE_661_PHRASES:
+        assert phrase not in normalized, (
+            f"stale pre-#661 wording returned to the coordinator doc: {phrase!r}"
+        )
+
+
+def test_dry_run_versus_apply_behavior_is_documented():
+    lower = _read(COORDINATOR_DOC).lower()
+    assert "dry_run=true" in lower and "dry_run=false" in lower
+    assert "apply_supported" in lower and "restart_performed" in lower
+    assert "never restarts anything" in lower
+
+
+def test_authorization_ordering_and_conjunction_are_documented():
+    text = _read(COORDINATOR_DOC)
+    lower = text.lower()
+    assert "authorization ordering" in lower
+    assert "allow_restart" in text
+    assert "apply_authorized" in text
+    # The conjunction itself, and the attribution fields behind it.
+    assert "gate.allow and allow_restart" in text
+    for field in ("drain_gate_allow", "restart_class_authorized"):
+        assert field in text, f"doc omits apply_gate.{field}"
+
+
+def test_break_glass_scope_is_documented_as_drain_proof_only():
+    text = _read(COORDINATOR_DOC)
+    lower = text.lower()
+    assert "break-glass" in lower
+    assert "drain proof only" in lower, (
+        "doc must state break-glass never bypasses the restart-class matrix"
+    )
+    assert "GITEA_BREAKGLASS_RESTART_AUTHORIZATION" in text
+
+
+def test_fail_closed_on_apply_is_documented():
+    lower = _read(COORDINATOR_DOC).lower()
+    assert "fail closed" in lower
+    for condition in ("expired", "unclean", "tampered", "stale"):
+        assert condition in lower, f"fail-closed list omits {condition!r}"
+
+
+def test_coordinator_doc_embeds_no_secrets():
+    text = _read(COORDINATOR_DOC)
+    for marker in ("ghp_", "BEGIN PRIVATE KEY", "Authorization: Bearer"):
+        assert marker not in text, f"{COORDINATOR_DOC} contains {marker!r}"

tests/test_restart_audit.py

@@ -0,0 +1,342 @@
+"""Tests for MCP restart lifecycle audit events and incidents (#665)."""
+
+from __future__ import annotations
+
+import json
+import os
+import tempfile
+import unittest
+from unittest.mock import patch
+
+import gitea_audit
+import restart_audit as ra
+
+
+class TestEventSchema(unittest.TestCase):
+    def test_all_lifecycle_event_types_are_named(self):
+        expected = {
+            "mcp.restart.impact_preview",
+            "mcp.restart.drain_enter",
+            "mcp.restart.drain_exit",
+            "mcp.restart.drain_proof",
+            "mcp.restart.apply_gate",
+            "mcp.restart.break_glass",
+            "mcp.restart.post_restart_reconcile",
+            "mcp.restart.narrower_recovery",
+            "mcp.restart.unguarded_detected",
+        }
+        self.assertEqual(set(ra.RESTART_EVENT_TYPES), expected)
+
+    def test_build_restart_event_core_fields(self):
+        event = ra.build_restart_event(
+            event_type=ra.EVENT_IMPACT_PREVIEW,
+            outcome="safe",
+            correlation_id="rst-abc123",
+            remote="prgs",
+            org="Scaled-Tech-Consulting",
+            repo="Gitea-Tools",
+            requesting_session_id="sess-1",
+            restart_class="full_mcp_restart",
+            profile_name="prgs-author",
+            authenticated_username="bot",
+            reasons=["inventory complete"],
+            details={"allow_restart": True},
+            now="2026-07-25T12:00:00+00:00",
+        )
+        self.assertEqual(event["event_type"], ra.EVENT_IMPACT_PREVIEW)
+        self.assertEqual(event["action"], ra.EVENT_IMPACT_PREVIEW)
+        self.assertEqual(event["action_type"], "restart_lifecycle")
+        self.assertEqual(event["result"], "safe")
+        self.assertEqual(event["correlation_id"], "rst-abc123")
+        self.assertEqual(event["profile_name"], "prgs-author")
+        self.assertEqual(event["authenticated_username"], "bot")
+        meta = event["request_metadata"]
+        self.assertEqual(meta["event_family"], "mcp.restart")
+        self.assertEqual(meta["correlation_id"], "rst-abc123")
+        self.assertEqual(meta["restart_class"], "full_mcp_restart")
+        self.assertEqual(meta["details"]["allow_restart"], True)
+
+    def test_unknown_event_type_raises(self):
+        with self.assertRaises(ValueError) as ctx:
+            ra.build_restart_event(
+                event_type="mcp.restart.not_a_real_event",
+                outcome="x",
+                correlation_id="rst-1",
+            )
+        self.assertIn("unknown restart audit event_type", str(ctx.exception))
+
+    def test_reasons_and_details_are_redacted(self):
+        event = ra.build_restart_event(
+            event_type=ra.EVENT_APPLY_GATE,
+            outcome="deny",
+            correlation_id="rst-sec",
+            reasons=["token secret-xyz rejected", "ok"],
+            details={"token": "leak-token", "status": "denied"},
+        )
+        self.assertNotIn("secret-xyz", event.get("reason") or "")
+        meta = event["request_metadata"]
+        self.assertEqual(meta["details"]["token"], gitea_audit.REDACTED)
+        self.assertEqual(meta["details"]["status"], "denied")
+        for reason in meta["reasons"]:
+            self.assertNotIn("secret-xyz", reason)
+
+    def test_new_correlation_id_shape(self):
+        cid = ra.new_correlation_id()
+        self.assertTrue(cid.startswith("rst-"))
+        self.assertEqual(len(cid), len("rst-") + 16)
+
+
+class TestEmitAndRequire(unittest.TestCase):
+    def test_emit_appends_json_line(self):
+        with tempfile.TemporaryDirectory() as d:
+            path = os.path.join(d, "audit.log")
+            event = ra.build_restart_event(
+                event_type=ra.EVENT_DRAIN_PROOF,
+                outcome="pass",
+                correlation_id="rst-write",
+            )
+            self.assertTrue(ra.emit_restart_event(event, path=path))
+            with open(path, encoding="utf-8") as fh:
+                lines = fh.read().splitlines()
+            self.assertEqual(len(lines), 1)
+            loaded = json.loads(lines[0])
+            self.assertEqual(loaded["event_type"], ra.EVENT_DRAIN_PROOF)
+            self.assertEqual(loaded["correlation_id"], "rst-write")
+
+    def test_emit_never_raises(self):
+        self.assertFalse(
+            ra.emit_restart_event({"action": "x"}, path="/no/such/dir/audit.log")
+        )
+
+    def test_require_audit_denies_privileged_when_write_fails_and_enabled(self):
+        deny = ra.require_audit_or_deny(
+            privileged=True, written=False, audit_enabled=True
+        )
+        self.assertEqual(len(deny), 1)
+        self.assertIn("fail closed", deny[0])
+
+    def test_require_audit_allows_when_audit_disabled(self):
+        # Rollout policy: enable audit before enforcing deny-on-audit-fail.
+        deny = ra.require_audit_or_deny(
+            privileged=True, written=False, audit_enabled=False
+        )
+        self.assertEqual(deny, [])
+
+    def test_require_audit_noop_for_non_privileged(self):
+        deny = ra.require_audit_or_deny(
+            privileged=False, written=False, audit_enabled=True
+        )
+        self.assertEqual(deny, [])
+
+    def test_require_audit_allows_when_written(self):
+        deny = ra.require_audit_or_deny(
+            privileged=True, written=True, audit_enabled=True
+        )
+        self.assertEqual(deny, [])
+
+
+class TestIncidents(unittest.TestCase):
+    def test_break_glass_descriptor(self):
+        desc = ra.build_incident_descriptor(
+            kind=ra.INCIDENT_BREAK_GLASS,
+            reasons=["break-glass authorized"],
+            correlation_id="rst-bg",
+            requesting_session_id="s1",
+            restart_class="full_mcp_restart",
+            remote="prgs",
+            org="O",
+            repo="R",
+        )
+        self.assertEqual(desc["kind"], ra.INCIDENT_BREAK_GLASS)
+        self.assertIn("Break-glass", desc["title"])
+        self.assertIn("mcp-health", desc["labels"])
+        self.assertEqual(desc["source"], "restart_audit#665")
+
+    def test_incident_body_redacts_and_includes_correlation(self):
+        desc = ra.build_incident_descriptor(
+            kind=ra.INCIDENT_FAILED_DRAIN,
+            reasons=["token secret-xyz failed proof"],
+            correlation_id="rst-body",
+            remote="prgs",
+            org="O",
+            repo="R",
+            proof_id="proof-1",
+        )
+        body = ra.incident_body(desc)
+        self.assertIn("rst-body", body)
+        self.assertIn("proof-1", body)
+        self.assertIn("mcp-restart-incident:v1", body)
+        self.assertNotIn("secret-xyz", body)
+
+    def test_materialize_dry_run(self):
+        desc = ra.build_incident_descriptor(
+            kind=ra.INCIDENT_FAILED_DRAIN,
+            reasons=["denied"],
+            correlation_id="rst-dr",
+        )
+        result = ra.materialize_incident(desc, create_issue_fn=lambda **k: {}, dry_run=True)
+        self.assertFalse(result["created"])
+        self.assertTrue(result["dry_run"])
+        self.assertIn("dry-run", result["reasons"][0])
+
+    def test_materialize_without_create_fn(self):
+        desc = ra.build_incident_descriptor(
+            kind=ra.INCIDENT_FAILED_DRAIN,
+            reasons=["denied"],
+            correlation_id="rst-nfn",
+        )
+        result = ra.materialize_incident(desc, create_issue_fn=None)
+        self.assertFalse(result["created"])
+        self.assertIn("create_issue_fn not provided", result["reasons"][0])
+
+    def test_materialize_creates_issue(self):
+        created = {}
+
+        def _create(*, title, body, labels, org=None, repo=None, **_kw):
+            created["title"] = title
+            created["body"] = body
+            created["labels"] = labels
+            created["org"] = org
+            created["repo"] = repo
+            return {"number": 999}
+
+        desc = ra.build_incident_descriptor(
+            kind=ra.INCIDENT_BREAK_GLASS,
+            reasons=["break-glass"],
+            correlation_id="rst-create",
+            org="O",
+            repo="R",
+        )
+        result = ra.materialize_incident(desc, create_issue_fn=_create)
+        self.assertTrue(result["created"])
+        self.assertEqual(result["issue_number"], 999)
+        self.assertIn("Break-glass", created["title"])
+        self.assertIn("rst-create", created["body"])
+        self.assertEqual(created["org"], "O")
+
+    def test_materialize_never_raises_on_create_failure(self):
+        def _boom(**_kw):
+            raise RuntimeError("token secret-xyz network")
+
+        desc = ra.build_incident_descriptor(
+            kind=ra.INCIDENT_FAILED_DRAIN,
+            reasons=["x"],
+            correlation_id="rst-boom",
+        )
+        result = ra.materialize_incident(desc, create_issue_fn=_boom)
+        self.assertFalse(result["created"])
+        self.assertIn("failed", result["reasons"][0])
+        self.assertNotIn("secret-xyz", result["reasons"][0])
+
+
+class TestIncidentFromApplyGate(unittest.TestCase):
+    def test_break_glass_always_incident(self):
+        desc = ra.incident_from_apply_gate(
+            gate_payload={"allow": True, "reasons": [], "proof_id": None},
+            break_glass=True,
+            correlation_id="rst-bg2",
+            requesting_session_id="s",
+            restart_class="full_mcp_restart",
+            remote="prgs",
+            org="O",
+            repo="R",
+        )
+        self.assertIsNotNone(desc)
+        self.assertEqual(desc["kind"], ra.INCIDENT_BREAK_GLASS)
+
+    def test_failed_drain_from_gate_incident(self):
+        desc = ra.incident_from_apply_gate(
+            gate_payload={
+                "allow": False,
+                "drain_gate_allow": False,
+                "reasons": ["proof expired"],
+                "incident": {
+                    "reasons": ["proof expired"],
+                    "proof_id": "p1",
+                },
+            },
+            break_glass=False,
+            correlation_id="rst-fd",
+            requesting_session_id="s",
+            restart_class="full_mcp_restart",
+            remote="prgs",
+            org="O",
+            repo="R",
+        )
+        self.assertIsNotNone(desc)
+        self.assertEqual(desc["kind"], ra.INCIDENT_FAILED_DRAIN)
+        self.assertEqual(desc["proof_id"], "p1")
+
+    def test_allow_without_break_glass_no_incident(self):
+        desc = ra.incident_from_apply_gate(
+            gate_payload={
+                "allow": True,
+                "drain_gate_allow": True,
+                "reasons": [],
+            },
+            break_glass=False,
+            correlation_id="rst-ok",
+            requesting_session_id="s",
+            restart_class="full_mcp_restart",
+            remote="prgs",
+            org="O",
+            repo="R",
+        )
+        self.assertIsNone(desc)
+
+
+class TestRecordLifecycle(unittest.TestCase):
+    def test_record_emits_and_materializes(self):
+        created = []
+
+        def _create(**kwargs):
+            created.append(kwargs)
+            return {"number": 42}
+
+        with tempfile.TemporaryDirectory() as d:
+            path = os.path.join(d, "audit.log")
+            with patch.dict(os.environ, {"GITEA_AUDIT_LOG": path}, clear=False):
+                incident = ra.build_incident_descriptor(
+                    kind=ra.INCIDENT_BREAK_GLASS,
+                    reasons=["bg"],
+                    correlation_id="rst-lc",
+                    org="O",
+                    repo="R",
+                )
+                out = ra.record_restart_lifecycle(
+                    event_type=ra.EVENT_BREAK_GLASS,
+                    outcome="break_glass",
+                    correlation_id="rst-lc",
+                    remote="prgs",
+                    org="O",
+                    repo="R",
+                    privileged=True,
+                    create_incident=incident,
+                    create_issue_fn=_create,
+                    audit_path=path,
+                )
+        self.assertTrue(out["audit_written"])
+        self.assertEqual(out["deny_reasons"], [])
+        self.assertTrue(out["incident_result"]["created"])
+        self.assertEqual(out["incident_result"]["issue_number"], 42)
+        self.assertEqual(len(created), 1)
+
+    def test_privileged_deny_when_audit_write_fails(self):
+        with patch.dict(
+            os.environ, {"GITEA_AUDIT_LOG": "/no/such/dir/a.log"}, clear=False
+        ):
+            with patch("restart_audit.emit_restart_event", return_value=False):
+                with patch("gitea_audit.audit_enabled", return_value=True):
+                    out = ra.record_restart_lifecycle(
+                        event_type=ra.EVENT_APPLY_GATE,
+                        outcome="deny",
+                        correlation_id="rst-deny",
+                        privileged=True,
+                        audit_path="/no/such/dir/a.log",
+                    )
+        self.assertFalse(out["audit_written"])
+        self.assertEqual(len(out["deny_reasons"]), 1)
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_restart_classes.py

