@open-agent-toolkit/cli 0.1.48 → 0.1.51

package/assets/skills/oat-project-implement/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: oat-project-implement
-version: 2.0.33
+version: 2.0.35
 description: Use when plan.md is ready for execution. Dispatches phase coordinators that select one exact target-pinned worker per task; supports bounded fix loops and plan-declared worktree-isolated parallel phases.
 oat_gateable: true
 argument-hint: '[--retry-limit <N>] [--dry-run]'
@@ -107,16 +107,21 @@ PROJECTS_ROOT="${PROJECTS_ROOT%/}"
 **Mandatory target-first order:** Complete Dispatch Policy Preflight and resolve
 the concrete managed Codex target before probing generic agent availability or
 locking Tier 1/Tier 2. A concrete target takes precedence over tier selection:
-use the exact registered role, or an explicitly pinned fresh Codex child when
-that role is unavailable. If neither route is possible, inline execution is
-allowed only with verified equivalent current-host model and effort controls;
-otherwise block before work starts. Explicit inherit/default and documented
-managed-uncapped reviewer behavior remain the only base-role exceptions.
-
-When a concrete managed Codex role is unavailable or the host cannot select
-it, launch a fresh Codex child pinned to the resolver-returned model and effort
-with canonical role instructions. If that fresh child cannot be launched, fail
-closed and block; never substitute the coordinator or base role.
+first send the exact registered role through native dispatch. Only a native
+role-selection rejection permits an explicitly pinned fresh Codex child. If
+neither exact route is possible, inline execution is allowed only with verified
+equivalent current-host model and effort controls; otherwise block before work
+starts. Explicit inherit/default and documented managed-uncapped reviewer
+behavior remain the only base-role exceptions.
+
+For a concrete managed Codex role, first send the resolver-returned Codex
+variant through the native spawn API as `agent_type`. Spawn acceptance plus the
+constructed launcher payload is configured invocation evidence; independent
+runtime telemetry or agent self-report is not required. Launch a fresh Codex
+child pinned to the resolver-returned model and effort with canonical role
+instructions only after a native role-selection rejection as defined below. If
+that fresh child cannot be launched, fail closed and block; never substitute
+the coordinator or base role.
 
 Use base `oat-phase-implementer` only for the allowed exceptions above:
 explicit inherit/default behavior. It is never a managed task-worker fallback.
@@ -142,9 +147,9 @@ Detection logic:
     - Declined → Tier 2.
 
 - If the host does not resolve either generic agent, first attempt the exact
-  registered role or explicitly pinned fresh-child route for any concrete
-  managed target. Select Tier 2 only when inline execution is allowed by the
-  target-first rule above.
+  registered role natively for any concrete managed target. Use the explicitly
+  pinned fresh-child route only after a native role-selection rejection. Select
+  Tier 2 only when inline execution is allowed by the target-first rule above.
 
 **Approval scope rule:** this Tier selection applies to both phase implementation and checkpoint review. Do not infer a mixed mode from conversational emphasis on review checkpoints. If the user has not explicitly approved Tier 1 for the run, stay Tier 2 throughout. Mixed mode is only valid when the user explicitly requests it.
 
@@ -181,6 +186,20 @@ If approved: Selected: Tier 1 — Subagents
 Forbidden: Selected: Tier 2 — Inline because the user did not separately mention subagents.
 ```
 
+**Native role-selection rejection:** This means the native host explicitly
+reports that the requested `agent_type` is unsupported, unknown, unregistered,
+or rejected before the child or agent starts. Missing runtime telemetry,
+missing agent self-report, a timeout after spawn acceptance, or any terminal
+result from an accepted child — including `BLOCKED` — is not role
+unavailability and is not a native role-selection rejection. Self-report is
+optional diagnostic data and cannot populate or overwrite launcher-owned
+`target`, `model_axis`, or `effort_axis` fields. An accepted child cannot
+trigger the fresh pinned-child or CLI fallback. If an accepted native reviewer
+later times out, retry the same already-selected native `agent_type` route. The
+fresh pinned-child route is eligible only when the original native attempt
+received explicit pre-start role-selection rejection; a timeout after native
+spawn acceptance never changes routes.
+
 **Legacy state migration:** If `state.md` contains `oat_execution_mode: subagent-driven`, silently ignore it. On the next bookkeeping write, remove that key. Do not redirect to `oat-project-subagent-implement` — that skill is deprecated.
 
 ### Dispatch Policy Preflight
@@ -193,13 +212,13 @@ Use the CLI resolver as the source of truth. The command name remains
 policy:
 