@@ -0,0 +1,232 @@
+"""Permission, drain, routing, and audit matrix for restart classes (#663)."""
+
+from __future__ import annotations
+
+import os
+from datetime import datetime, timezone
+
+import restart_coordinator as rc
+
+NOW = datetime(2026, 7, 24, 20, 0, tzinfo=timezone.utc)
+
+
+def _inventory() -> dict:
+    return {
+        "inventory_complete": True,
+        "sessions": [
+            {
+                "session_id": "requester",
+                "role": "author",
+                "profile": "prgs-author",
+                "pid": os.getpid(),
+                "status": "active",
+                "last_heartbeat_at": NOW.isoformat(),
+            },
+            {
+                "session_id": "reviewer",
+                "role": "reviewer",
+                "profile": "prgs-reviewer",
+                "pid": os.getpid(),
+                "status": "active",
+                "last_heartbeat_at": NOW.isoformat(),
+            },
+        ],
+        "leases": [
+            {
+                "lease_id": "review-lease",
+                "session_id": "reviewer",
+                "role": "reviewer",
+                "phase": "reviewing",
+                "work_kind": "pr",
+                "work_number": 900,
+                "worktree_path": "/tmp/review-900",
+                "freshness": {"freshness": "active"},
+            }
+        ],
+    }
+
+
+def _evaluate(
+    restart_class: rc.RestartClass,
+    *,
+    role: str = "controller",
+    permissions: tuple[str, ...] | None = None,
+    approved: bool = True,
+    operator: bool = True,
+    **targets,
+):
+    return rc.evaluate_restart_impact(
+        _inventory(),
+        now=NOW,
+        requesting_session_id="requester",
+        restart_class=restart_class,
+        requester_role=role,
+        requester_permissions=(
+            permissions if permissions is not None
+            else rc.permissions_for_role(role)
+        ),
+        controller_approved=approved,
+        operator_authorized=operator,
+        **targets,
+    )
+
+
+def test_policy_table_covers_exactly_all_nine_classes():
+    assert set(rc.RESTART_CLASS_POLICIES) == set(rc.RestartClass)
+    assert len(rc.RESTART_CLASS_POLICIES) == 9
+    for restart_class, policy in rc.RESTART_CLASS_POLICIES.items():
+        assert policy.restart_class is restart_class
+        assert policy.required_permission
+        assert policy.expected_blast_radius in {
+            rc.BLAST_NONE, rc.BLAST_LOW, rc.BLAST_MEDIUM, rc.BLAST_HIGH
+        }
+        assert policy.drain_requirement
+        assert policy.approval_requirement
+        assert policy.audit_requirement
+        assert policy.recovery_behavior
+
+
+def test_permission_matrix_allows_each_class_with_exact_permission():
+    targets = {
+        rc.RestartClass.WORKER_RESTART: {"target_session_id": "reviewer"},
+        rc.RestartClass.ROLE_RUNTIME_RESTART: {"target_role": "reviewer"},
+        rc.RestartClass.CONNECTOR_RESTART: {"target_connector": "github"},
+    }
+    for restart_class, policy in rc.RESTART_CLASS_POLICIES.items():
+        report = _evaluate(
+            restart_class,
+            permissions=(policy.required_permission,),
+            **targets.get(restart_class, {}),
+        )
+        assert report.permission_authorized, restart_class
+        assert report.role_authorized, restart_class
+        assert report.approval_satisfied, restart_class
+        assert report.audit_record["restart_class"] == restart_class.value
+        assert (
+            report.audit_record["required_permission"]
+            == policy.required_permission
+        )
+
+
+def test_missing_or_nearby_permission_denies():
+    report = _evaluate(
+        rc.RestartClass.ROLE_RUNTIME_RESTART,
+        permissions=("mcp.restart.worker.request",),
+        target_role="reviewer",
+    )
+    assert report.verdict == rc.VERDICT_UNSAFE
+    assert not report.allow_restart
+    assert not report.permission_authorized
+    assert any("missing required permission" in r for r in report.reasons)
+
+
+def test_unknown_restart_class_denies_fail_closed():
+    report = rc.evaluate_restart_impact(
+        _inventory(),
+        now=NOW,
+        restart_class="surprise_reboot",
+        requester_role="admin",
+        requester_permissions=("mcp.restart.host.request",),
+        controller_approved=True,
+        operator_authorized=True,
+    )
+    assert report.verdict == rc.VERDICT_UNSAFE
+    assert not report.allow_restart
+    assert report.restart_policy == {}
+    assert any("unknown restart class" in r for r in report.reasons)
+
+
+def test_worker_roles_cannot_request_full_or_host_restart():
+    for role in rc.WORKER_ROLES:
+        granted = rc.permissions_for_role(role)
+        assert "mcp.restart.full.request" not in granted
+        assert "mcp.restart.host.request" not in granted
+        report = _evaluate(
+            rc.RestartClass.FULL_MCP_RESTART,
+            role=role,
+            permissions=granted,
+        )
+        assert not report.role_authorized
+        assert not report.allow_restart
+
+
+def test_controller_approval_is_independent_of_permission():
+    report = _evaluate(
+        rc.RestartClass.WORKER_RESTART,
+        approved=False,
+        target_session_id="reviewer",
+    )
+    assert report.permission_authorized
+    assert not report.approval_satisfied
+    assert not report.allow_restart
+
+
+def test_narrow_classes_do_not_inherit_full_drain_or_peer_lease_block():
+    for restart_class in (
+        rc.RestartClass.CLIENT_RECONNECT,
+        rc.RestartClass.SESSION_RECONNECT,
+        rc.RestartClass.CONFIGURATION_RELOAD,
+    ):
+        report = _evaluate(restart_class)
+        assert not report.restart_policy["full_drain_required"]
+        assert report.counts["leases_disruptive"] == 0
+        assert report.counts["sessions_live_other"] == 0
+        assert report.counts["critical_sections"] == 0
+        assert report.counts["mutations"] == 0
+        assert report.allow_restart, (restart_class, report.reasons)
+
+
+def test_client_reconnect_does_not_wait_for_unrelated_terminal_lock():
+    inventory = _inventory()
+    inventory["terminal_lock"] = {"terminal_pr": 901}
+    report = rc.evaluate_restart_impact(
+        inventory,
+        now=NOW,
+        requesting_session_id="requester",
+        restart_class=rc.RestartClass.CLIENT_RECONNECT,
+        requester_role="author",
+        requester_permissions=rc.permissions_for_role("author"),
+    )
+    assert report.allow_restart
+    assert report.terminal_lock is None
+
+
+def test_scoped_restart_only_counts_named_target():
+    report = _evaluate(
+        rc.RestartClass.ROLE_RUNTIME_RESTART,
+        target_role="author",
+    )
+    assert report.counts["leases_disruptive"] == 0
+    assert report.affected_prs == []
+    assert report.allow_restart
+
+    reviewer = _evaluate(
+        rc.RestartClass.ROLE_RUNTIME_RESTART,
+        target_role="reviewer",
+    )
+    assert reviewer.counts["leases_disruptive"] == 1
+    assert reviewer.affected_prs == [900]
+    assert not reviewer.allow_restart
+
+
+def test_missing_scoped_target_denies_instead_of_widening():
+    for restart_class in (
+        rc.RestartClass.WORKER_RESTART,
+        rc.RestartClass.ROLE_RUNTIME_RESTART,
+        rc.RestartClass.CONNECTOR_RESTART,
+    ):
+        report = _evaluate(restart_class)
+        assert not report.allow_restart
+        assert any("target required" in r for r in report.reasons)
+
+
+def test_only_full_and_host_classes_require_full_drain():
+    requiring_full = {
+        restart_class
+        for restart_class, policy in rc.RESTART_CLASS_POLICIES.items()
+        if policy.full_drain_required
+    }
+    assert requiring_full == {
+        rc.RestartClass.FULL_MCP_RESTART,
+        rc.RestartClass.HOST_RESTART,
+    }

tests/test_webui_sanctioned_restart.py

@@ -444,6 +444,12 @@ class TestAuditEmission(unittest.TestCase):
         )
         self.assertEqual(record["target"]["namespace"], NAMESPACE)
         self.assertEqual(record["target"]["mode"], "restart")
+        self.assertEqual(
+            record["target"]["restart_class"], "role_runtime_restart"
+        )
+        self.assertEqual(
+            record["metadata"]["restart_class"], "role_runtime_restart"
+        )
         self.assertEqual(record["result"], console_audit.RESULT_ALLOWED)
         self.assertEqual(record["actor"]["subject"], "[email protected]")
         self.assertFalse(record["metadata"]["process_kill_executed"])

tests/test_webui_sessions_view.py

@@ -0,0 +1,739 @@
+"""Tests for the Runtime and session view (Phase 1, #641).
+
+Covers clean and stale session rendering, contamination marker surfacing,
+worktree binding display, sanctioned recovery links (no pkill), nav/live
+status, and the JSON API export.
+
+Also pins the two invariants a reviewer found violated at head a81db754:
+degraded ownership sections must render as *unknown* rather than as an
+affirmative "none"/"unbound", and contamination payload text must be redacted
+at the display boundary rather than trusted from the write-time denylist.
+"""
+
+from __future__ import annotations
+
+import os
+import sys
+import unittest
+from pathlib import Path
+from unittest import mock
+
+sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
+
+from tests.webui_testclient import TestClient
+
+from webui.app import create_app
+from webui.inventory import (
+    AUTHORITY_CONTROL_PLANE_DB,
+    AUTHORITY_FILESYSTEM,
+    InventorySection,
+    InventorySnapshot,
+    STATUS_DEGRADED,
+    STATUS_OK,
+    STATUS_UNAVAILABLE,
+)
+from webui.nav import NAV_GROUPS, STUB_PAGES, iter_nav_items
+from webui.runtime_health import FileHash, RuntimeSnapshot
+from webui.session_loader import (
+    ContaminationMarker,
+    SessionRow,
+    SessionViewSnapshot,
+    _build_session_rows,
+    _inspect_contamination,
+    load_session_view_snapshot,
+    snapshot_to_dict,
+)
+from webui.session_views import render_sessions_page
+
+
+def _runtime(
+    *,
+    stale: str | None = None,
+    profile: str = "prgs-author",
+    role: str = "author",
+) -> RuntimeSnapshot:
+    return RuntimeSnapshot(
+        project_id="gitea-tools",
+        repo_root="/tmp/repo",
+        remote="prgs",
+        host="gitea.prgs.cc",
+        profile_name=profile,
+        role_kind=role,
+        config_model="v2-contexts",
+        profile_mode="dynamic-profile",
+        profile_source="config file profile",
+        authenticated_username="jcwalker3",
+        identity_error=None,
+        repo_sha="a" * 40,
+        remote_master_sha="a" * 40,
+        commits_behind_master=0,
+        stale_runtime_warning=stale,
+        shell_health={"shell_use_allowed": True, "consecutive_spawn_failures": 0},
+        workflow_hashes=(
+            FileHash(label="SKILL.md", path="skills/llm-project-workflow/SKILL.md", sha256="abc"),
+        ),
+        schema_hashes=(),
+        restart_guidance="docs/mcp-namespace-eof-recovery.md",
+        fetch_error=None,
+    )
+
+
+def _inventory(
+    *,
+    sessions: tuple[dict, ...] = (),
+    leases: tuple[dict, ...] = (),
+    locks: tuple[dict, ...] = (),
+    worktrees: tuple[dict, ...] = (),
+    namespaces: tuple[dict, ...] = (),
+    statuses: dict[str, str] | None = None,
+) -> InventorySnapshot:
+    """Build a snapshot; ``statuses`` degrades named sections (default all ok)."""
+    status_of = statuses or {}
+
+    def _status(name: str) -> str:
+        return status_of.get(name, STATUS_OK)
+
+    sections = (
+        InventorySection(
+            name="sessions",
+            authority=AUTHORITY_CONTROL_PLANE_DB,
+            status=_status("sessions"),
+            items=sessions,
+        ),
+        InventorySection(
+            name="leases",
+            authority=AUTHORITY_CONTROL_PLANE_DB,
+            status=_status("leases"),
+            items=leases,
+        ),
+        InventorySection(
+            name="locks",
+            authority=AUTHORITY_FILESYSTEM,
+            status=_status("locks"),
+            items=locks,
+        ),
+        InventorySection(
+            name="worktrees",
+            authority=AUTHORITY_FILESYSTEM,
+            status=_status("worktrees"),
+            items=worktrees,
+        ),
+        InventorySection(
+            name="namespaces",
+            authority=AUTHORITY_FILESYSTEM,
+            status=_status("namespaces"),
+            items=namespaces
+            or (
+                {
+                    "profile_name": "prgs-author",
+                    "role": "author",
+                    "mcp_namespace": "gitea-author",
+                    "capability_summary": {
+                        "can_author": True,
+                        "can_review": False,
+                        "can_merge": False,
+                    },
+                    "active": True,
+                },
+            ),
+            reason="only the profile serving this web process is observable",
+        ),
+    )
+    index = {section.name: section for section in sections}
+    return InventorySnapshot(
+        generated_at="2026-07-25T00:00:00+00:00",
+        sections=sections,
+        collisions=(),
+        correlations=(),
+        scan_ms=1.0,
+        _section_index=index,
+    )
+
+
+def _clean_session() -> dict:
+    return {
+        "session_id": "prgs-author-111-clean",
+        "role": "author",
+        "profile": "prgs-author",
+        "namespace": "gitea-author",
+        "pid": 1111,
+        "pid_alive": True,
+        "status": "active",
+        "started_at": "2026-07-25T00:00:00Z",
+        "last_heartbeat_at": "2026-07-25T01:00:00Z",
+    }
+
+
+def _stale_session() -> dict:
+    return {
+        "session_id": "prgs-author-222-stale",
+        "role": "author",
+        "profile": "prgs-author",
+        "namespace": "gitea-author",
+        "pid": 2222,
+        "pid_alive": False,
+        "status": "active",
+        "started_at": "2026-07-24T00:00:00Z",
+        "last_heartbeat_at": "2026-07-24T01:00:00Z",
+    }
+
+
+class TestBuildSessionRows(unittest.TestCase):
+    def test_clean_session_has_no_stale_or_contamination_flags(self):
+        inventory = _inventory(
+            sessions=(_clean_session(),),
+            leases=(
+                {
+                    "lease_id": "lease-clean",
+                    "session_id": "prgs-author-111-clean",
+                    "status": "active",
+                    "expired": False,
+                    "work_kind": "issue",
+                    "work_number": 641,
+                },
+            ),
+            locks=(
+                {
+                    "issue_number": 641,
+                    "branch_name": "feat/issue-641-runtime-session-view",
+                    "worktree_path": "~/Development/Gitea-Tools/branches/feat-issue-641",
+                    "live": True,
+                },
+            ),
+        )
+        rows = _build_session_rows(inventory, contamination=())
+        self.assertEqual(len(rows), 1)
+        row = rows[0]
+        self.assertEqual(row.session_id, "prgs-author-111-clean")
+        self.assertEqual(row.role, "author")
+        self.assertEqual(row.namespace, "gitea-author")
+        self.assertEqual(row.pid_alive, True)
+        self.assertEqual(row.lease_ids, ("lease-clean",))
+        self.assertEqual(row.work_refs, ("issue#641",))
+        self.assertTrue(row.worktree_paths)
+        self.assertEqual(row.stale_flags, ())
+        self.assertEqual(row.contamination_flags, ())
+
+    def test_stale_session_flags_dead_pid(self):
+        inventory = _inventory(sessions=(_stale_session(),))
+        rows = _build_session_rows(inventory, contamination=())
+        self.assertEqual(rows[0].stale_flags, ("pid-dead",))
+
+    def test_contamination_marker_binds_to_session(self):
+        inventory = _inventory(sessions=(_clean_session(),))
+        marker = ContaminationMarker(
+            kind="runtime_recovery_contamination",
+            on_disk=True,
+            has_payload=True,
+            summary="manual daemon kill",
+            reason_class="manual_daemon_kill",
+            session_id="prgs-author-111-clean",
+            role="author",
+            command_summary="pkill -f mcp_server.py",
+            cleared=False,
+        )
+        rows = _build_session_rows(inventory, contamination=(marker,))
+        self.assertIn("runtime_recovery_contamination", rows[0].contamination_flags)
+
+    def test_process_wide_contamination_surfaces_on_all_sessions(self):
+        inventory = _inventory(sessions=(_clean_session(), _stale_session()))
+        marker = ContaminationMarker(
+            kind="stable_branch_contamination",
+            on_disk=True,
+            has_payload=True,
+            summary="direct master push attempt",
+            reason_class="stable_branch_push",
+            session_id=None,
+            cleared=False,
+        )
+        rows = _build_session_rows(inventory, contamination=(marker,))
+        self.assertEqual(len(rows), 2)
+        for row in rows:
+            self.assertTrue(
+                any("stable_branch_contamination" in f for f in row.contamination_flags)
+            )
+
+
+class TestRenderSessionsPage(unittest.TestCase):
+    def _snapshot(
+        self,
+        *,
+        sessions: tuple[dict, ...],
+        contamination: tuple[ContaminationMarker, ...] = (),
+        stale_runtime: str | None = None,
+    ) -> SessionViewSnapshot:
+        inventory = _inventory(
+            sessions=sessions,
+            leases=(
+                {
+                    "lease_id": "lease-1",
+                    "session_id": sessions[0]["session_id"] if sessions else "",
+                    "status": "active",
+                    "expired": False,
+                    "work_kind": "issue",
+                    "work_number": 641,
+                },
+            )
+            if sessions
+            else (),
+            locks=(
+                {
+                    "issue_number": 641,
+                    "worktree_path": "branches/feat-issue-641",
+                },
+            )
+            if sessions
+            else (),
+            worktrees=(
+                {
+                    "rel_path": "branches/feat-issue-641",
+                    "branch": "feat/issue-641-runtime-session-view",
+                    "classification": "active_issue_work",
+                    "registered_worktree": True,
+                    "dirty": False,
+                },
+            ),
+        )
+        rows = _build_session_rows(inventory, contamination)
+        return SessionViewSnapshot(
+            runtime=_runtime(stale=stale_runtime),
+            inventory=inventory,
+            sessions=rows,
+            contamination_markers=contamination,
+        )
+
+    def test_clean_session_render(self):
+        html = render_sessions_page(self._snapshot(sessions=(_clean_session(),)))
+        self.assertIn("Runtime and sessions", html)
+        self.assertIn("prgs-author-111-clean", html)
+        self.assertIn("gitea-author", html)
+        self.assertIn("branches/feat-issue-641", html)
+        self.assertIn("Sanctioned recovery", html)
+        self.assertIn("docs/mcp-namespace-eof-recovery.md", html)
+        # Recovery section must name reconnect and forbid manual kill.
+        recovery_idx = html.lower().find("sanctioned recovery")
+        self.assertGreaterEqual(recovery_idx, 0)
+        recovery = html[recovery_idx:].lower()
+        self.assertIn("reconnect", recovery)
+        self.assertIn("contamination", recovery)
+        self.assertIn("not recovery", recovery)
+        self.assertNotIn("run pkill", recovery)
+        self.assertNotIn("killall", recovery)
+
+    def test_stale_session_render(self):
+        html = render_sessions_page(self._snapshot(sessions=(_stale_session(),)))
+        self.assertIn("prgs-author-222-stale", html)
+        self.assertIn("pid-dead", html)
+        self.assertIn("badge-stale", html)
+
+    def test_contamination_render_is_not_silent(self):
+        marker = ContaminationMarker(
+            kind="runtime_recovery_contamination",
+            on_disk=True,
+            has_payload=True,
+            summary="manual kill",
+            reason_class="manual_daemon_kill",
+            session_id="prgs-author-111-clean",
+            command_summary="pkill -f mcp_server.py",
+            cleared=False,
+        )
+        html = render_sessions_page(
+            self._snapshot(sessions=(_clean_session(),), contamination=(marker,))
+        )
+        self.assertIn("Contamination markers", html)
+        self.assertIn("runtime_recovery_contamination", html)
+        self.assertIn("ACTIVE", html)
+        self.assertIn("badge-blocked", html)
+
+    def test_stale_runtime_banner(self):
+        html = render_sessions_page(
+            self._snapshot(
+                sessions=(_clean_session(),),
+                stale_runtime="server behind master by 3 commits",
+            )
+        )
+        self.assertIn("Stale runtime", html)
+        self.assertIn("server behind master", html)
+
+
+class TestSessionLoaderComposition(unittest.TestCase):
+    def test_load_with_injected_sources(self):
+        inventory = _inventory(sessions=(_clean_session(), _stale_session()))
+        snap = load_session_view_snapshot(
+            load_runtime=lambda: _runtime(),
+            load_inventory=lambda: inventory,
+            inspect_contamination=lambda **_k: {
+                "on_disk": False,
+                "has_payload": False,
+                "summary": "absent",
+            },
+            load_contamination_payload=lambda **_k: None,
+        )
+        self.assertEqual(len(snap.sessions), 2)
+        self.assertEqual(snap.stale_session_count, 1)
+        self.assertEqual(snap.contaminated_session_count, 0)
+        data = snapshot_to_dict(snap)
+        self.assertEqual(data["view"], "runtime-sessions")
+        self.assertEqual(data["issue"], 641)
+        self.assertTrue(data["read_only"])
+        self.assertEqual(data["session_counts"]["total"], 2)
+        self.assertEqual(data["session_counts"]["stale"], 1)
+        self.assertIn("recovery_docs", data)
+
+
+class TestSessionsRoutes(unittest.TestCase):
+    def setUp(self):
+        self.client = TestClient(create_app())
+        inventory = _inventory(
+            sessions=(_clean_session(), _stale_session()),
+            leases=(
+                {
+                    "lease_id": "lease-x",
+                    "session_id": "prgs-author-111-clean",
+                    "status": "active",
+                    "expired": False,
+                    "work_kind": "issue",
+                    "work_number": 641,
+                },
+            ),
+            locks=(
+                {
+                    "issue_number": 641,
+                    "worktree_path": "branches/feat-issue-641",
+                },
+            ),
+            worktrees=(
+                {
+                    "rel_path": "branches/feat-issue-641",
+                    "branch": "feat/issue-641-runtime-session-view",
+                    "classification": "active_issue_work",
+                    "registered_worktree": True,
+                    "dirty": False,
+                },
+            ),
+        )
+        rows = _build_session_rows(inventory, contamination=())
+        self.snapshot = SessionViewSnapshot(
+            runtime=_runtime(stale="stale for test"),
+            inventory=inventory,
+            sessions=rows,
+            contamination_markers=(),
+        )
+        self._patch = mock.patch(
+            "webui.app.load_session_view_snapshot",
+            return_value=self.snapshot,
+        )
+        self._patch.start()
+
+    def tearDown(self):
+        self._patch.stop()
+
+    def test_sessions_page_live(self):
+        response = self.client.get("/sessions")
+        self.assertEqual(response.status_code, 200)
+        self.assertIn("Runtime and sessions", response.text)
+        self.assertIn("prgs-author-111-clean", response.text)
+        self.assertIn("prgs-author-222-stale", response.text)
+        self.assertIn("pid-dead", response.text)
+        self.assertIn("Sanctioned recovery", response.text)
+        self.assertNotIn("Phase 1 shell placeholder", response.text)
+        self.assertNotIn("child issue of #425", response.text.lower())
+
+    def test_api_sessions_json(self):
+        for path in ("/api/sessions", "/api/v1/sessions"):
+            response = self.client.get(path)
+            self.assertEqual(response.status_code, 200, path)
+            data = response.json()
+            self.assertEqual(data["view"], "runtime-sessions")
+            self.assertEqual(data["session_counts"]["total"], 2)
+            self.assertEqual(data["session_counts"]["stale"], 1)
+            self.assertTrue(data["read_only"])
+            self.assertEqual(data["mutations"], [])
+
+    def test_nav_marks_sessions_live(self):
+        sessions_items = [
+            item for item in iter_nav_items() if item.href == "/sessions"
+        ]
+        self.assertEqual(len(sessions_items), 1)
+        self.assertEqual(sessions_items[0].status, "live")
+        self.assertNotIn("/sessions", STUB_PAGES)
+        # Home page should not mark Sessions as stub.
+        home = self.client.get("/")
+        self.assertEqual(home.status_code, 200)
+        self.assertIn('href="/sessions"', home.text)
+        # Stub marker only appears next to remaining stub destinations.
+        self.assertNotIn(
+            'href="/sessions">Sessions</a> <span class="muted">(stub)</span>',
+            home.text,
+        )
+
+
+class TestDegradedOwnershipAuthority(unittest.TestCase):
+    """B1: a section that could not be read must never render as absence."""
+
+    def _snapshot(self, inventory: InventorySnapshot) -> SessionViewSnapshot:
+        return SessionViewSnapshot(
+            runtime=_runtime(),
+            inventory=inventory,
+            sessions=_build_session_rows(inventory, contamination=()),
+            contamination_markers=(),
+        )
+
+    def test_unavailable_leases_mark_row_authority_unproven(self):
+        inventory = _inventory(
+            sessions=(_clean_session(),),
+            statuses={"leases": STATUS_UNAVAILABLE},
+        )
+        row = _build_session_rows(inventory, contamination=())[0]
+        self.assertEqual(row.lease_ids, ())
+        self.assertEqual(row.lease_authority, STATUS_UNAVAILABLE)
+        # Worktree binding is correlated through lease work numbers, so it
+        # inherits the unreadable lease section.
+        self.assertEqual(row.worktree_authority, STATUS_UNAVAILABLE)
+        self.assertFalse(row.ownership_authority_complete)
+
+    def test_readable_locks_are_not_reported_unbound_when_leases_degrade(self):
+        # The narrow variant: locks hold a real worktree_path and read cleanly,
+        # but the lease section that supplies the correlating work number does
+        # not. The row must say unknown, not "unbound".
+        inventory = _inventory(
+            sessions=(_clean_session(),),
+            locks=(
+                {
+                    "issue_number": 641,
+                    "worktree_path": "branches/feat-issue-641",
+                },
+            ),
+            statuses={"leases": STATUS_DEGRADED},
+        )
+        row = _build_session_rows(inventory, contamination=())[0]
+        self.assertEqual(row.worktree_paths, ())
+        self.assertEqual(row.worktree_authority, STATUS_DEGRADED)
+        self.assertFalse(row.ownership_authority_complete)
+
+    def test_degraded_render_says_unknown_not_none_or_unbound(self):
+        inventory = _inventory(
+            sessions=(_clean_session(),),
+            statuses={"leases": STATUS_UNAVAILABLE, "locks": STATUS_UNAVAILABLE},
+        )
+        html = render_sessions_page(self._snapshot(inventory))
+        self.assertIn("unknown (inventory unavailable)", html)
+        self.assertIn("authority unproven", html)
+        self.assertIn("Ownership authority incomplete", html)
+        # The affirmative-absence strings must be gone from the row entirely.
+        self.assertNotIn(">none<", html)
+        self.assertNotIn(">unbound<", html)
+
+    def test_clean_inventory_still_renders_affirmative_absence(self):
+        # Guards against over-correcting B1 into "everything is unknown".
+        inventory = _inventory(sessions=(_clean_session(),))
+        html = render_sessions_page(self._snapshot(inventory))
+        self.assertIn(">none<", html)
+        self.assertIn(">unbound<", html)
+        # The column legend mentions "unknown (inventory …)" as static copy, so
+        # assert on the per-row marker and the concrete statuses instead.
+        self.assertNotIn("authority unproven", html)
+        self.assertNotIn("unknown (inventory unavailable)", html)
+        self.assertNotIn("unknown (inventory degraded)", html)
+        self.assertNotIn("Ownership authority incomplete", html)
+
+    def test_json_export_carries_snapshot_and_per_row_authority(self):
+        inventory = _inventory(
+            sessions=(_clean_session(),),
+            statuses={"locks": STATUS_UNAVAILABLE},
+        )
+        data = snapshot_to_dict(self._snapshot(inventory))
+        self.assertFalse(data["ownership_authority_complete"])
+        self.assertEqual(
+            data["ownership_section_status"]["locks"], STATUS_UNAVAILABLE
+        )
+        self.assertEqual(data["ownership_section_status"]["leases"], STATUS_OK)
+        self.assertIn("unknown, not unowned", data["ownership_note"])
+
+        row = data["sessions"][0]
+        self.assertTrue(row["lease_authority_complete"])
+        self.assertFalse(row["worktree_authority_complete"])
+        self.assertEqual(row["worktree_authority"], STATUS_UNAVAILABLE)
+        self.assertFalse(row["ownership_authority_complete"])
+        self.assertIn("unknown, not unowned", row["ownership_note"])
+
+    def test_json_export_is_affirmative_when_every_source_reads(self):
+        inventory = _inventory(sessions=(_clean_session(),))
+        data = snapshot_to_dict(self._snapshot(inventory))
+        self.assertTrue(data["ownership_authority_complete"])
+        self.assertTrue(data["sessions"][0]["ownership_authority_complete"])
+
+    def test_missing_session_list_is_not_reported_as_no_sessions(self):
+        inventory = _inventory(statuses={"sessions": STATUS_UNAVAILABLE})
+        html = render_sessions_page(self._snapshot(inventory))
+        self.assertIn("could not be read", html)
+        self.assertIn("not evidence that no sessions exist", html)
+
+    def test_expired_lease_flags_row_as_stale(self):
+        inventory = _inventory(
+            sessions=(_clean_session(),),
+            leases=(
+                {
+                    "lease_id": "lease-expired-1",
+                    "session_id": "prgs-author-111-clean",
+                    "status": "active",
+                    "expired": True,
+                    "work_kind": "issue",
+                    "work_number": 641,
+                },
+            ),
+        )
+        row = _build_session_rows(inventory, contamination=())[0]
+        self.assertIn("lease-expired", row.stale_flags)
+        self.assertIn("active-lease-past-expiry", row.stale_flags)
+
+
+class TestContaminationRedaction(unittest.TestCase):
+    """B2: marker payload text is redacted at the display boundary."""
+
+    def _marker(self, payload: dict) -> ContaminationMarker:
+        return _inspect_contamination(
+            "runtime_recovery_contamination",
+            remote="prgs",
+            inspect=lambda **_k: {
+                "on_disk": True,
+                "has_payload": True,
+                "summary": "",
+            },
+            load=lambda **_k: payload,
+        )
+
+    def test_home_paths_are_collapsed(self):
+        home = os.path.expanduser("~")
+        marker = self._marker(
+            {"command_summary": f"pkill -f {home}/Development/Gitea-Tools/x.py"}
+        )
+        self.assertNotIn(home, marker.command_summary)
+        self.assertIn("~/Development/Gitea-Tools/x.py", marker.command_summary)
+
+    def test_secrets_missed_by_the_write_time_denylist_are_redacted(self):
+        # Each of these was verified in review to survive
+        # stable_branch_push_guard.redact_command untouched.
+        cases = (
+            ("curl -H 'X-Api-Key: SUPERSECRET123' https://example.invalid", "SUPERSECRET123"),
+            ("cmd --password hunter2 origin master", "hunter2"),
+            ("PRIVATE_KEY=abc123 python deploy.py", "abc123"),
+            ("fetch https://user:[email protected]/x.git", "user:pw"),
+        )
+        for raw, secret in cases:
+            with self.subTest(raw=raw):
+                marker = self._marker({"command_summary": raw})
+                self.assertNotIn(secret, marker.command_summary)
+                self.assertIn("[redacted]", marker.command_summary)
+
+    def test_command_summary_is_redacted_not_removed(self):
+        # It is legitimate #630 evidence: the operator must still see which
+        # daemon was killed.
+        marker = self._marker(
+            {
+                "command_summary": "pkill -f gitea_mcp_server.py",
+                "reason_class": "manual_daemon_kill",
+                "session_id": "prgs-author-111-clean",
+                "role": "author",
+            }
+        )
+        self.assertIn("pkill -f gitea_mcp_server.py", marker.command_summary)
+        self.assertEqual(marker.reason_class, "manual_daemon_kill")
+        self.assertEqual(marker.session_id, "prgs-author-111-clean")
+        self.assertEqual(marker.role, "author")
+
+    def test_rendered_page_exposes_no_home_path_from_a_marker(self):
+        home = os.path.expanduser("~")
+        marker = self._marker(
+            {
+                "command_summary": f"pkill -f {home}/Development/Gitea-Tools/x.py",
+                "reason_class": "manual_daemon_kill",
+            }
+        )
+        inventory = _inventory(sessions=(_clean_session(),))
+        html = render_sessions_page(
+            SessionViewSnapshot(
+                runtime=_runtime(),
+                inventory=inventory,
+                sessions=_build_session_rows(inventory, (marker,)),
+                contamination_markers=(marker,),
+            )
+        )
+        self.assertIn("Contamination markers", html)
+        self.assertNotIn(home, html)
+
+
+class TestSessionsPageEscaping(unittest.TestCase):
+    """Hostile values from every rendered source stay inert (N2)."""
+
+    HOSTILE = '<script>alert("xss")</script>'
+
+    def test_hostile_session_and_marker_values_are_escaped(self):
+        session = dict(_clean_session())
+        session["session_id"] = f"sid-{self.HOSTILE}"
+        session["role"] = self.HOSTILE
+        session["profile"] = self.HOSTILE
+        session["namespace"] = self.HOSTILE
+        session["status"] = self.HOSTILE
+        inventory = _inventory(
+            sessions=(session,),
+            leases=(
+                {
+                    "lease_id": self.HOSTILE,
+                    "session_id": session["session_id"],
+                    "status": "active",
+                    "expired": False,
+                    "work_kind": self.HOSTILE,
+                    "work_number": 641,
+                },
+            ),
+            locks=(
+                {
+                    "issue_number": 641,
+                    "worktree_path": self.HOSTILE,
+                },
+            ),
+        )
+        marker = ContaminationMarker(
+            kind="runtime_recovery_contamination",
+            on_disk=True,
+            has_payload=True,
+            summary=self.HOSTILE,
+            reason_class=self.HOSTILE,
+            session_id=session["session_id"],
+            role=self.HOSTILE,
+            command_summary=self.HOSTILE,
+            cleared=False,
+        )
+        html = render_sessions_page(
+            SessionViewSnapshot(
+                runtime=_runtime(),
+                inventory=inventory,
+                sessions=_build_session_rows(inventory, (marker,)),
+                contamination_markers=(marker,),
+            )
+        )
+        self.assertNotIn("<script>", html)
+        self.assertNotIn('alert("xss")', html)
+        self.assertIn("&lt;script&gt;", html)
+
+    def test_hostile_values_in_a_degraded_render_are_escaped(self):
+        inventory = _inventory(
+            sessions=(dict(_clean_session(), session_id=f"sid-{self.HOSTILE}"),),
+            statuses={"leases": STATUS_UNAVAILABLE, "locks": STATUS_DEGRADED},
+        )
+        html = render_sessions_page(
+            SessionViewSnapshot(
+                runtime=_runtime(),
+                inventory=inventory,
+                sessions=_build_session_rows(inventory, contamination=()),
+                contamination_markers=(),
+            )
+        )
+        self.assertNotIn("<script>", html)
+        self.assertIn("&lt;script&gt;", html)
+        self.assertIn("unknown (inventory", html)
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_webui_traffic_control.py