 ```bash
-oat project dispatch-ceiling resolve --provider <active-provider> --preflight --json
+oat project dispatch-ceiling resolve --provider <active-provider> --preflight --report-scope implementation-preflight --report-action implementation --json
 ```
 
 If `oat` is not in PATH, use:
 
 ```bash
-pnpm run cli -- project dispatch-ceiling resolve --provider <active-provider> --preflight --json
+pnpm run cli -- project dispatch-ceiling resolve --provider <active-provider> --preflight --report-scope implementation-preflight --report-action implementation --json
 ```
 
 Resolution order:
@@ -345,7 +364,7 @@ exists, rerun the resolver with non-interactive behavior and stop before work
 starts if it blocks:
 
 ```bash
-oat project dispatch-ceiling resolve --provider <active-provider> --preflight --non-interactive
+oat project dispatch-ceiling resolve --provider <active-provider> --preflight --non-interactive --report-scope implementation-preflight --report-action implementation
 ```
 
 ```text
@@ -390,6 +409,32 @@ an ordered matrix route:
   entry before retrying, up to the last available route entry and within
   `oat_orchestration_retry_limit`.
 
+#### Dispatch Report V1 contract
+
+Every implementation, fix, and review resolver invocation MUST pass explicit
+report context:
+
+- implementation: `--report-scope <phase-or-task> --report-action implementation`
+- fix: `--report-scope <phase-or-task> --report-action fix`
+- review: `--report-scope <phase-or-review-scope> --report-action review`
+
+Require `dispatchReport.schemaVersion: 1` in the completed resolver JSON before
+dispatch. Consume the report as the human/audit source: render the versioned
+block with `formatDispatchReport(dispatchReport)` semantics, and derive the
+formal compatibility line only through
+`formatDispatchStamp(dispatchReport)` / `toDispatchStampRecord(dispatchReport)`.
+Never hand-assemble a second `Dispatch:` schema from policy labels, role names,
+candidate strings, or target names.
+
+The exact provider invocation remains authoritative in
+`providers.<provider>.dispatchArgs` and `providers.<provider>.selection.target`;
+the report does not replace or weaken target-pinned dispatch. Add independently
+observed runtime identity to `dispatchReport.runtimeIdentity` only when such an
+observation exists. Requested/configured controls are not runtime observation.
+For gate-originated review, keep `dispatchReport.gateInvocation`, existing
+work-producer `diversity`, and reviewer `runtimeIdentity` as three distinct
+facts; producer stamps or self-report never overwrite configured invocation.
+
 Axis states:
 
 - `selected:<value>` - host exposes the axis and the orchestrator chose a value.
@@ -401,14 +446,19 @@ Axis states:
 Codex rules:
 
 **Managed Codex execution invariant:** When the resolver returns a model+effort
-target, prefer the exact registered role or variant from
-`providers.codex.dispatchArgs.variant`. If the current host cannot select that
-exact registered target, launch a fresh Codex child with the resolver target's
-explicit model, reasoning effort, and canonical role instructions from
+target, the resolver-returned Codex variant from
+`providers.codex.dispatchArgs.variant` must first be sent through the native
+spawn API as `agent_type`. Spawn acceptance plus the launcher payload is
+configured invocation evidence with launcher-selected/config-declared
+provenance. If and only if the host returns a native role-selection rejection,
+launch a fresh Codex child with the resolver target's explicit model, reasoning
+effort, and canonical role instructions from
 `.agents/agents/oat-phase-implementer.md` or
-`.agents/agents/oat-reviewer.md`. Workflow correctness must not require provider
-restart or hot reload. A managed base role is forbidden when a concrete target
-was requested; never silently downgrade to it. Base roles remain valid only for
+`.agents/agents/oat-reviewer.md`. Missing runtime telemetry or agent self-report
+is not role unavailability, and an accepted child result such as `BLOCKED`
+cannot trigger fallback. Workflow correctness must not require provider restart
+or hot reload. A managed base role is forbidden when a concrete target was
+requested; never silently downgrade to it. Base roles remain valid only for
 explicit inherit/default behavior and the documented managed-uncapped reviewer
 fallback.
 