@@ -0,0 +1,408 @@
+"""Tests for web UI workflow traffic-control view (#640)."""
+
+import sys
+import unittest
+from pathlib import Path
+from unittest import mock
+
+sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
+
+from starlette.testclient import TestClient
+
+from webui.app import create_app
+from webui.traffic_loader import (
+    TrafficItem,
+    TrafficSnapshot,
+    load_traffic_snapshot,
+    snapshot_to_dict,
+)
+from webui.traffic_views import render_traffic_page
+from allocator_service import WorkCandidate
+
+
+class TestTrafficClassification(unittest.TestCase):
+    def test_runnable_candidate_classification(self):
+        cand = WorkCandidate(
+            kind="issue",
+            number=640,
+            state="open",
+            labels=("status:ready",),
+            title="Web Console: Workflow traffic-control view (Phase 1)",
+            priority=20,
+        )
+        snap = load_traffic_snapshot(candidates=[cand])
+        self.assertEqual(len(snap.runnable), 1)
+        self.assertEqual(snap.runnable[0].number, 640)
+        self.assertTrue(snap.runnable[0].is_safe)
+        self.assertEqual(snap.runnable[0].traffic_state, "runnable")
+
+    def test_blocked_dependency_candidate_classification(self):
+        cand = WorkCandidate(
+            kind="issue",
+            number=643,
+            state="open",
+            labels=("status:ready",),
+            title="Web Console: Requests & intent preview (Phase 2)",
+            priority=20,
+            dependency_unmet=True,
+            dependency_reason="issue#643 depends on unresolved issue(s) #640; they are not closed",
+        )
+        snap = load_traffic_snapshot(candidates=[cand])
+        self.assertEqual(len(snap.blocked), 1)
+        self.assertEqual(snap.blocked[0].number, 643)
+        self.assertFalse(snap.blocked[0].is_safe)
+        self.assertEqual(snap.blocked[0].traffic_state, "blocked")
+        self.assertIn("depends on unresolved issue(s) #640", snap.blocked[0].block_reason)
+
+    def test_leased_candidate_classification(self):
+        cand = WorkCandidate(
+            kind="issue",
+            number=640,
+            state="open",
+            labels=("status:in-progress",),
+            title="Web Console: Workflow traffic-control view (Phase 1)",
+            priority=20,
+        )
+        lease = {
+            "kind": "issue",
+            "number": 640,
+            "session_id": "prgs-author-12345",
+            "role": "author",
+            "status": "active",
+        }
+        snap = load_traffic_snapshot(candidates=[cand], leases=[lease])
+        self.assertEqual(len(snap.leased), 1)
+        self.assertEqual(snap.leased[0].number, 640)
+        self.assertEqual(snap.leased[0].traffic_state, "leased")
+        self.assertIsNotNone(snap.leased[0].lease_info)
+
+    def test_needs_controller_candidate_classification(self):
+        cand = WorkCandidate(
+            kind="issue",
+            number=700,
+            state="open",
+            labels=("status:blocked",),
+            title="Controller intervention needed",
+            priority=10,
+            blocked=True,
+        )
+        snap = load_traffic_snapshot(candidates=[cand])
+        self.assertEqual(len(snap.needs_controller), 1)
+        self.assertEqual(snap.needs_controller[0].number, 700)
+
+
+class TestTrafficLoader(unittest.TestCase):
+    def test_snapshot_to_dict_export(self):
+        cand = WorkCandidate(
+            kind="issue",
+            number=640,
+            state="open",
+            labels=("status:ready",),
+            title="Traffic control test",
+            priority=20,
+        )
+        snap = load_traffic_snapshot(candidates=[cand])
+        data = snapshot_to_dict(snap)
+        self.assertEqual(data["project_id"], "gitea-tools")
+        self.assertEqual(len(data["runnable"]), 1)
+        self.assertTrue(data["inventory_complete"])
+
+    def test_fail_closed_error_handling(self):
+        with mock.patch("webui.traffic_loader.load_queue_snapshot", side_effect=RuntimeError("Gitea connection failed")):
+            snap = load_traffic_snapshot()
+        self.assertIsNotNone(snap.fetch_error)
+        self.assertIn("Failed to load traffic state", snap.fetch_error)
+        self.assertEqual(len(snap.runnable), 0)
+        self.assertFalse(snap.inventory_complete)
+
+
+class TestTrafficLivePath(unittest.TestCase):
+    """Live path tests: inject QueueSnapshot + LeaseSnapshot (no candidates=).
+
+    Covers the production ``load_traffic_snapshot()`` branch that ``/traffic``
+    and ``/api/traffic`` actually execute (#640 B1–B5).
+    """
+
+    FULL_SHA = "069a9af7e6aa2c2994e07199d1b0814819457017"
+
+    def _queue(
+        self,
+        *,
+        prs=(),
+        issues=(),
+    ):
+        from webui.queue_loader import QueueSnapshot
+
+        return QueueSnapshot(
+            project_id="gitea-tools",
+            repo_label="Scaled-Tech-Consulting/Gitea-Tools",
+            prs=tuple(prs),
+            issues=tuple(issues),
+            pr_pagination=None,
+            issue_pagination=None,
+            fetch_error=None,
+        )
+
+    def _lease(
+        self,
+        *,
+        claim_inventory=None,
+        reviewer_leases=(),
+    ):
+        from webui.lease_loader import LeaseSnapshot
+
+        return LeaseSnapshot(
+            project_id="gitea-tools",
+            repo_label="Scaled-Tech-Consulting/Gitea-Tools",
+            issue_lock=None,
+            claim_inventory=claim_inventory or {"entries": [], "counts": {}},
+            reviewer_leases=tuple(reviewer_leases),
+            duplicate_prs=(),
+            duplicate_branches=(),
+            collision_history=(),
+            fetch_error=None,
+        )
+
+    def test_live_pr_uses_full_head_sha_and_is_runnable(self):
+        from webui.queue_loader import QueueItem
+
+        pr = QueueItem(
+            number=885,
+            title="traffic control",
+            badges=("in-review",),
+            extra={"head_sha": self.FULL_SHA[:12], "linked_issue": "640"},
+            signals={
+                "head_sha": self.FULL_SHA,
+                "mergeable": True,
+                "labels": (),
+                "linked_issue": 640,
+            },
+        )
+        q = self._queue(prs=[pr])
+        l = self._lease()
+        snap = load_traffic_snapshot(
+            fetch_queue_snapshot=lambda: q,
+            fetch_lease_snapshot=lambda: l,
+        )
+        self.assertIsNone(snap.fetch_error)
+        self.assertEqual(len(snap.runnable), 1)
+        item = snap.runnable[0]
+        self.assertEqual(item.kind, "pr")
+        self.assertEqual(item.number, 885)
+        self.assertEqual(item.head_sha, self.FULL_SHA)
+        self.assertNotEqual(item.head_sha, self.FULL_SHA[:12])
+        self.assertIsNone(item.block_reason)
+        self.assertEqual(len(snap.blocked), 0)
+
+    def test_live_pr_without_head_sha_is_blocked(self):
+        from webui.queue_loader import QueueItem
+
+        pr = QueueItem(
+            number=1,
+            title="missing pin",
+            badges=("open",),
+            extra={"head_sha": ""},
+            signals={"head_sha": "", "mergeable": True, "labels": ()},
+        )
+        snap = load_traffic_snapshot(
+            fetch_queue_snapshot=lambda: self._queue(prs=[pr]),
+            fetch_lease_snapshot=lambda: self._lease(),
+        )
+        self.assertEqual(len(snap.blocked) + len(snap.needs_controller), 1)
+        item = (snap.blocked or snap.needs_controller)[0]
+        self.assertIn("missing head_sha", (item.block_reason or "").lower())
+
+    def test_reviewer_lease_keys_by_pr_not_linked_issue(self):
+        from webui.queue_loader import QueueItem
+
+        pr = QueueItem(
+            number=885,
+            title="leased pr",
+            badges=("in-review",),
+            extra={"head_sha": self.FULL_SHA[:12]},
+            signals={"head_sha": self.FULL_SHA, "mergeable": True, "labels": ()},
+        )
+        issue = QueueItem(
+            number=640,
+            title="linked issue",
+            badges=("open",),
+            extra={},
+            signals={"labels": ()},
+        )
+        # Marker-shaped record: has both pr_number and issue_number; must
+        # attach to the PR only (B2).
+        reviewer_lease = {
+            "pr_number": 885,
+            "issue_number": 640,
+            "phase": "validating",
+            "reviewer_identity": "sysadmin",
+            "session_id": "review-sess-1",
+        }
+        snap = load_traffic_snapshot(
+            fetch_queue_snapshot=lambda: self._queue(prs=[pr], issues=[issue]),
+            fetch_lease_snapshot=lambda: self._lease(reviewer_leases=[reviewer_lease]),
+        )
+        leased_prs = [i for i in snap.leased if i.kind == "pr" and i.number == 885]
+        self.assertEqual(len(leased_prs), 1)
+        self.assertEqual(leased_prs[0].lease_info.get("pr_number"), 885)
+        # Issue 640 must not inherit the reviewer lease just because issue_number
+        # is present on the marker.
+        for item in list(snap.leased) + list(snap.runnable) + list(snap.blocked):
+            if item.kind == "issue" and item.number == 640:
+                self.assertIsNone(
+                    item.lease_info,
+                    "reviewer lease must not attach to linked issue #640",
+                )
+                break
+        else:
+            self.fail("expected issue #640 in traffic snapshot")
+
+    def test_claim_inventory_entries_key_marks_issue_leased(self):
+        from webui.queue_loader import QueueItem
+
+        issue = QueueItem(
+            number=640,
+            title="claimed issue",
+            badges=("claimed",),
+            extra={},
+            signals={"labels": ("status:in-progress",)},
+        )
+        inventory = {
+            "entries": [
+                {
+                    "issue_number": 640,
+                    "status": "active",
+                    "latest_heartbeat": {"session_id": "author-sess-9"},
+                    "reasons": ["claim has structured heartbeat proof"],
+                }
+            ],
+            "counts": {"active": 1},
+            "in_progress_total": 1,
+        }
+        snap = load_traffic_snapshot(
+            fetch_queue_snapshot=lambda: self._queue(issues=[issue]),
+            fetch_lease_snapshot=lambda: self._lease(claim_inventory=inventory),
+        )
+        leased_issues = [i for i in snap.leased if i.kind == "issue" and i.number == 640]
+        self.assertEqual(len(leased_issues), 1)
+        self.assertEqual(leased_issues[0].traffic_state, "leased")
+
+    def test_active_claims_key_is_ignored(self):
+        """B3 regression: fictional ``active_claims`` must not create lease_info."""
+        from webui.queue_loader import QueueItem
+
+        issue = QueueItem(
+            number=640,
+            title="open issue",
+            badges=("open",),
+            extra={},
+            signals={"labels": ()},
+        )
+        # Only the broken key — must NOT produce lease_info. Entries-less
+        # inventory is empty (entries is the real claim_inventory key).
+        inventory = {
+            "active_claims": [
+                {
+                    "kind": "issue",
+                    "number": 640,
+                    "issue_number": 640,
+                    "status": "active",
+                },
+            ],
+            "counts": {},
+        }
+        snap = load_traffic_snapshot(
+            fetch_queue_snapshot=lambda: self._queue(issues=[issue]),
+            fetch_lease_snapshot=lambda: self._lease(claim_inventory=inventory),
+        )
+        items = [
+            i
+            for i in (
+                list(snap.runnable)
+                + list(snap.leased)
+                + list(snap.blocked)
+                + list(snap.needs_controller)
+            )
+            if i.kind == "issue" and i.number == 640
+        ]
+        self.assertEqual(len(items), 1)
+        self.assertIsNone(
+            items[0].lease_info,
+            "active_claims is not a real inventory key; entries-only",
+        )
+
+
+class TestTrafficRoutesAndRendering(unittest.TestCase):
+    def setUp(self):
+        self.client = TestClient(create_app())
+
+    def test_traffic_html_page_rendering(self):
+        cand1 = WorkCandidate(
+            kind="issue",
+            number=640,
+            state="open",
+            labels=("status:ready",),
+            title="Traffic View Implementation",
+            priority=20,
+        )
+        cand2 = WorkCandidate(
+            kind="issue",
+            number=643,
+            state="open",
+            labels=("status:ready",),
+            title="Dependent Feature",
+            priority=20,
+            dependency_unmet=True,
+            dependency_reason="issue#643 depends on unresolved issue(s) #640; they are not closed",
+        )
+        snap = load_traffic_snapshot(candidates=[cand1, cand2])
+        with mock.patch("webui.app.load_traffic_snapshot", return_value=snap):
+            response = self.client.get("/traffic")
+        
+        self.assertEqual(response.status_code, 200)
+        self.assertIn("Workflow Traffic Control", response.text)
+        self.assertIn("1. Runnable Lanes", response.text)
+        self.assertIn("3. Blocked Items", response.text)
+        self.assertIn("Traffic View Implementation", response.text)
+        self.assertIn("depends on unresolved issue(s) #640", response.text)
+
+    def test_api_traffic_json_route(self):
+        cand = WorkCandidate(
+            kind="issue",
+            number=640,
+            state="open",
+            labels=("status:ready",),
+            title="Traffic View API Test",
+            priority=20,
+        )
+        snap = load_traffic_snapshot(candidates=[cand])
+        with mock.patch("webui.app.load_traffic_snapshot", return_value=snap):
+            response = self.client.get("/api/traffic")
+        
+        self.assertEqual(response.status_code, 200)
+        data = response.json()
+        self.assertEqual(data["project_id"], "gitea-tools")
+        self.assertEqual(len(data["runnable"]), 1)
+        self.assertEqual(data["runnable"][0]["number"], 640)
+
+    def test_render_traffic_fail_closed_page(self):
+        snap = TrafficSnapshot(
+            project_id="gitea-tools",
+            repo_label="Scaled-Tech-Consulting/Gitea-Tools",
+            runnable=(),
+            leased=(),
+            blocked=(),
+            needs_controller=(),
+            terminal_complete=(),
+            next_roles=(),
+            fetch_error="Gitea credentials unavailable for gitea.prgs.cc",
+            inventory_complete=False,
+        )
+        html = render_traffic_page(snap)
+        self.assertIn("Traffic data unavailable", html)
+        self.assertIn("Fail closed", html)
+        self.assertNotIn("1. Runnable Lanes", html)
+
+
+if __name__ == "__main__":
+    unittest.main()

webui/app.py

@@ -42,10 +42,17 @@ from webui.lease_loader import load_lease_snapshot, snapshot_to_dict as lease_sn
 from webui.lease_views import render_leases_page
 from webui.queue_loader import load_queue_snapshot, snapshot_to_dict as queue_snapshot_to_dict
 from webui.queue_views import render_queue_page
+from webui.traffic_loader import load_traffic_snapshot, snapshot_to_dict as traffic_snapshot_to_dict
+from webui.traffic_views import render_traffic_page
 from webui.worktree_scanner import load_hygiene_snapshot, snapshot_to_dict as worktree_snapshot_to_dict
 from webui.worktree_views import render_worktrees_page
 from webui.runtime_health import load_runtime_snapshot, snapshot_to_dict as runtime_snapshot_to_dict
 from webui.runtime_views import render_runtime_page
+from webui.session_loader import (
+    load_session_view_snapshot,
+    snapshot_to_dict as session_view_snapshot_to_dict,
+)
+from webui.session_views import render_sessions_page
 from webui.inventory import (
     SECTION_NAMES as _INVENTORY_SECTIONS,
     load_inventory_snapshot,
@@ -80,10 +87,12 @@ def _stub_page(title: str, description: str) -> HTMLResponse:
 
 
 _LEGACY_PAGES = (
+    ("/traffic", "Traffic", "workflow traffic-control view (#640)"),
     ("/queue", "Queue", "live PR and issue dashboard (#429)"),
     ("/projects", "Projects", "registry and onboarding (#427)"),
     ("/prompts", "Prompts", "canonical workflow prompt library (#428)"),
     ("/runtime", "Runtime", "MCP health and stale-runtime detection (#430)"),
+    ("/sessions", "Sessions", "runtime and session view (#641)"),
     ("/audit", "Audit", "final-report paste and validator preview (#431)"),
     ("/worktrees", "Worktrees", "branch hygiene dashboard (#432)"),
     ("/leases", "Leases", "collision and lease visibility (#433)"),
@@ -200,6 +209,15 @@ async def api_queue(_request: Request) -> JSONResponse:
     return JSONResponse(queue_snapshot_to_dict(load_queue_snapshot()))
 
 
+async def traffic(_request: Request) -> HTMLResponse:
+    snapshot = load_traffic_snapshot()
+    return HTMLResponse(render_traffic_page(snapshot))
+
+
+async def api_traffic(_request: Request) -> JSONResponse:
+    return JSONResponse(traffic_snapshot_to_dict(load_traffic_snapshot()))
+
+
 def _load_project_registry() -> tuple[ProjectRegistry | None, RegistryError | None]:
     """Load the registry, converting validation failure into a fail-closed pair."""
     try:
@@ -313,6 +331,17 @@ async def api_runtime(_request: Request) -> JSONResponse:
     return JSONResponse(runtime_snapshot_to_dict(load_runtime_snapshot()))
 
 
+async def sessions(_request: Request) -> HTMLResponse:
+    """Runtime and session view (#641) — read-only composition of health + inventory."""
+    snapshot = load_session_view_snapshot()
+    return HTMLResponse(render_sessions_page(snapshot))
+
+
+async def api_sessions(_request: Request) -> JSONResponse:
+    """JSON export for the runtime/session view (#641)."""
+    return JSONResponse(session_view_snapshot_to_dict(load_session_view_snapshot()))
+
+
 async def _parse_audit_form(request: Request) -> tuple[str, str | None]:
     if request.method == "GET":
         return "", None
@@ -736,6 +765,8 @@ def create_app(*, bind_host: str | None = None) -> Starlette:
             Route("/system-health", system_health, methods=["GET"]),
             Route("/queue", queue, methods=["GET"]),
             Route("/api/queue", api_queue, methods=["GET"]),
+            Route("/traffic", traffic, methods=["GET"]),
+            Route("/api/traffic", api_traffic, methods=["GET"]),
             Route("/projects", projects, methods=["GET"]),
             Route("/projects/{project_id}", project_detail, methods=["GET"]),
             Route("/api/projects", api_projects, methods=["GET"]),
@@ -750,6 +781,9 @@ def create_app(*, bind_host: str | None = None) -> Starlette:
             Route("/api/prompts", api_prompts, methods=["GET"]),
             Route("/runtime", runtime, methods=["GET"]),
             Route("/api/runtime", api_runtime, methods=["GET"]),
+            Route("/sessions", sessions, methods=["GET"]),
+            Route("/api/sessions", api_sessions, methods=["GET"]),
+            Route("/api/v1/sessions", api_sessions, methods=["GET"]),
             Route("/api/v1/timeline", api_v1_timeline, methods=["GET"]),
             Route("/analytics", analytics, methods=["GET"]),
             Route("/api/analytics", api_v1_analytics, methods=["GET"]),

webui/inventory.py

@@ -76,6 +76,35 @@ _CREDENTIAL_KEY_RE = re.compile(
 )
 _REDACTED = "[redacted]"
 
+#: Credential-shaped *name* as it appears inside a free-form command line. This
+#: is deliberately broader than :data:`_CREDENTIAL_KEY_RE` — it also matches a
+#: bare ``key`` component, so ``PRIVATE_KEY=`` is caught. Over-redacting a
+#: displayed string is safe; under-redacting one is not.
+_TEXT_CREDENTIAL_NAME = (
+    r"[A-Za-z0-9_.\-]*"
+    r"(?:token|secret|password|passwd|key|authorization|bearer|credential)"
+    r"[A-Za-z0-9_.\-]*"
+)
+#: A value following such a name: single-quoted, double-quoted, or bare. The
+#: bare form stops at a quote so an enclosing quote survives the redaction.
+_TEXT_CREDENTIAL_VALUE = r"'[^']*'|\"[^\"]*\"|[^\s'\"]+"
+
+_TEXT_CREDENTIAL_FLAG_RE = re.compile(
+    rf"(?P<key>(?<![\w\-])--?{_TEXT_CREDENTIAL_NAME})"
+    rf"(?P<sep>[=\s]+)"
+    rf"(?P<value>{_TEXT_CREDENTIAL_VALUE})",
+    re.IGNORECASE,
+)
+_TEXT_CREDENTIAL_ASSIGN_RE = re.compile(
+    rf"(?P<key>(?<![\w\-]){_TEXT_CREDENTIAL_NAME})"
+    rf"(?P<sep>\s*[:=]\s*)"
+    rf"(?P<value>{_TEXT_CREDENTIAL_VALUE})",
+    re.IGNORECASE,
+)
+_TEXT_URL_USERINFO_RE = re.compile(
+    r"(?P<scheme>\b[A-Za-z][A-Za-z0-9+.\-]*://)[^\s/@]+@"
+)
+
 