@@ -422,12 +472,12 @@ fallback.
 3. For capped managed implementer/fix work, selected effort is `min(preferred, resolved_cap)`.
 4. For managed `Uncapped` implementer/fix work, selected effort is the preferred effort with no cap.
 5. For inherit/default mode, the resolver returns no selected dispatch args. Use the base/unpinned Codex role, log `Selected effort: provider-default`, display provider default effort when known, and do not describe this as managed uncapped behavior.
-6. For managed capped task-worker/fix dispatch, choose an exact configured candidate and call `oat project dispatch-ceiling resolve --provider codex --role implementer --ceiling-tier <project-or-phase-tier> --candidate-model <model> --candidate-effort <effort> --escalation-level <route-level> --json`. Read `providers.codex.dispatchArgs.variant` and `providers.codex.selection.target`; never reuse the coordinator role or a cap-only variant. `--preferred` remains compatibility behavior outside the exact task-worker path.
-7. For review dispatch: call `oat project dispatch-ceiling resolve --provider codex --role reviewer`; read `providers.codex.dispatchArgs.variant` and `providers.codex.selection.target`.
+6. For managed capped task-worker/fix dispatch, choose an exact configured candidate. For implementation, call `oat project dispatch-ceiling resolve --provider codex --role implementer --ceiling-tier <project-or-phase-tier> --candidate-model <model> --candidate-effort <effort> --escalation-level <route-level> --report-scope <task-id> --report-action implementation --json`. For a bounded fix, use `oat project dispatch-ceiling resolve --provider codex --role implementer --ceiling-tier <project-or-phase-tier> --candidate-model <model> --candidate-effort <effort> --escalation-level <route-level> --report-scope <task-id> --report-action fix --json`. Read `providers.codex.dispatchArgs.variant` and `providers.codex.selection.target`; never reuse the coordinator role or a cap-only variant. `--preferred` remains compatibility behavior outside the exact task-worker path.
+7. For review dispatch: call `oat project dispatch-ceiling resolve --provider codex --role reviewer --report-scope <phase-or-review-scope> --report-action review --json`; read `providers.codex.dispatchArgs.variant` and `providers.codex.selection.target`.
    - Capped managed policy: reviewer targets the configured cap for deterministic quality gate behavior.
    - Managed `Uncapped`: no reviewer target exists; use base/unpinned reviewer fallback and log `selectionMode=no-review-target`, `selectedValue=null`, and `effort_axis=provider-default`.
    - Inherit/default: no reviewer target exists; use base/unpinned reviewer fallback and log `selectionMode=inherit-default`, `selectedValue=null`, and `effort_axis=provider-default`.
-8. Codex payload-first assertion applies whenever the resolver returns a materialized model+effort target. If `providers.codex.dispatchArgs.variant` is present, the actual `spawn_agent` payload MUST use it as `agent_type`; when that variant came from a Codex model+effort target, log `model_axis=selected:<model>` and `effort_axis=selected:<effort>` from resolver output. If a concrete managed target has no usable variant, use the explicitly pinned fresh-child route or block. Use the base role and log provider-default only for explicit inherit/default behavior or the documented managed-uncapped reviewer exception. Always derive `model_axis` and `effort_axis` from resolver output, not from legacy role-name parsing.
+8. Codex payload-first assertion applies whenever the resolver returns a materialized model+effort target. If `providers.codex.dispatchArgs.variant` is present, the actual `spawn_agent` payload MUST first use it as native `agent_type`; when that variant came from a Codex model+effort target, log `model_axis=selected:<model>` and `effort_axis=selected:<effort>` from resolver output and the constructed launcher payload. Spawn acceptance is sufficient configured invocation evidence. Missing telemetry or self-report does not make the variant unusable. If native role selection explicitly rejects the variant, use the explicitly pinned fresh-child route or block. Use the base role and log provider-default only for explicit inherit/default behavior or the documented managed-uncapped reviewer exception. Always derive `model_axis` and `effort_axis` from resolver output, not from legacy role-name parsing or agent self-report.
 9. Do not use top-level per-call `reasoning_effort` as the standard OAT selected-effort path; dogfooding showed that path can be inconsistent.
 