 @dataclass(frozen=True)
 class InventorySection:
@@ -225,6 +254,42 @@ def scrub(value: Any, *, key: str | None = None) -> Any:
     return repr(value)
 
 
+def collapse_home(text: str) -> str:
+    """Collapse every ``$HOME`` occurrence *inside* a string, not just a prefix."""
+    home = os.path.expanduser("~")
+    if not home or home == "/":
+        return text
+    return text.replace(home, "~")
+
+
+def scrub_text(value: Any) -> Any:
+    """Redact a free-form text blob such as a recorded command line.
+
+    :func:`scrub` keys off structured field *names* and whole-value prefixes,
+    which is right for inventory records but blind to a secret embedded in the
+    middle of a sentence. This collapses ``$HOME`` and redacts credential-shaped
+    tokens and URL userinfo *anywhere* in the string, so operator-supplied text
+    rendered verbatim — contamination ``command_summary`` (#630) above all — is
+    held to the same standard as every other field on the page.
+
+    Returns ``None`` unchanged so callers can keep "absent" distinct from "".
+    """
+    if value is None:
+        return None
+    text = value if isinstance(value, str) else str(value)
+    text = collapse_home(text)
+    text = _TEXT_URL_USERINFO_RE.sub(
+        lambda m: f"{m.group('scheme')}{_REDACTED}@", text
+    )
+    text = _TEXT_CREDENTIAL_FLAG_RE.sub(
+        lambda m: f"{m.group('key')}{m.group('sep')}{_REDACTED}", text
+    )
+    text = _TEXT_CREDENTIAL_ASSIGN_RE.sub(
+        lambda m: f"{m.group('key')}{m.group('sep')}{_REDACTED}", text
+    )
+    return text
+
+
 # ── control-plane database (read-only) ───────────────────────────────────────
 
 

webui/lease_loader.py

@@ -182,10 +182,17 @@ def _extract_reviewer_leases(
             parsed = parse_reviewer_lease_comment(comment.get("body") or "")
             if not parsed:
                 continue
+            subject_pr = parsed.get("pr_number") or pr_number
             leases.append(
                 {
                     **parsed,
-                    "pr_number": parsed.get("pr_number") or pr_number,
+                    "pr_number": subject_pr,
+                    # The lease subject is the PR, never the linked issue: a
+                    # reviewer lease on PR #N must not be attributed to issue #N
+                    # or to the issue that PR closes (#640).
+                    "kind": "pr",
+                    "number": subject_pr,
+                    "role": "reviewer",
                     "comment_id": comment.get("id"),
                     "author": (comment.get("user") or {}).get("login"),
                     "created_at": comment.get("created_at"),

webui/nav.py

@@ -41,13 +41,14 @@ NAV_GROUPS: tuple[NavGroup, ...] = (
         NavItem("/system-health", "System health"),
     )),
     NavGroup("Traffic", (
+        NavItem("/traffic", "Traffic control"),
         NavItem("/queue", "Queue"),
         NavItem("/leases", "Leases"),
         NavItem("/actions", "Actions"),
     )),
     NavGroup("Runtime/Sessions", (
         NavItem("/runtime", "Runtime health"),
-        NavItem("/sessions", "Sessions", "stub"),
+        NavItem("/sessions", "Sessions"),
     )),
     NavGroup("Projects", (
         NavItem("/projects", "Projects"),
@@ -75,11 +76,6 @@ NAV_GROUPS: tuple[NavGroup, ...] = (
 # issues of epic #631. Each maps a path to (title, description). Routes are
 # registered so nav links resolve to a graceful, read-only stub page.
 STUB_PAGES: dict[str, tuple[str, str]] = {
-    "/sessions": (
-        "Sessions",
-        "Active session, capability, and role inventory. Backed by the unified "
-        "inventory API (#636) once it lands.",
-    ),
     "/inventory": (
         "Inventory",
         "Unified sessions, leases, locks, namespaces, and worktree inventory. "

webui/queue_loader.py

@@ -4,7 +4,7 @@ from __future__ import annotations
 
 import os
 import re
-from dataclasses import dataclass
+from dataclasses import dataclass, field
 from datetime import datetime, timezone
 from typing import Any, Callable
 from urllib.parse import urlparse
@@ -31,10 +31,20 @@ class PaginationMeta:
 
 @dataclass(frozen=True)
 class QueueItem:
+    """One queue row.
+
+    ``extra`` holds *display* strings for the queue page (values are truncated
+    or humanized for rendering). ``signals`` holds the *authoritative* typed
+    values taken straight from the Gitea payload, for consumers that classify
+    or pin state rather than render it (#640). Never derive identity or
+    concurrency decisions from ``extra``.
+    """
+
     number: int
     title: str
     badges: tuple[str, ...]
     extra: dict[str, str]
+    signals: dict[str, Any] = field(default_factory=dict)
 
 
 @dataclass(frozen=True)
@@ -134,21 +144,37 @@ def _format_pr_item(pr: dict, badges: tuple[str, ...]) -> QueueItem:
         "mergeable" if mergeable is True else "conflicted" if mergeable is False else "unknown"
     )
     linked = _extract_linked_issue(pr.get("title"), pr.get("body"))
+    head_sha = str(head.get("sha") or "")
+    labels = tuple(
+        str(lb.get("name") or "") for lb in (pr.get("labels") or []) if lb.get("name")
+    )
     return QueueItem(
         number=int(pr["number"]),
         title=str(pr.get("title") or ""),
         badges=badges,
         extra={
             "branch": f"{head.get('ref', '?')} → {base.get('ref', '?')}",
-            "head_sha": str(head.get("sha") or "")[:12],
+            # Display only — truncated. Pin against signals["head_sha"] instead.
+            "head_sha": head_sha[:12],
             "mergeable": merge_label,
             "linked_issue": str(linked) if linked is not None else "",
         },
+        signals={
+            "head_sha": head_sha,
+            "head_ref": str(head.get("ref") or ""),
+            "base_ref": str(base.get("ref") or ""),
+            "mergeable": mergeable if isinstance(mergeable, bool) else None,
+            "labels": labels,
+            "linked_issue": linked,
+        },
     )
 
 
 def _format_issue_item(issue: dict, badges: tuple[str, ...]) -> QueueItem:
-    labels = ", ".join(lb.get("name", "") for lb in issue.get("labels", []))
+    label_names = tuple(
+        str(lb.get("name") or "") for lb in (issue.get("labels") or []) if lb.get("name")
+    )
+    labels = ", ".join(label_names)
     assignee = (issue.get("assignee") or {}).get("login", "")
     return QueueItem(
         number=int(issue["number"]),
@@ -159,6 +185,11 @@ def _format_issue_item(issue: dict, badges: tuple[str, ...]) -> QueueItem:
             "assignee": assignee or "unassigned",
             "state": str(issue.get("state") or ""),
         },
+        signals={
+            "labels": label_names,
+            "assignee": assignee,
+            "state": str(issue.get("state") or ""),
+        },
     )
 
 

webui/runtime_views.py

@@ -88,5 +88,7 @@ def render_runtime_page(snapshot: RuntimeSnapshot) -> str:
         "<p class='muted'>MVP is read-only — restart MCP servers from your IDE/operator "
         "workflow. Related issue: <code>#420</code>. Guidance: "
         f"<code>{html.escape(snapshot.restart_guidance)}</code></p>"
-        "<p class='muted'>This page does not expose tokens or perform MCP restarts.</p>"
-    )
+        "<p class='muted'>This page does not expose tokens or perform MCP restarts. "
+        "Correlated sessions, worktree bindings, and contamination markers: "
+        "<a href='/sessions'>/sessions</a> (#641).</p>"
+    )

webui/sanctioned_restart.py

@@ -38,6 +38,7 @@ from dataclasses import asdict, dataclass
 from typing import Any
 
 import mcp_namespace_health
+import restart_coordinator
 import runtime_recovery_guard
 from webui import console_audit, console_authz
 
@@ -99,6 +100,14 @@ def _clean(value: Any) -> str:
     return str(value or "").strip()
 
 
+def restart_class_for_mode(mode: str) -> str:
+    """Map the existing namespace controls onto the #663 class taxonomy."""
+
+    if _clean(mode) == MODE_RELOAD:
+        return restart_coordinator.RestartClass.CONFIGURATION_RELOAD.value
+    return restart_coordinator.RestartClass.ROLE_RUNTIME_RESTART.value
+
+
 # --- Mutation ledger --------------------------------------------------------
 
 
@@ -256,6 +265,7 @@ def build_restart_preview(
 
     return {
         "action_id": action_id,
+        "restart_class": restart_class_for_mode(md),
         "namespace": ns,
         "mode": md,
         "scope_valid": scope_error is None,
@@ -309,6 +319,7 @@ def assess_restart_request(
             "reason_code": reason_code,
             "detail": detail,
             "action_id": action_id,
+            "restart_class": restart_class_for_mode(md),
             "namespace": ns,
             "mode": md,
             "preview": preview,
@@ -393,6 +404,7 @@ def assess_restart_request(
             "process."
         ),
         "action_id": action_id,
+        "restart_class": restart_class_for_mode(md),
         "namespace": ns,
         "mode": md,
         "preview": preview,
@@ -441,7 +453,11 @@ def execute_restart(
             else console_audit.RESULT_DENIED
         ),
         principal=principal,
-        target={"namespace": assessment["namespace"], "mode": assessment["mode"]},
+        target={
+            "namespace": assessment["namespace"],
+            "mode": assessment["mode"],
+            "restart_class": assessment["restart_class"],
+        },
         reason_code=assessment["reason_code"],
         detail=assessment["detail"],
         request_id=request_id,
@@ -450,6 +466,7 @@ def execute_restart(
             "gates_passed": assessment["gates_passed"],
             "process_kill_executed": False,
             "post_restart_verification_required": True,
+            "restart_class": assessment["restart_class"],
         },
     )
 
@@ -463,6 +480,7 @@ def execute_restart(
         "namespace": assessment["namespace"],
         "mode": assessment["mode"],
         "action_id": action_id,
+        "restart_class": assessment["restart_class"],
         "process_kill_executed": False,
         "host_hook": assessment["preview"]["restart_hook"],
         "next_action": (