 Claude rules:
@@ -440,7 +490,7 @@ Claude rules:
 - Review dispatch:
   - Capped managed policy: target the configured policy cap directly.
   - Managed `Uncapped` or inherit/default: no reviewer target exists; omit `model` and log inherited/default model behavior.
-- For managed capped task-worker/fix dispatch, call `oat project dispatch-ceiling resolve --provider claude --role implementer --ceiling-tier <project-or-phase-tier> --candidate-model <model> --orchestrator-tier <current-orchestrator-tier> --escalation-level <route-level> --json`; for review dispatch, call the resolver with `--role reviewer` and no candidate flags. Read `providers.claude.dispatchArgs.model` and pass it exactly on the actual Task invocation.
+- For managed capped task-worker/fix dispatch, call `oat project dispatch-ceiling resolve --provider claude --role implementer --ceiling-tier <project-or-phase-tier> --candidate-model <model> --orchestrator-tier <current-orchestrator-tier> --escalation-level <route-level> --report-scope <task-id> --report-action implementation --json` for implementation. For a bounded fix, call `oat project dispatch-ceiling resolve --provider claude --role implementer --ceiling-tier <project-or-phase-tier> --candidate-model <model> --orchestrator-tier <current-orchestrator-tier> --escalation-level <route-level> --report-scope <task-id> --report-action fix --json`. For review dispatch, call the resolver with `--role reviewer --report-scope <phase-or-review-scope> --report-action review --json` and no candidate flags. Read `providers.claude.dispatchArgs.model` and pass it exactly on the actual Task invocation.
 - Pass `model: "<value>"` when `model_axis=selected:<value>` on the Task tool call.
 - Keep `effort_axis=not-applicable`; Claude Code has no separate per-dispatch effort axis.
 
@@ -448,9 +498,10 @@ Cursor rules:
 
 - Treat every configured Cursor candidate string as opaque. Do not normalize it
   or infer capability from its spelling.
-- For managed capped task-worker/fix dispatch, call the resolver with
-  `--provider cursor --role implementer --ceiling-tier
-<project-or-phase-tier> --candidate-model <opaque-model> --json`.
+- For managed capped task-worker/fix dispatch, call
+  `oat project dispatch-ceiling resolve --provider cursor --role implementer --ceiling-tier <project-or-phase-tier> --candidate-model <opaque-model> --report-scope <task-id> --report-action implementation --json`
+  for implementation. For a bounded fix, call
+  `oat project dispatch-ceiling resolve --provider cursor --role implementer --ceiling-tier <project-or-phase-tier> --candidate-model <opaque-model> --report-scope <task-id> --report-action fix --json`.
 - Require `providers.cursor.dispatchArgs.model` and pass that exact byte-for-byte
   string as the actual Cursor invocation model. If the host cannot apply it,
   fail closed.
@@ -459,20 +510,23 @@ Payload-first invariant:
 
 - Build the actual host dispatch argument map before logging.
 - Do not emit `selected:<value>` unless the host invocation contains the corresponding role/model selection.
+- For every coordinator, task-worker, fix, and review launch, record `target`,
+  `model_axis`, and `effort_axis` from resolver output and the actual launcher
+  payload after payload construction.
+- Those fields are launcher-owned. Agent self-report cannot populate or
+  overwrite them; it may only be retained separately as optional diagnostics.
 - Derive `Dispatch target` and `Effort axis` / `Model axis` from the payload.
-- After the payload is built, append a formal dispatch stamp to Dispatch Notes
-  for every implementation, fix, and review dispatch. Use the p01 grammar
-  exactly:
+- After the payload is built, append the compatibility stamp returned from
+  `formatDispatchStamp(dispatchReport)` to Dispatch Notes for every
+  implementation, fix, and review dispatch. The derived line retains the p01
+  grammar exactly:
   `Dispatch: scope=<phase-or-task> action=<implementation|fix|review> role=<implementer|fix|reviewer> producer=<slug|unknown> provenance=<declared|observed|inferred|unknown> model_axis=<axis> effort_axis=<axis> dispatch_policy=<policy|unknown> dispatch_ceiling=<value|none> target=<target|unknown>`.
-  Derive `producer` and `provenance` from the resolver payload and actual host
-  arguments. Only concrete model arguments, including same-harness route model
-  args for model-arg providers, declare producer identity. Codex materialized
-  model+effort variants declare `model_axis=selected:<model>` and
-  `effort_axis=selected:<effort>` from resolver output, but keep
-  `producer=unknown provenance=unknown` unless an observed/inferred model
-  identity is available. Base/unpinned or deferred cross-harness paths are also
-  `producer=unknown provenance=unknown` unless an observed/inferred identity is
-  available. Do not write prose-only or legacy comma-separated stamp forms.
+  Populate the report from the completed resolver and actual host arguments.
+  Only independently observed or otherwise supported runtime evidence may
+  populate runtime producer identity. Codex materialized model+effort variants
+  retain selected model/effort controls while keeping runtime identity
+  not-reported unless evidence exists. Do not write prose-only, hand-built, or
+  legacy comma-separated stamp forms.
 
 Human-facing dispatch display rules:
 
@@ -1043,10 +1097,15 @@ coordinator_target: {resolver-selected coordinator target}
 
 Tier 1 uses the already resolved exact coordinator role/model payload and sends
 the Phase Scope. A concrete Codex coordinator uses
-`providers.codex.dispatchArgs.variant`; when the registered role cannot be
-selected, use the fresh child pinned to the resolver's model and effort with
-canonical coordinator instructions. Claude and Cursor coordinator calls pass
-their exact resolver model argument.
+`providers.codex.dispatchArgs.variant` first as native `agent_type`. After the
+host accepts the spawn, record the coordinator `target`, `model_axis`, and
+`effort_axis` from resolver output and that constructed launcher payload; do not
+wait for or accept a coordinator self-report as proof. Only an actual native
+role-selection rejection permits the fresh child pinned to the resolver's model
+and effort with canonical coordinator instructions. An accepted coordinator
+that later returns `BLOCKED` has produced a coordinator outcome and cannot
+trigger fallback. Claude and Cursor coordinator calls pass their exact resolver
+model argument.
 
 Tier 2 may run the coordinator instructions in the current context only when
 that context can still dispatch every exact task worker. Tier 2 never permits
@@ -1067,6 +1126,8 @@ For each task in dependency order, the coordinator must:
      --role implementer \
      --ceiling-tier <project-or-phase-named-tier> \
      --candidate-model <exact-model> \
+     --report-scope <task-id> \
+     --report-action implementation \
      --project-path "$PROJECT_PATH" \
      --json
    ```
@@ -1077,14 +1138,19 @@ For each task in dependency order, the coordinator must:
    is read-only and must never persist its override.
 
 3. Build the actual provider invocation before logging:
-   - Codex uses `providers.codex.dispatchArgs.variant` as `agent_type`, or the
-     exact fresh pinned-child model/effort route when native role selection is
-     unavailable.
+   - Codex first uses `providers.codex.dispatchArgs.variant` as native
+     `agent_type`. Spawn acceptance establishes the configured invocation; only
+     a native role-selection rejection permits the exact fresh pinned-child
+     model/effort route.
    - Claude passes `providers.claude.dispatchArgs.model` as the actual Task
      `model`.
    - Cursor passes `providers.cursor.dispatchArgs.model` byte-for-byte as the
      actual invocation model. Treat the string as opaque and never normalize or
      infer capability from it.
+   - After construction, record the task-worker `target`, `model_axis`, and
+     `effort_axis` from resolver output and the launcher payload. Missing worker
+     telemetry or self-report is not unavailability, and self-report cannot
+     replace those launcher-owned values.
 4. Send one bounded Task Scope, never the full phase task list:
 