webui/session_loader.py

@@ -0,0 +1,517 @@
+"""Compose runtime health + inventory into a sessions/runtime view (#641).
+
+Phase 1 is read-only. It correlates namespaces, sessions, capabilities,
+worktree bindings, lease ownership, stale flags, and contamination markers
+when they are detectable on disk (#630 / #671). It never restarts, kills, or
+takes over a session.
+
+Sources:
+
+* :mod:`webui.runtime_health` — profile, role, stale runtime, shell health.
+* :mod:`webui.inventory` — sessions, leases, locks, worktrees, namespaces
+  and collision signals from the control-plane DB + filesystem.
+* :mod:`mcp_session_state` — durable contamination markers (inspect only).
+
+Secrets are never read. Inventory-sourced values arrive already redacted by
+:func:`webui.inventory.scrub`; free-form marker text this module loads itself is
+put through :func:`webui.inventory.scrub_text`, which collapses ``$HOME`` and
+redacts credential-shaped tokens *inside* a string rather than only at its start.
+
+Ownership columns are authority-aware. A lease or lock section that could not be
+read renders as ``unknown``, never as ``none`` or ``unbound``: a lease the reader
+could not load is not an absent lease (see
+:attr:`webui.inventory.InventorySnapshot.ownership_authority_complete`).
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any, Callable
+
+import mcp_session_state
+from webui.inventory import (
+    OWNERSHIP_SECTIONS,
+    STATUS_OK,
+    InventorySection,
+    InventorySnapshot,
+    load_inventory_snapshot,
+    scrub_text,
+    snapshot_to_dict as inventory_snapshot_to_dict,
+)
+from webui.runtime_health import (
+    RuntimeSnapshot,
+    load_runtime_snapshot,
+    snapshot_to_dict as runtime_snapshot_to_dict,
+)
+
+# Sanctioned recovery pointers only — never pkill / killall (#630).
+SANCTIONED_RECOVERY_DOCS: tuple[dict[str, str], ...] = (
+    {
+        "label": "MCP namespace EOF recovery (reconnect only)",
+        "path": "docs/mcp-namespace-eof-recovery.md",
+        "note": "IDE/client reconnect or operator-owned restart; never kill daemons.",
+    },
+    {
+        "label": "MCP namespace health",
+        "path": "docs/mcp-namespace-health.md",
+        "note": "client_namespace probe proves namespace health.",
+    },
+    {
+        "label": "Restart path inventory",
+        "path": "docs/mcp-restart-path-inventory.md",
+        "note": "Catalog of sanctioned reconnect/restart paths.",
+    },
+    {
+        "label": "Local web UI recovery",
+        "path": "docs/webui-local-dev.md",
+        "note": "Operator console start and documented recovery sequence.",
+    },
+)
+
+_CONTAMINATION_KINDS: tuple[str, ...] = (
+    mcp_session_state.KIND_RUNTIME_RECOVERY_CONTAMINATION,
+    mcp_session_state.KIND_STABLE_BRANCH_CONTAMINATION,
+)
+
+#: Status recorded on a row when the backing inventory section is absent
+#: entirely — distinct from a section that reported itself degraded.
+AUTHORITY_MISSING = "missing"
+
+
+def _section_status(section: InventorySection | None) -> str:
+    """Status of an ownership section, treating an absent section as missing."""
+    if section is None:
+        return AUTHORITY_MISSING
+    return section.status
+
+
+def _combined_authority(*statuses: str) -> str:
+    """Worst status of the sections a derived column depends on.
+
+    A column proved from two sections is only trustworthy when *both* read
+    cleanly, so the first non-``ok`` status wins.
+    """
+    for status in statuses:
+        if status != STATUS_OK:
+            return status
+    return STATUS_OK
+
+
+@dataclass(frozen=True)
+class ContaminationMarker:
+    """A detectable durable contamination marker (audit-safe summary)."""
+
+    kind: str
+    on_disk: bool
+    has_payload: bool
+    summary: str
+    reason_class: str | None = None
+    session_id: str | None = None
+    role: str | None = None
+    command_summary: str | None = None
+    cleared: bool = False
+
+    def to_dict(self) -> dict[str, Any]:
+        return {
+            "kind": self.kind,
+            "on_disk": self.on_disk,
+            "has_payload": self.has_payload,
+            "summary": self.summary,
+            "reason_class": self.reason_class,
+            "session_id": self.session_id,
+            "role": self.role,
+            "command_summary": self.command_summary,
+            "cleared": self.cleared,
+            "active": self.on_disk and self.has_payload and not self.cleared,
+        }
+
+
+@dataclass(frozen=True)
+class SessionRow:
+    """One correlated session row for the sessions table."""
+
+    session_id: str
+    role: str | None
+    profile: str | None
+    namespace: str | None
+    pid: int | None
+    pid_alive: bool | None
+    status: str | None
+    started_at: str | None
+    last_heartbeat_at: str | None
+    lease_ids: tuple[str, ...] = ()
+    work_refs: tuple[str, ...] = ()
+    worktree_paths: tuple[str, ...] = ()
+    stale_flags: tuple[str, ...] = ()
+    contamination_flags: tuple[str, ...] = ()
+    #: Status of the section backing ``lease_ids``/``work_refs``. While this is
+    #: not ``ok`` those tuples mean "could not be read", never "none held".
+    lease_authority: str = STATUS_OK
+    #: Worst status across the sections backing ``worktree_paths`` (locks are
+    #: correlated through lease work numbers, so both must read cleanly).
+    worktree_authority: str = STATUS_OK
+
+    @property
+    def ownership_authority_complete(self) -> bool:
+        """True only when this row's ownership columns are provable."""
+        return self.lease_authority == STATUS_OK and self.worktree_authority == STATUS_OK
+
+    def to_dict(self) -> dict[str, Any]:
+        return {
+            "session_id": self.session_id,
+            "role": self.role,
+            "profile": self.profile,
+            "namespace": self.namespace,
+            "pid": self.pid,
+            "pid_alive": self.pid_alive,
+            "status": self.status,
+            "started_at": self.started_at,
+            "last_heartbeat_at": self.last_heartbeat_at,
+            "lease_ids": list(self.lease_ids),
+            "work_refs": list(self.work_refs),
+            "worktree_paths": list(self.worktree_paths),
+            "stale_flags": list(self.stale_flags),
+            "contamination_flags": list(self.contamination_flags),
+            "is_stale": bool(self.stale_flags),
+            "is_contaminated": bool(self.contamination_flags),
+            "lease_authority": self.lease_authority,
+            "worktree_authority": self.worktree_authority,
+            "lease_authority_complete": self.lease_authority == STATUS_OK,
+            "worktree_authority_complete": self.worktree_authority == STATUS_OK,
+            "ownership_authority_complete": self.ownership_authority_complete,
+            "ownership_note": (
+                "Lease and worktree columns are proved from sections that read "
+                "cleanly."
+                if self.ownership_authority_complete
+                else "An ownership source could not be read; empty lease_ids or "
+                "worktree_paths on this row mean unknown, not unowned."
+            ),
+        }
+
+
+@dataclass(frozen=True)
+class SessionViewSnapshot:
+    """Composed runtime + session inventory view (#641)."""
+
+    runtime: RuntimeSnapshot
+    inventory: InventorySnapshot
+    sessions: tuple[SessionRow, ...]
+    contamination_markers: tuple[ContaminationMarker, ...]
+    recovery_docs: tuple[dict[str, str], ...] = SANCTIONED_RECOVERY_DOCS
+    fetch_error: str | None = None
+
+    @property
+    def stale_session_count(self) -> int:
+        return sum(1 for row in self.sessions if row.stale_flags)
+
+    @property
+    def contaminated_session_count(self) -> int:
+        return sum(1 for row in self.sessions if row.contamination_flags)
+
+    @property
+    def active_contamination(self) -> tuple[ContaminationMarker, ...]:
+        return tuple(m for m in self.contamination_markers if m.to_dict()["active"])
+
+    @property
+    def ownership_authority_complete(self) -> bool:
+        """False while any ownership section is degraded, absent, or unavailable."""
+        return self.inventory.ownership_authority_complete
+
+    @property
+    def ownership_section_status(self) -> dict[str, str]:
+        """Per-section status for the three ownership-bearing sections."""
+        return {
+            name: _section_status(self.inventory.section(name))
+            for name in OWNERSHIP_SECTIONS
+        }
+
+
+def _inspect_contamination(
+    kind: str,
+    *,
+    remote: str | None,
+    inspect: Callable[..., dict[str, Any]] | None = None,
+    load: Callable[..., dict[str, Any] | None] | None = None,
+) -> ContaminationMarker:
+    """Inspect one contamination kind; never raises into the page render path."""
+    inspect_fn = inspect or mcp_session_state.inspect_state_envelope
+    load_fn = load or mcp_session_state.load_state
+    try:
+        envelope = inspect_fn(kind=kind, remote=remote)
+    except Exception as exc:  # noqa: BLE001 — fail soft for the dashboard
+        return ContaminationMarker(
+            kind=kind,
+            on_disk=False,
+            has_payload=False,
+            summary=f"contamination inspect failed: {type(exc).__name__}",
+        )
+
+    reason_class = None
+    session_id = None
+    role = None
+    command_summary = None
+    cleared = False
+    summary = scrub_text(str(envelope.get("summary") or ""))
+
+    if envelope.get("on_disk") and envelope.get("has_payload"):
+        try:
+            payload = load_fn(kind=kind, remote=remote) or {}
+        except Exception:  # noqa: BLE001
+            payload = {}
+        if isinstance(payload, dict):
+            reason_class = payload.get("reason_class")
+            session_id = payload.get("session_id")
+            role = payload.get("role")
+            command_summary = payload.get("command_summary") or payload.get("detail")
+            cleared = bool(payload.get("cleared_by_reconciler"))
+            if not summary:
+                summary = scrub_text(
+                    f"{kind}: {reason_class or 'present'}"
+                    + (" (cleared)" if cleared else "")
+                )
+
+    # Marker payloads are operator-supplied free text and are the one thing on
+    # this page that does not arrive through inventory scrubbing. The write-time
+    # redactor is a narrow denylist, so redact again at the display boundary:
+    # it leaves $HOME paths, `-H 'X-Api-Key: …'`, `--password …`, and
+    # `PRIVATE_KEY=…` intact. The field itself stays — it is #630 evidence.
+    return ContaminationMarker(
+        kind=kind,
+        on_disk=bool(envelope.get("on_disk")),
+        has_payload=bool(envelope.get("has_payload")),
+        summary=summary or f"{kind}: not present",
+        reason_class=scrub_text(str(reason_class)) if reason_class else None,
+        session_id=scrub_text(str(session_id)) if session_id else None,
+        role=scrub_text(str(role)) if role else None,
+        command_summary=scrub_text(str(command_summary)) if command_summary else None,
+        cleared=cleared,
+    )
+
+
+def _load_contamination_markers(
+    *,
+    remote: str | None,
+    inspect: Callable[..., dict[str, Any]] | None = None,
+    load: Callable[..., dict[str, Any] | None] | None = None,
+) -> tuple[ContaminationMarker, ...]:
+    return tuple(
+        _inspect_contamination(kind, remote=remote, inspect=inspect, load=load)
+        for kind in _CONTAMINATION_KINDS
+    )
+
+
+def _build_session_rows(
+    inventory: InventorySnapshot,
+    contamination: tuple[ContaminationMarker, ...],
+) -> tuple[SessionRow, ...]:
+    sessions_section = inventory.section("sessions")
+    leases_section = inventory.section("leases")
+    locks_section = inventory.section("locks")
+
+    # Ownership columns may only assert absence when their source read cleanly.
+    lease_authority = _section_status(leases_section)
+    # Locks carry claimant profile/username, not control-plane session ids, so a
+    # worktree binding is correlated through lease work numbers: it depends on
+    # the locks *and* the leases section.
+    worktree_authority = _combined_authority(
+        lease_authority, _section_status(locks_section)
+    )
+
+    leases_by_session: dict[str, list[dict[str, Any]]] = {}
+    for lease in (leases_section.items if leases_section else ()):
+        sid = str(lease.get("session_id") or "")
+        if sid:
+            leases_by_session.setdefault(sid, []).append(lease)
+
+    active_markers = [m for m in contamination if m.to_dict()["active"]]
+    marker_session_ids = {
+        m.session_id for m in active_markers if m.session_id
+    }
+
+    rows: list[SessionRow] = []
+    for raw in sessions_section.items if sessions_section else ():
+        sid = str(raw.get("session_id") or "")
+        if not sid:
+            continue
+        session_leases = leases_by_session.get(sid, [])
+        lease_ids = tuple(
+            str(lease["lease_id"])
+            for lease in session_leases
+            if lease.get("lease_id")
+        )
+        work_refs: list[str] = []
+        work_numbers: list[int] = []
+        for lease in session_leases:
+            kind = lease.get("work_kind")
+            number = lease.get("work_number")
+            if kind and number is not None:
+                work_refs.append(f"{kind}#{number}")
+                try:
+                    work_numbers.append(int(number))
+                except (TypeError, ValueError):
+                    pass
+
+        worktree_paths: list[str] = []
+        for lock in (locks_section.items if locks_section else ()):
+            try:
+                issue_no = int(lock.get("issue_number"))
+            except (TypeError, ValueError):
+                continue
+            if issue_no in work_numbers and lock.get("worktree_path"):
+                worktree_paths.append(str(lock["worktree_path"]))
+
+        stale_flags: list[str] = []
+        if raw.get("pid_alive") is False:
+            stale_flags.append("pid-dead")
+        status = str(raw.get("status") or "").lower()
+        if status and status not in {"active", "alive", "running", "ok"}:
+            stale_flags.append(f"status:{status}")
+        for lease in session_leases:
+            if lease.get("expired") is True:
+                stale_flags.append("lease-expired")
+            if str(lease.get("status") or "").lower() == "active" and lease.get(
+                "expired"
+            ) is True:
+                stale_flags.append("active-lease-past-expiry")
+
+        contamination_flags: list[str] = []
+        if sid in marker_session_ids:
+            for marker in active_markers:
+                if marker.session_id == sid:
+                    contamination_flags.append(marker.kind)
+        # Process-wide contamination with no session binding still surfaces
+        # against every live session so it cannot be silent (#630).
+        for marker in active_markers:
+            if not marker.session_id and marker.kind not in contamination_flags:
+                contamination_flags.append(f"{marker.kind}:process-wide")
+
+        rows.append(
+            SessionRow(
+                session_id=sid,
+                role=raw.get("role"),
+                profile=raw.get("profile"),
+                namespace=raw.get("namespace"),
+                pid=raw.get("pid") if isinstance(raw.get("pid"), int) else None,
+                pid_alive=raw.get("pid_alive")
+                if isinstance(raw.get("pid_alive"), bool)
+                else None,
+                status=raw.get("status"),
+                started_at=raw.get("started_at"),
+                last_heartbeat_at=raw.get("last_heartbeat_at"),
+                lease_ids=lease_ids,
+                work_refs=tuple(work_refs),
+                worktree_paths=tuple(worktree_paths),
+                stale_flags=tuple(dict.fromkeys(stale_flags)),
+                contamination_flags=tuple(dict.fromkeys(contamination_flags)),
+                lease_authority=lease_authority,
+                worktree_authority=worktree_authority,
+            )
+        )
+
+    return tuple(rows)
+
+
+def load_session_view_snapshot(
+    *,
+    load_runtime: Callable[..., RuntimeSnapshot] | None = None,
+    load_inventory: Callable[..., InventorySnapshot] | None = None,
+    inspect_contamination: Callable[..., dict[str, Any]] | None = None,
+    load_contamination_payload: Callable[..., dict[str, Any] | None] | None = None,
+) -> SessionViewSnapshot:
+    """Build the composed sessions/runtime view. Fail-soft on partial sources."""
+    runtime_loader = load_runtime or load_runtime_snapshot
+    inventory_loader = load_inventory or load_inventory_snapshot
+    fetch_error: str | None = None
+
+    try:
+        runtime = runtime_loader()
+    except Exception as exc:  # noqa: BLE001
+        fetch_error = f"runtime snapshot failed: {type(exc).__name__}: {exc}"
+        # Minimal placeholder so the page still renders inventory + recovery.
+        from webui.runtime_health import RuntimeSnapshot as _RS
+
+        runtime = _RS(
+            project_id="unknown",
+            repo_root="",
+            remote="",
+            host="",
+            profile_name="unknown",
+            role_kind="unknown",
+            config_model="unknown",
+            profile_mode="unknown",
+            profile_source="unknown",
+            authenticated_username=None,
+            identity_error=str(exc),
+            repo_sha=None,
+            remote_master_sha=None,
+            commits_behind_master=None,
+            stale_runtime_warning=None,
+            shell_health={},
+            workflow_hashes=(),
+            schema_hashes=(),
+            restart_guidance="docs/mcp-namespace-eof-recovery.md",
+            fetch_error=str(exc),
+        )
+
+    try:
+        inventory = inventory_loader()
+    except Exception as exc:  # noqa: BLE001
+        msg = f"inventory snapshot failed: {type(exc).__name__}: {exc}"
+        fetch_error = f"{fetch_error}; {msg}" if fetch_error else msg
+        inventory = load_inventory_snapshot(
+            db_path="/nonexistent-for-fail-soft",
+            lock_dir="/nonexistent-for-fail-soft",
+        )
+
+    remote = getattr(runtime, "remote", None)
+    contamination = _load_contamination_markers(
+        remote=remote,
+        inspect=inspect_contamination,
+        load=load_contamination_payload,
+    )
+    sessions = _build_session_rows(inventory, contamination)
+    return SessionViewSnapshot(
+        runtime=runtime,
+        inventory=inventory,
+        sessions=sessions,
+        contamination_markers=contamination,
+        fetch_error=fetch_error,
+    )
+
+
+def snapshot_to_dict(snapshot: SessionViewSnapshot) -> dict[str, Any]:
+    """JSON export for ``/api/sessions`` (read-only)."""
+    return {
+        "api_version": "v1",
+        "view": "runtime-sessions",
+        "issue": 641,
+        "fetch_error": snapshot.fetch_error,
+        "runtime": runtime_snapshot_to_dict(snapshot.runtime),
+        "inventory": inventory_snapshot_to_dict(snapshot.inventory),
+        "sessions": [row.to_dict() for row in snapshot.sessions],
+        "session_counts": {
+            "total": len(snapshot.sessions),
+            "stale": snapshot.stale_session_count,
+            "contaminated": snapshot.contaminated_session_count,
+        },
+        "ownership_authority_complete": snapshot.ownership_authority_complete,
+        "ownership_section_status": snapshot.ownership_section_status,
+        "ownership_note": (
+            "Every ownership source read cleanly; a session with no lease and no "
+            "worktree path genuinely holds neither."
+            if snapshot.ownership_authority_complete
+            else "An ownership source is degraded or unavailable. Empty lease_ids "
+            "and worktree_paths mean unknown, not unowned; consult each row's "
+            "lease_authority and worktree_authority."
+        ),
+        "contamination_markers": [
+            marker.to_dict() for marker in snapshot.contamination_markers
+        ],
+        "active_contamination": [
+            marker.to_dict() for marker in snapshot.active_contamination
+        ],
+        "recovery_docs": [dict(doc) for doc in snapshot.recovery_docs],
+        "read_only": True,
+        "phase": 1,
+        "mutations": [],
+    }