    ```yaml
@@ -1135,8 +1201,8 @@ After the implementer returns DONE (or DONE_WITH_CONCERNS without correctness co
 **Dispatch:**
 
 - Use the same tier that was selected at start.
-- For Codex with a capped managed policy, dispatch the materialized reviewer role returned in `providers.codex.dispatchArgs.variant` for deterministic quality gates.
-- If that exact registered reviewer role is not selectable in the current host session, launch a fresh Codex child with explicit model, reasoning effort, and canonical role instructions from `.agents/agents/oat-reviewer.md`; never substitute the managed base role and never require restart/hot reload.
+- For Codex with a capped managed policy, first dispatch the materialized reviewer role returned in `providers.codex.dispatchArgs.variant` as native `agent_type` for deterministic quality gates. After spawn acceptance, record the review `target`, `model_axis`, and `effort_axis` from resolver output and the constructed launcher payload; reviewer self-report cannot populate or overwrite them.
+- Only if the exact registered reviewer role receives a native role-selection rejection may the launcher start a fresh Codex child with explicit model, reasoning effort, and canonical role instructions from `.agents/agents/oat-reviewer.md`; never substitute the managed base role and never require restart/hot reload. Missing reviewer telemetry or self-report is not a rejection, and an accepted reviewer that later returns `BLOCKED` cannot trigger fallback.
 - For Codex with managed `Uncapped` or inherit/default mode, no reviewer target exists; use base `oat-reviewer`, log `effort_axis=provider-default`, and explain that the base role follows the provider default.
 - For Claude Code with a capped managed policy, require `providers.claude.dispatchArgs.model` and pass that exact value as the review `model`; managed `Uncapped` or inherit/default mode omits `model` because no reviewer target exists. Always keep `effort_axis=not-applicable`.
 - For Cursor with a concrete managed reviewer target, require `providers.cursor.dispatchArgs.model` and pass that exact opaque, unnormalized string as the actual review invocation's `model` argument.
@@ -1165,7 +1231,7 @@ After the implementer returns DONE (or DONE_WITH_CONCERNS without correctness co
   - For Codex Tier 1 dispatches, send the Review Scope block as a self-contained packet and keep fresh context (`fork_context: false`). The reviewer is expected to reconstruct context from git state and the OAT artifacts listed above.
   - For Codex Tier 1 review dispatches, use the materialized Codex role name from `providers.codex.dispatchArgs.variant` only when the resolver returns a reviewer variant for a capped managed policy. A Codex materialized reviewer role selected from a model+effort target must carry `model_axis=selected:<model>` and `effort_axis=selected:<effort>` from resolver output. Use base `oat-reviewer` only when the resolver returns no `dispatchArgs.variant` for managed `Uncapped`, inherit/default mode, or provider-default fallback, and log `effort_axis=provider-default`. For Claude Code, pass `model: providers.claude.dispatchArgs.model` for a concrete managed reviewer and never pass a per-review effort override. For Cursor, pass `model: providers.cursor.dispatchArgs.model` byte-for-byte for a concrete managed reviewer.
   - Treat the commit range as authoritative for review scope. `files_changed` is optional orientation metadata only.
-  - If a reviewer does not return a terminal result on the first wait, poll once more. If it still has not concluded, send one concise nudge to return immediately with current findings. If the reviewer still does not conclude, treat the target-preserving review dispatch as failed for this phase. Retry the same exact role, pinned fresh-child route, or complete Claude/Cursor invocation payload within the retry bound, preserving the exact model argument; never downgrade a timed-out managed reviewer to unpinned inline execution.
+  - If a reviewer does not return a terminal result on the first wait, poll once more. If it still has not concluded, send one concise nudge to return immediately with current findings. If the reviewer still does not conclude, treat the target-preserving review dispatch as failed for this phase. When the original native reviewer spawn was accepted, retry the same already-selected native `agent_type` route within the retry bound; do not switch that timed-out reviewer to a fresh pinned child. The fresh pinned-child route is eligible only when the original native attempt received explicit pre-start role-selection rejection. Retry an already-selected pinned fresh-child route or complete Claude/Cursor invocation payload without changing routes, preserving the exact model argument; never downgrade a timed-out managed reviewer to unpinned inline execution.
 