webui/session_views.py

@@ -0,0 +1,403 @@
+"""HTML views for the Runtime and session view (Phase 1, #641).
+
+Read-only composition of runtime health (#430) and inventory sessions /
+namespaces / worktrees (#636). Surfaces stale and contamination indicators
+when detectable. Recovery links point only at sanctioned reconnect/restart
+docs — never at manual process kill (#630).
+"""
+
+from __future__ import annotations
+
+from html import escape
+from typing import Sequence
+
+from webui.inventory import STATUS_OK
+from webui.layout import render_page
+from webui.session_loader import (
+    ContaminationMarker,
+    SessionRow,
+    SessionViewSnapshot,
+)
+
+
+def _badge(text: str, css: str) -> str:
+    return f'<span class="badge {css}">{escape(text)}</span>'
+
+
+def _flags(flags: Sequence[str], *, css: str) -> str:
+    if not flags:
+        return '<span class="muted">—</span>'
+    return " ".join(_badge(flag, css) for flag in flags)
+
+
+def _unproven_cell(status: str) -> str:
+    """Render an ownership column whose backing inventory section failed to read.
+
+    Never "none" and never "unbound": an unreadable source proves nothing about
+    ownership, and claiming otherwise is the exact failure the
+    ``ownership_authority_complete`` invariant exists to prevent.
+    """
+    return (
+        f'<span class="muted">unknown (inventory {escape(status)})</span><br>'
+        f'{_badge("authority unproven", "badge-health-degraded")}'
+    )
+
+
+def _runtime_banner(snapshot: SessionViewSnapshot) -> str:
+    runtime = snapshot.runtime
+    stale = runtime.stale_runtime_warning
+    stale_html = ""
+    if stale:
+        stale_html = (
+            f'<div class="health-card health-stale" style="margin-top:0.75rem;">'
+            f"<strong>Stale runtime:</strong> {escape(stale)}</div>"
+        )
+    identity = runtime.authenticated_username or "unresolved"
+    if runtime.identity_error:
+        identity = f"unresolved ({runtime.identity_error})"
+
+    return f"""<div class="health-card">
+  <h3>Runtime context</h3>
+  <table class="detail">
+    <tr><th>Profile</th><td><code>{escape(runtime.profile_name)}</code></td></tr>
+    <tr><th>Role kind</th><td>{escape(runtime.role_kind)}</td></tr>
+    <tr><th>Identity</th><td>{escape(str(identity))}</td></tr>
+    <tr><th>Remote / host</th>
+      <td><code>{escape(runtime.remote)}</code> · <code>{escape(runtime.host)}</code></td>
+    </tr>
+    <tr><th>Local HEAD</th>
+      <td><code>{escape(runtime.repo_sha or "unknown")}</code></td>
+    </tr>
+    <tr><th>Remote master</th>
+      <td><code>{escape(runtime.remote_master_sha or "unknown")}</code></td>
+    </tr>
+    <tr><th>Commits behind</th>
+      <td>{escape(str(runtime.commits_behind_master if runtime.commits_behind_master is not None else "unknown"))}</td>
+    </tr>
+  </table>
+  <p class="muted">Full runtime detail: <a href="/runtime">/runtime</a> ·
+  Inventory API: <a href="/api/v1/inventory"><code>/api/v1/inventory</code></a></p>
+  {stale_html}
+</div>"""
+
+
+def _summary_bar(snapshot: SessionViewSnapshot) -> str:
+    total = len(snapshot.sessions)
+    stale = snapshot.stale_session_count
+    contaminated = snapshot.contaminated_session_count
+    active_markers = len(snapshot.active_contamination)
+    inv_status = snapshot.inventory.status
+    authority_complete = snapshot.ownership_authority_complete
+    authority_text = "complete" if authority_complete else "incomplete"
+    authority_css = "badge-health-ok" if authority_complete else "badge-health-degraded"
+    return f"""<div class="health-card" style="display:flex; flex-wrap:wrap; gap:1rem; align-items:center;">
+  <div><strong>Sessions:</strong> <span class="badge badge-health-ok">{total}</span></div>
+  <div><strong>Stale:</strong> <span class="badge badge-stale">{stale}</span></div>
+  <div><strong>Contaminated:</strong> <span class="badge badge-blocked">{contaminated}</span></div>
+  <div><strong>Active markers:</strong> <span class="badge badge-health-unproven">{active_markers}</span></div>
+  <div><strong>Inventory:</strong> <span class="badge badge-health-skipped">{escape(inv_status)}</span></div>
+  <div><strong>Ownership authority:</strong> <span class="badge {authority_css}">{authority_text}</span></div>
+</div>"""
+
+
+def _ownership_caveat(snapshot: SessionViewSnapshot) -> str:
+    """Name the unreadable ownership sections, or render nothing when all read."""
+    if snapshot.ownership_authority_complete:
+        return ""
+    degraded = ", ".join(
+        f"{name}: {status}"
+        for name, status in snapshot.ownership_section_status.items()
+        if status != STATUS_OK
+    )
+    return (
+        '<div class="health-card health-stale">'
+        "<strong>Ownership authority incomplete:</strong> "
+        f"{escape(degraded)}. Columns marked <em>unknown</em> could not be read. "
+        "No session below may be treated as holding no lease or no worktree "
+        "binding — absence of evidence is not evidence of absence.</div>"
+    )
+
+
+def _render_session_row(row: SessionRow) -> str:
+    pid = "—" if row.pid is None else str(row.pid)
+    pid_alive = "—" if row.pid_alive is None else ("alive" if row.pid_alive else "dead")
+    pid_css = (
+        "badge-health-ok"
+        if row.pid_alive is True
+        else ("badge-blocked" if row.pid_alive is False else "badge-health-skipped")
+    )
+    # An empty tuple only means "holds none" when its source read cleanly.
+    if row.lease_authority != STATUS_OK:
+        lease_cell = _unproven_cell(row.lease_authority)
+    else:
+        leases = (
+            ", ".join(f"<code>{escape(lid)}</code>" for lid in row.lease_ids)
+            if row.lease_ids
+            else '<span class="muted">none</span>'
+        )
+        work = (
+            ", ".join(escape(ref) for ref in row.work_refs)
+            if row.work_refs
+            else '<span class="muted">—</span>'
+        )
+        lease_cell = (
+            f'{leases}<div class="muted" style="font-size:0.82rem; '
+            f'margin-top:0.2rem;">{work}</div>'
+        )
+
+    if row.worktree_authority != STATUS_OK:
+        worktrees = _unproven_cell(row.worktree_authority)
+    else:
+        worktrees = (
+            "<br>".join(f"<code>{escape(path)}</code>" for path in row.worktree_paths)
+            if row.worktree_paths
+            else '<span class="muted">unbound</span>'
+        )
+    return f"""<tr>
+  <td><code>{escape(row.session_id)}</code></td>
+  <td>
+    <div><code>{escape(str(row.role or "—"))}</code> / <code>{escape(str(row.profile or "—"))}</code></div>
+    <div class="muted" style="font-size:0.82rem;">ns: <code>{escape(str(row.namespace or "—"))}</code></div>
+  </td>
+  <td>
+    <code>{escape(pid)}</code>
+    {_badge(pid_alive, pid_css)}
+  </td>
+  <td>{escape(str(row.status or "—"))}<div class="muted" style="font-size:0.82rem;">{escape(str(row.last_heartbeat_at or ""))}</div></td>
+  <td>{lease_cell}</td>
+  <td>{worktrees}</td>
+  <td>{_flags(row.stale_flags, css="badge-stale")}</td>
+  <td>{_flags(row.contamination_flags, css="badge-blocked")}</td>
+</tr>"""
+
+
+def _sessions_table(snapshot: SessionViewSnapshot) -> str:
+    rows: Sequence[SessionRow] = snapshot.sessions
+    if not rows:
+        sessions_status = snapshot.ownership_section_status.get("sessions", STATUS_OK)
+        if sessions_status != STATUS_OK:
+            return (
+                '<p class="muted">Session inventory is '
+                f"<strong>{escape(sessions_status)}</strong> — the session list "
+                "could not be read. This is not evidence that no sessions "
+                "exist.</p>"
+            )
+        return (
+            '<p class="muted">No control-plane sessions recorded. Inventory may '
+            "be unavailable, or no MCP workers have registered yet.</p>"
+        )
+    body = "".join(_render_session_row(row) for row in rows)
+    return f"""<table class="registry">
+  <thead>
+    <tr>
+      <th>Session</th>
+      <th>Role / profile / namespace</th>
+      <th>PID</th>
+      <th>Status</th>
+      <th>Leases / work</th>
+      <th>Worktree binding</th>
+      <th>Stale</th>
+      <th>Contamination</th>
+    </tr>
+  </thead>
+  <tbody>
+    {body}
+  </tbody>
+</table>"""
+
+
+def _namespaces_section(snapshot: SessionViewSnapshot) -> str:
+    section = snapshot.inventory.section("namespaces")
+    if section is None:
+        return (
+            '<div class="prompt-card"><h3>Namespaces</h3>'
+            '<p class="muted">Namespaces section not loaded.</p></div>'
+        )
+    if not section.ok:
+        return f"""<div class="prompt-card">
+  <h3>Namespaces {_badge(section.status, "badge-health-degraded")}</h3>
+  <p class="muted">{escape(section.reason or "unavailable")}</p>
+</div>"""
+
+    rows = []
+    for item in section.items:
+        caps = item.get("capability_summary") or {}
+        cap_bits = ", ".join(
+            name for name, ok in sorted(caps.items()) if ok
+        ) or "none"
+        rows.append(
+            "<tr>"
+            f"<td><code>{escape(str(item.get('mcp_namespace') or '—'))}</code></td>"
+            f"<td><code>{escape(str(item.get('profile_name') or '—'))}</code></td>"
+            f"<td>{escape(str(item.get('role') or '—'))}</td>"
+            f"<td>{escape(cap_bits)}</td>"
+            f"<td>{'yes' if item.get('active') else 'no'}</td>"
+            "</tr>"
+        )
+    reason = (
+        f'<p class="muted">{escape(section.reason)}</p>'
+        if section.reason
+        else ""
+    )
+    return f"""<div class="prompt-card">
+  <h3>Namespaces / capabilities</h3>
+  {reason}
+  <table class="registry">
+    <thead>
+      <tr>
+        <th>Namespace</th>
+        <th>Profile</th>
+        <th>Role</th>
+        <th>Capabilities</th>
+        <th>Active in process</th>
+      </tr>
+    </thead>
+    <tbody>
+      {"".join(rows) if rows else '<tr><td colspan="5" class="muted">No namespace rows.</td></tr>'}
+    </tbody>
+  </table>
+</div>"""
+
+
+def _worktrees_section(snapshot: SessionViewSnapshot) -> str:
+    section = snapshot.inventory.section("worktrees")
+    if section is None:
+        return ""
+    if not section.ok and not section.items:
+        return f"""<div class="prompt-card">
+  <h3>Worktrees {_badge(section.status, "badge-health-degraded")}</h3>
+  <p class="muted">{escape(section.reason or "unavailable")}</p>
+</div>"""
+
+    rows = []
+    for item in section.items[:50]:
+        rows.append(
+            "<tr>"
+            f"<td><code>{escape(str(item.get('rel_path') or item.get('path') or '—'))}</code></td>"
+            f"<td><code>{escape(str(item.get('branch') or '—'))}</code></td>"
+            f"<td>{escape(str(item.get('classification') or '—'))}</td>"
+            f"<td>{'yes' if item.get('registered_worktree') else 'no'}</td>"
+            f"<td>{'dirty' if item.get('dirty') else 'clean'}</td>"
+            "</tr>"
+        )
+    more = ""
+    if len(section.items) > 50:
+        more = f'<p class="muted">Showing 50 of {len(section.items)}. Full list: <a href="/worktrees">/worktrees</a>.</p>'
+    return f"""<div class="prompt-card">
+  <h3>Worktree bindings</h3>
+  <p class="muted">Registered issue worktrees under <code>branches/</code>. Hygiene detail: <a href="/worktrees">/worktrees</a>.</p>
+  <table class="registry">
+    <thead>
+      <tr>
+        <th>Path</th>
+        <th>Branch</th>
+        <th>Classification</th>
+        <th>Registered</th>
+        <th>State</th>
+      </tr>
+    </thead>
+    <tbody>
+      {"".join(rows) if rows else '<tr><td colspan="5" class="muted">No worktrees recorded.</td></tr>'}
+    </tbody>
+  </table>
+  {more}
+</div>"""
+
+
+def _contamination_section(markers: Sequence[ContaminationMarker]) -> str:
+    if not markers:
+        return (
+            '<div class="prompt-card"><h3>Contamination markers</h3>'
+            '<p class="muted">No contamination kinds inspected.</p></div>'
+        )
+    rows = []
+    for marker in markers:
+        active = marker.to_dict()["active"]
+        status = "ACTIVE" if active else ("cleared" if marker.cleared else "absent")
+        css = "badge-blocked" if active else "badge-health-ok"
+        rows.append(
+            "<tr>"
+            f"<td><code>{escape(marker.kind)}</code></td>"
+            f"<td>{_badge(status, css)}</td>"
+            f"<td>{escape(marker.reason_class or '—')}</td>"
+            f"<td><code>{escape(marker.session_id or '—')}</code></td>"
+            f"<td>{escape(marker.command_summary or marker.summary)}</td>"
+            "</tr>"
+        )
+    return f"""<div class="prompt-card">
+  <h3>Contamination markers (#630 / #671)</h3>
+  <p class="muted">Durable markers only — never silent when present. Clearance is reconciler-only.</p>
+  <table class="registry">
+    <thead>
+      <tr>
+        <th>Kind</th>
+        <th>State</th>
+        <th>Reason class</th>
+        <th>Session</th>
+        <th>Summary</th>
+      </tr>
+    </thead>
+    <tbody>
+      {"".join(rows)}
+    </tbody>
+  </table>
+</div>"""
+
+
+def _recovery_section(snapshot: SessionViewSnapshot) -> str:
+    items = []
+    for doc in snapshot.recovery_docs:
+        items.append(
+            "<li>"
+            f"<code>{escape(doc['path'])}</code> — "
+            f"<strong>{escape(doc['label'])}</strong>: {escape(doc['note'])}"
+            "</li>"
+        )
+    return f"""<div class="prompt-card">
+  <h3>Sanctioned recovery (read-only)</h3>
+  <p class="muted">This view does <strong>not</strong> restart, kill, or take over sessions.
+  Manual <code>pkill</code> / <code>kill</code> of MCP daemons is contamination (#630), not recovery.</p>
+  <ul class="reasons">
+    {"".join(items)}
+    <li>Prefer IDE/client reconnect (<code>/mcp reconnect</code>) or an operator-owned restart recorded in the restart inventory.</li>
+  </ul>
+</div>"""
+
+
+def render_sessions_page(snapshot: SessionViewSnapshot) -> str:
+    """Render the full HTML body for the runtime/session view."""
+    error_block = ""
+    if snapshot.fetch_error:
+        error_block = (
+            f'<div class="health-card health-stale"><strong>Partial load:</strong> '
+            f"{escape(snapshot.fetch_error)}</div>"
+        )
+    if snapshot.runtime.fetch_error:
+        error_block += (
+            f'<div class="health-card health-stale"><strong>Runtime note:</strong> '
+            f"{escape(snapshot.runtime.fetch_error)}</div>"
+        )
+
+    body = f"""
+{error_block}
+{_runtime_banner(snapshot)}
+{_summary_bar(snapshot)}
+
+<div class="prompt-card">
+  <h3>Sessions</h3>
+  <p class="muted">Control-plane sessions correlated with leases and worktree bindings.
+  Stale and contamination flags are fail-soft: absence of a marker is not proof of cleanliness when inventory is degraded.
+  Lease and worktree columns read <em>unknown (inventory …)</em> when their source could not be loaded.</p>
+  {_ownership_caveat(snapshot)}
+  {_sessions_table(snapshot)}
+</div>
+
+{_namespaces_section(snapshot)}
+{_worktrees_section(snapshot)}
+{_contamination_section(snapshot.contamination_markers)}
+{_recovery_section(snapshot)}
+"""
+    return render_page(title="Sessions", body_html=f"""<h2>Runtime and sessions</h2>
+<p class="meta">Phase 1 read-only view (#641). Combines runtime health (#430) with
+unified inventory sessions/namespaces/worktrees (#636). No restart or session-takeover controls.</p>
+{body}""")

webui/system_health_views.py

@@ -278,8 +278,10 @@ def _recovery_card() -> str:
         "controls arrive in Phase 2 (#642); until then recovery runs through "
         "the sanctioned client reconnect / operator restart path.</p>"
         "<ul class='reasons'>"
-        "<li><a href='/runtime'>Runtime and session view</a> — active profile, "
-        "workflow hashes, and shell health.</li>"
+        "<li><a href='/runtime'>Runtime health</a> — active profile, workflow "
+        "hashes, and shell health.</li>"
+        "<li><a href='/sessions'>Runtime and sessions</a> — namespaces, session "
+        "rows, worktree bindings, and contamination markers (#641).</li>"
         "<li>Reconnect the MCP client from the IDE, then re-run the blocked "
         "cycle. Never kill the daemon process manually: unmanaged kills are "
         "recorded as runtime contamination (#630).</li>"

webui/traffic_loader.py

@@ -0,0 +1,449 @@
+"""Traffic-control view loader for Phase 1 operator web console (#640).
+
+Combines queue snapshots, inventory leases, dependency graph classifications,
+and workflow dashboard rules to deliver full traffic-control visibility:
+runnable, leased (in-progress), blocked (dependency/lock), needs-controller,
+and terminal-complete candidates.
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any, Callable, Sequence
+
+from webui.project_registry import find_project, load_registry
+from webui.queue_loader import load_queue_snapshot, QueueSnapshot
+from webui.lease_loader import load_lease_snapshot, LeaseSnapshot
+from workflow_dashboard import (
+    DashboardSnapshot,
+    QueueEntry,
+    RoleNextAction,
+    build_workflow_dashboard,
+    DASHBOARD_ROLES,
+)
+from allocator_service import WorkCandidate
+
+
+@dataclass(frozen=True)
+class TrafficItem:
+    kind: str  # "issue" or "pr"
+    number: int
+    title: str
+    traffic_state: str  # "runnable", "leased", "blocked", "needs_controller", "terminal_complete"
+    expected_role: str
+    safe_for_roles: tuple[str, ...]
+    badges: tuple[str, ...]
+    block_reason: str | None = None
+    lease_info: dict[str, Any] | None = None
+    head_sha: str | None = None
+
+    @property
+    def is_safe(self) -> bool:
+        return self.block_reason is None and bool(self.safe_for_roles)
+
+    def as_dict(self) -> dict[str, Any]:
+        return {
+            "kind": self.kind,
+            "number": self.number,
+            "title": self.title,
+            "traffic_state": self.traffic_state,
+            "expected_role": self.expected_role,
+            "safe_for_roles": list(self.safe_for_roles),
+            "badges": list(self.badges),
+            "block_reason": self.block_reason,
+            "lease_info": self.lease_info,
+            "head_sha": self.head_sha,
+            "is_safe": self.is_safe,
+        }
+
+
+@dataclass(frozen=True)
+class TrafficSnapshot:
+    project_id: str
+    repo_label: str
+    runnable: tuple[TrafficItem, ...]
+    leased: tuple[TrafficItem, ...]
+    blocked: tuple[TrafficItem, ...]
+    needs_controller: tuple[TrafficItem, ...]
+    terminal_complete: tuple[TrafficItem, ...]
+    next_roles: tuple[dict[str, Any], ...]
+    fetch_error: str | None = None
+    inventory_complete: bool = True
+
+    def as_dict(self) -> dict[str, Any]:
+        return {
+            "project_id": self.project_id,
+            "repo_label": self.repo_label,
+            "runnable": [i.as_dict() for i in self.runnable],
+            "leased": [i.as_dict() for i in self.leased],
+            "blocked": [i.as_dict() for i in self.blocked],
+            "needs_controller": [i.as_dict() for i in self.needs_controller],
+            "terminal_complete": [i.as_dict() for i in self.terminal_complete],
+            "next_roles": list(self.next_roles),
+            "fetch_error": self.fetch_error,
+            "inventory_complete": self.inventory_complete,
+        }
+
+
+def _classify_traffic_item(
+    entry: QueueEntry,
+    *,
+    lease_info: dict[str, Any] | None = None,
+) -> TrafficItem:
+    """Classify a QueueEntry into a TrafficItem with explicit traffic state."""
+    badges = list(entry.badges)
+    block_reason = entry.block_reason
+    expected_role = entry.expected_role
+
+    entry_is_safe = entry.block_reason is None and bool(entry.safe_for_roles)
+    # Lease state is checked first: an item that is both leased and blocked is
+    # reported as leased. That is safe by construction — a leased item is never
+    # placed in the runnable lane — and it keeps the operator's attention on the
+    # session that currently owns the work. The blocker text still renders.
+    if lease_info is not None or "in-progress" in badges or "claimed" in badges:
+        state = "leased"
+    elif expected_role == "reconciler" or "terminal-lock" in badges:
+        state = "terminal_complete"
+    elif expected_role == "controller" or "contaminated" in badges or "needs-controller" in badges:
+        state = "needs_controller"
+    elif (
+        block_reason is not None
+        or "blocked" in badges
+        or "dependency-unmet" in badges
+        or "blocked-by-terminal" in badges
+        or "status:blocked" in badges
+    ):
+        state = "blocked"
+    elif entry_is_safe:
+        state = "runnable"
+    else:
+        state = "needs_controller"
+
+    return TrafficItem(
+        kind=entry.kind,
+        number=entry.number,
+        title=entry.title,
+        traffic_state=state,
+        expected_role=expected_role,
+        safe_for_roles=entry.safe_for_roles,
+        badges=tuple(badges),
+        block_reason=block_reason,
+        lease_info=lease_info,
+        head_sha=entry.head_sha,
+    )
+
+
+# Claim statuses from ``issue_claim_heartbeat.build_claim_inventory`` that mean
+# a live worker currently holds the issue. Everything else (``stale``,
+# ``phantom``, ``reclaimable``, ``not_claimed``) is reported through the
+# dashboard's stale-lease channel and is never rendered as an active lease.
+_ACTIVE_CLAIM_STATUSES = frozenset({"active", "awaiting_review"})
+
+# Statuses that positively mean "not an active lease" for any lease record.
+_INACTIVE_LEASE_STATUSES = frozenset(
+    {"expired", "stale", "released", "moot", "reclaimable", "phantom", "not_claimed"}
+)
+
+
+def _candidates_from_queue_snapshot(q_snap: QueueSnapshot) -> list[WorkCandidate]:
+    """Build allocator candidates from the queue loader's authoritative signals.
+
+    Display badges (``blocked``/``claimed``/``duplicate``/``stale``/
+    ``in-review``/``open``) are rendering hints, not routing state, so nothing
+    here branches on them. Every routing field comes from
+    ``QueueItem.signals`` — the raw Gitea payload values.
+
+    The queue loader reads ``/pulls`` and ``/issues`` only; it never fetches
+    review verdicts. ``request_changes_current_head`` / ``approval_on_current_head``
+    are therefore left at their fail-safe ``False`` rather than being guessed
+    from badges: an unproven approval must never route a PR to the merger.
+    """
+    candidates: list[WorkCandidate] = []
+
+    for pr in q_snap.prs:
+        signals = pr.signals or {}
+        head_sha = str(signals.get("head_sha") or "").strip()
+        mergeable = signals.get("mergeable")
+        labels = tuple(str(x) for x in (signals.get("labels") or ()))
+        candidates.append(
+            WorkCandidate(
+                kind="pr",
+                number=pr.number,
+                state="open",
+                labels=labels,
+                title=pr.title,
+                # Full 40-char SHA from head.sha — never the 12-char display value.
+                head_sha=head_sha or None,
+                priority=5,
+                mergeable=mergeable is True,
+                blocked=mergeable is False or "status:blocked" in labels,
+            )
+        )
+
+    for issue in q_snap.issues:
+        signals = issue.signals or {}
+        labels = tuple(str(x) for x in (signals.get("labels") or ()))
+        lowered = {label.lower() for label in labels}
+        candidates.append(
+            WorkCandidate(
+                kind="issue",
+                number=issue.number,
+                state="open",
+                labels=labels,
+                title=issue.title,
+                priority=20 if "status:ready" in lowered else 10,
+                blocked="status:blocked" in lowered,
+                # A live claim by another session is not this session's work.
+                already_claimed_elsewhere="status:in-progress" in lowered,
+            )
+        )
+
+    return candidates
+
+
+def _claim_lease_records(inventory: dict[str, Any] | None) -> list[dict[str, Any]]:
+    """Normalize ``build_claim_inventory`` entries into lease records.
+
+    The inventory contract is ``{"entries", "counts", "heartbeat_lease_minutes",
+    "reclaim_after_minutes", "in_progress_total"}``. Each entry is keyed by
+    ``issue_number``; the subject kind is therefore always ``issue``.
+    """
+    entries = (inventory or {}).get("entries") or ()
+    records: list[dict[str, Any]] = []
+    for entry in entries:
+        if not isinstance(entry, dict):
+            continue
+        number = entry.get("issue_number")
+        if number is None:
+            continue
+        try:
+            number_int = int(number)
+        except (TypeError, ValueError):
+            continue
+        heartbeat = entry.get("latest_heartbeat") or {}
+        record = dict(entry)
+        record.update(
+            {
+                "kind": "issue",
+                "number": number_int,
+                "role": "author",
+                "lease_source": "issue-claim-heartbeat",
+            }
+        )
+        if isinstance(heartbeat, dict):
+            if heartbeat.get("session_id") and not record.get("session_id"):
+                record["session_id"] = heartbeat.get("session_id")
+            if heartbeat.get("author") and not record.get("author"):
+                record["author"] = heartbeat.get("author")
+        records.append(record)
+    return records
+
+
+def _lease_subject(lease: dict[str, Any]) -> tuple[str, int] | None:
+    """Return the ``(kind, number)`` a lease record actually covers.
+
+    Fails closed: a record that does not identify exactly one subject is
+    dropped rather than attributed to a guessed work item (#640 — never invent
+    a lease, and never attach a PR lease to a same-numbered issue).
+    """
+    kind = str(lease.get("kind") or lease.get("work_kind") or "").strip().lower()
+    pr_number = lease.get("pr_number")
+    issue_number = lease.get("issue_number")
+
+    if kind not in ("pr", "issue"):
+        if pr_number is not None and issue_number is None:
+            kind = "pr"
+        elif issue_number is not None and pr_number is None:
+            kind = "issue"
+        else:
+            return None
+
+    number = lease.get("number")
+    if number is None:
+        number = lease.get("work_number")
+    if number is None:
+        number = pr_number if kind == "pr" else issue_number
+    if number is None:
+        return None
+    try:
+        return kind, int(number)
+    except (TypeError, ValueError):
+        return None
+
+
+def _is_active_lease(lease: dict[str, Any]) -> bool:
+    """True when the record proves a worker currently holds the item."""
+    if lease.get("stale") or lease.get("expired"):
+        return False
+    status = str(lease.get("status") or lease.get("lease_status") or "").strip().lower()
+    if status in _INACTIVE_LEASE_STATUSES:
+        return False
+    if lease.get("lease_source") == "issue-claim-heartbeat":
+        return status in _ACTIVE_CLAIM_STATUSES
+    return True
+
+
+def load_traffic_snapshot(
+    *,
+    candidates: Sequence[WorkCandidate] | None = None,
+    leases: Sequence[dict[str, Any]] | None = None,
+    terminal_pr: int | None = None,
+    fetch_queue_snapshot: Callable[[], QueueSnapshot] | None = None,
+    fetch_lease_snapshot: Callable[[], LeaseSnapshot] | None = None,
+    project_id: str = "gitea-tools",
+) -> TrafficSnapshot:
+    """Load and compute the traffic-control snapshot."""
+    try:
+        reg = load_registry()
+        proj = find_project(reg, project_id)
+        repo_label = proj.remote_repo if proj else "Scaled-Tech-Consulting/Gitea-Tools"
+    except Exception:
+        repo_label = "Scaled-Tech-Consulting/Gitea-Tools"
+
+    # Injected candidates path (pure unit testing)
+    if candidates is not None:
+        dashboard = build_workflow_dashboard(
+            candidates=candidates,
+            leases=leases,
+            terminal_pr=terminal_pr,
+            inventory_complete=True,
+        )
+        return _build_traffic_snapshot_from_dashboard(
+            project_id=project_id,
+            repo_label=repo_label,
+            dashboard=dashboard,
+            leases=leases or (),
+        )
+
+    # Live snapshot loading
+    q_loader = fetch_queue_snapshot or load_queue_snapshot
+    l_loader = fetch_lease_snapshot or load_lease_snapshot
+
+    try:
+        q_snap = q_loader()
+        l_snap = l_loader()
+    except Exception as exc:  # noqa: BLE001
+        return TrafficSnapshot(
+            project_id=project_id,
+            repo_label=repo_label,
+            runnable=(),
+            leased=(),
+            blocked=(),
+            needs_controller=(),
+            terminal_complete=(),
+            next_roles=(),
+            fetch_error=f"Failed to load traffic state: {exc}",
+            inventory_complete=False,
+        )
+
+    if q_snap.fetch_error or l_snap.fetch_error:
+        err = q_snap.fetch_error or l_snap.fetch_error
+        return TrafficSnapshot(
+            project_id=project_id,
+            repo_label=repo_label,
+            runnable=(),
+            leased=(),
+            blocked=(),
+            needs_controller=(),
+            terminal_complete=(),
+            next_roles=(),
+            fetch_error=err,
+            inventory_complete=False,
+        )
+
+    candidate_list = _candidates_from_queue_snapshot(q_snap)
+
+    raw_leases: list[dict[str, Any]] = _claim_lease_records(l_snap.claim_inventory)
+    for r_lease in l_snap.reviewer_leases or ():
+        if not isinstance(r_lease, dict):
+            continue
+        # Always pin reviewer leases to the PR subject, even if a linked
+        # issue_number is present on the marker (#640 B2).
+        normalized = dict(r_lease)
+        subject = normalized.get("pr_number") or normalized.get("number")
+        if subject is None:
+            continue
+        try:
+            pr_num = int(subject)
+        except (TypeError, ValueError):
+            continue
+        normalized["kind"] = "pr"
+        normalized["number"] = pr_num
+        normalized["pr_number"] = pr_num
+        normalized.setdefault("role", "reviewer")
+        raw_leases.append(normalized)
+
+    dashboard = build_workflow_dashboard(
+        candidates=candidate_list,
+        leases=raw_leases,
+        inventory_complete=q_snap.pr_pagination.inventory_complete if q_snap.pr_pagination else True,
+    )
+
+    return _build_traffic_snapshot_from_dashboard(
+        project_id=project_id,
+        repo_label=repo_label,
+        dashboard=dashboard,
+        leases=raw_leases,
+    )
+
+
+def _build_traffic_snapshot_from_dashboard(
+    *,
+    project_id: str,
+    repo_label: str,
+    dashboard: DashboardSnapshot,
+    leases: Sequence[dict[str, Any]],
+) -> TrafficSnapshot:
+    """Classify dashboard entries into the 5 traffic state buckets."""
+    all_entries = dashboard.open_prs + dashboard.open_issues
+
+    # Map each active lease onto the exact work item it covers. Records whose
+    # subject cannot be determined, and claims that are stale/phantom/
+    # reclaimable, are deliberately dropped instead of guessed.
+    lease_map: dict[tuple[str, int], dict[str, Any]] = {}
+    for lease in leases:
+        if not isinstance(lease, dict) or not _is_active_lease(lease):
+            continue
+        subject = _lease_subject(lease)
+        if subject is not None:
+            lease_map[subject] = lease
+
+    runnable: list[TrafficItem] = []
+    leased: list[TrafficItem] = []
+    blocked: list[TrafficItem] = []
+    needs_controller: list[TrafficItem] = []
+    terminal_complete: list[TrafficItem] = []
+
+    for entry in all_entries:
+        l_info = lease_map.get((entry.kind, entry.number))
+        item = _classify_traffic_item(entry, lease_info=l_info)
+
+        if item.traffic_state == "leased":
+            leased.append(item)
+        elif item.traffic_state == "terminal_complete":
+            terminal_complete.append(item)
+        elif item.traffic_state == "blocked":
+            blocked.append(item)
+        elif item.traffic_state == "needs_controller":
+            needs_controller.append(item)
+        else:
+            runnable.append(item)
+
+    next_roles = [dashboard.next_safe_by_role[r].as_dict() for r in DASHBOARD_ROLES if r in dashboard.next_safe_by_role]
+
+    return TrafficSnapshot(
+        project_id=project_id,
+        repo_label=repo_label,
+        runnable=tuple(runnable),
+        leased=tuple(leased),
+        blocked=tuple(blocked),
+        needs_controller=tuple(needs_controller),
+        terminal_complete=tuple(terminal_complete),
+        next_roles=tuple(next_roles),
+        fetch_error=None,
+        inventory_complete=dashboard.inventory_complete,
+    )
+
+
+def snapshot_to_dict(snapshot: TrafficSnapshot) -> dict[str, Any]:
+    return snapshot.as_dict()

webui/traffic_views.py

@@ -0,0 +1,170 @@
+"""HTML rendering for Phase 1 Traffic-Control View (#640)."""
+
+from __future__ import annotations
+
+from html import escape
+from typing import Sequence
+
+from webui.layout import render_page
+from webui.traffic_loader import TrafficItem, TrafficSnapshot
+
+
+def _render_badges(badges: Sequence[str]) -> str:
+    if not badges:
+        return ""
+    out = []
+    for b in badges:
+        cls = "badge"
+        b_lower = b.lower()
+        if "blocked" in b_lower or "unmet" in b_lower:
+            cls += " badge-blocked"
+        elif "claimed" in b_lower or "in-progress" in b_lower or "leased" in b_lower:
+            cls += " badge-claimed"
+        elif "review" in b_lower or "ready" in b_lower:
+            cls += " badge-in-review"
+        elif "duplicate" in b_lower:
+            cls += " badge-duplicate"
+        elif "stale" in b_lower:
+            cls += " badge-stale"
+        out.append(f'<span class="{cls}">{escape(b)}</span>')
+    return f'<div class="badges">{"".join(out)}</div>'
+
+
+def _render_traffic_item_row(item: TrafficItem) -> str:
+    kind_label = escape(item.kind.upper())
+    num_str = f"#{item.number}"
+    title_str = escape(item.title)
+    role_str = escape(item.expected_role)
+    badges_html = _render_badges(item.badges)
+
+    reason_html = ""
+    if item.block_reason:
+        reason_html = f'<div class="muted" style="font-size:0.82rem; margin-top:0.2rem;"><strong>Blocker:</strong> {escape(item.block_reason)}</div>'
+
+    lease_html = ""
+    if item.lease_info:
+        owner = escape(str(item.lease_info.get("session_id") or item.lease_info.get("reviewer_identity") or "active worker"))
+        lease_html = f'<div class="muted" style="font-size:0.82rem; margin-top:0.2rem;"><strong>Lease:</strong> {owner}</div>'
+
+    return f"""<tr>
+  <td><code>{kind_label} {num_str}</code></td>
+  <td>
+    <div><strong>{title_str}</strong> {badges_html}</div>
+    {reason_html}
+    {lease_html}
+  </td>
+  <td><code>{role_str}</code></td>
+</tr>"""
+
+
+def _render_traffic_table(items: Sequence[TrafficItem], empty_message: str) -> str:
+    if not items:
+        return f'<p class="muted">{escape(empty_message)}</p>'
+
+    rows = "".join(_render_traffic_item_row(item) for item in items)
+    return f"""<table class="registry">
+  <thead>
+    <tr>
+      <th style="width: 15%;">Item</th>
+      <th style="width: 65%;">Title & Details</th>
+      <th style="width: 20%;">Next Role</th>
+    </tr>
+  </thead>
+  <tbody>
+    {rows}
+  </tbody>
+</table>"""
+
+
+def _render_next_roles(next_roles: Sequence[dict]) -> str:
+    if not next_roles:
+        return ""
+
+    cards = []
+    for r in next_roles:
+        role = escape(r.get("role", "unknown"))
+        status = r.get("status", "idle")
+        prompt = escape(r.get("prompt", ""))
+
+        status_cls = "badge-health-ok" if status == "safe" else ("badge-blocked" if "blocked" in status else "badge-health-skipped")
+        cards.append(f"""<div class="health-card" style="margin-bottom:0.75rem;">
+  <div style="display:flex; justify-content:space-between; align-items:center;">
+    <h3>Role: <code>{role}</code></h3>
+    <span class="badge {status_cls}">status: {escape(status)}</span>
+  </div>
+  <p class="meta" style="margin:0.35rem 0 0;">{prompt}</p>
+</div>""")
+
+    return f"""<div style="margin: 1.5rem 0;">
+  <h3>Next Safe Role Actions</h3>
+  {"".join(cards)}
+</div>"""
+
+
+def render_traffic_page(snapshot: TrafficSnapshot) -> str:
+    """Render the full HTML view for workflow traffic control."""
+    if snapshot.fetch_error:
+        body = f"""<h2>Workflow Traffic Control</h2>
+<p class="meta">Repository: <code>{escape(snapshot.repo_label)}</code></p>
+<div class="health-card health-stale">
+  <h3>Traffic data unavailable</h3>
+  <p class="health-headline">{escape(snapshot.fetch_error)}</p>
+  <p class="muted">Fail closed: traffic state cannot be established cleanly. Check credentials or remote connectivity.</p>
+</div>"""
+        return render_page(title="Traffic Control", body_html=body)
+
+    runnable_count = len(snapshot.runnable)
+    leased_count = len(snapshot.leased)
+    blocked_count = len(snapshot.blocked)
+    controller_count = len(snapshot.needs_controller)
+    terminal_count = len(snapshot.terminal_complete)
+
+    summary_bar = f"""<div class="health-card" style="display:flex; flex-wrap:wrap; gap:1rem; align-items:center;">
+  <div><strong>Runnable:</strong> <span class="badge badge-health-ok">{runnable_count}</span></div>
+  <div><strong>Leased:</strong> <span class="badge badge-claimed">{leased_count}</span></div>
+  <div><strong>Blocked:</strong> <span class="badge badge-blocked">{blocked_count}</span></div>
+  <div><strong>Needs Controller:</strong> <span class="badge badge-duplicate">{controller_count}</span></div>
+  <div><strong>Terminal Complete:</strong> <span class="badge badge-stale">{terminal_count}</span></div>
+</div>"""
+
+    next_roles_html = _render_next_roles(snapshot.next_roles)
+
+    sections_html = f"""
+<div class="prompt-card">
+  <h3>1. Runnable Lanes (Ready for Allocation)</h3>
+  <p class="muted">Safe work items with no unmet dependencies or active leases. Safe for allocation.</p>
+  {_render_traffic_table(snapshot.runnable, "No runnable items ready for allocation.")}
+</div>
+
+<div class="prompt-card">
+  <h3>2. In-Progress Work (Active Leases)</h3>
+  <p class="muted">Work items currently leased and actively being worked by an assigned role session.</p>
+  {_render_traffic_table(snapshot.leased, "No active leases in flight.")}
+</div>
+
+<div class="prompt-card">
+  <h3>3. Blocked Items (Dependencies / Locks)</h3>
+  <p class="muted">Items blocked by unmet dependency issues, a missing head pin, a merge conflict, or an active terminal review lock. Items labelled status:blocked route to section 4. Never presented as safe.</p>
+  {_render_traffic_table(snapshot.blocked, "No blocked items.")}
+</div>
+
+<div class="prompt-card">
+  <h3>4. Needs Controller Intervention</h3>
+  <p class="muted">Items requiring controller routing, diagnosis, or cross-role assignment.</p>
+  {_render_traffic_table(snapshot.needs_controller, "No items requiring controller intervention.")}
+</div>
+
+<div class="prompt-card">
+  <h3>5. Terminal / Complete Candidates</h3>
+  <p class="muted">Items ready for terminal reconciliation or post-merge worktree cleanup.</p>
+  {_render_traffic_table(snapshot.terminal_complete, "No terminal complete candidates.")}
+</div>
+"""
+
+    body = f"""<h2>Workflow Traffic Control</h2>
+<p class="meta">Repository: <code>{escape(snapshot.repo_label)}</code></p>
+{summary_bar}
+{next_roles_html}
+{sections_html}"""
+
+    return render_page(title="Traffic Control", body_html=body)