 - Tier 2: read `.agents/agents/oat-reviewer.md` and review inline only with verified equivalent current-host model and effort controls, explicit inherit/default behavior, or the documented managed-uncapped reviewer behavior. Otherwise block.
 
@@ -1175,6 +1241,9 @@ Parse the reviewer's confirmation for verdict + finding severities. Map to pass
 
 - **pass:** zero Critical and zero Important findings.
 - **fail:** one or more Critical or Important findings.
+- **blocked:** An accepted reviewer `BLOCKED` terminal blocks this phase review.
+  It does not invoke fallback and must not be interpreted as a pass due to
+  absent findings. Stop and surface the review target and blocker reason.
 
 Medium / Minor findings do not block the phase but are recorded.
 
@@ -1185,7 +1254,7 @@ On reviewer verdict `fail`, run a bounded fix loop.
 1. Read `oat_orchestration_retry_limit` from `state.md` frontmatter (default: `2`, range 0–5).
 2. For each retry (up to the limit):
    a. Convert Critical/Important findings into bounded fix scopes associated with one planned task/file boundary at a time. Do not hand one worker the full phase finding list.
-   b. Reuse the phase coordinator in `fix` mode. It selects an exact candidate under the same project or phase named ceiling with `--ceiling-tier`, then emits one Task Scope per bounded fix. Codex uses `providers.codex.dispatchArgs.variant`; Claude and Cursor pass their exact `providers.<provider>.dispatchArgs.model` value on the actual invocation. Every fix worker writes the formal `Dispatch: scope=<phase-or-task> action=fix role=fix producer=<slug|unknown> provenance=<declared|observed|inferred|unknown> model_axis=<axis> effort_axis=<axis> dispatch_policy=<policy|unknown> dispatch_ceiling=<value|none> target=<target|unknown>` stamp before execution.
+   b. Reuse the phase coordinator in `fix` mode. It selects an exact candidate under the same project or phase named ceiling with `--ceiling-tier`, then emits one Task Scope per bounded fix. Codex first uses `providers.codex.dispatchArgs.variant` as native `agent_type`; only a native role-selection rejection permits the exact fresh-child fallback. Claude and Cursor pass their exact `providers.<provider>.dispatchArgs.model` value on the actual invocation. After constructing the launcher payload, record the fix `target`, `model_axis`, and `effort_axis` from that payload and resolver output. Missing fix-worker telemetry or self-report is not unavailability, and an accepted fix worker — including one that returns `BLOCKED` — cannot trigger fallback. Every fix worker writes the formal `Dispatch: scope=<phase-or-task> action=fix role=fix producer=<slug|unknown> provenance=<declared|observed|inferred|unknown> model_axis=<axis> effort_axis=<axis> dispatch_policy=<policy|unknown> dispatch_ceiling=<value|none> target=<target|unknown>` stamp before execution.
    c. Receive and verify each fix result and commit. The coordinator must not apply fixes itself, and Tier 2 does not authorize inline task edits.
    d. Re-dispatch the reviewer with the updated commit range.
    e. Parse the new verdict.
@@ -1628,6 +1697,10 @@ All must pass before proceeding.
 
 **At the final plan phase boundary, a code review is required before PR.**
 
+An accepted reviewer `BLOCKED` terminal blocks final review. It does not invoke
+fallback and must not be interpreted as a pass due to absent findings. Stop and
+surface the review target and blocker reason.
+
 Before requesting final review, ensure the latest project-artifact bookkeeping is already committed. Review should evaluate the implementation state as it actually stands on the branch, not a half-tracked working tree.
 
 Check if final review already completed (preferred source of truth: plan.md Reviews table):
@@ -1654,13 +1727,22 @@ echo "$FINAL_ROW"
 **Workflow preference check (before prompting):**
 
 First resolve the final reviewer target with the same target-first contract as
-per-phase review. A concrete managed Codex target must use its exact registered
-reviewer or an explicitly pinned fresh child. A concrete managed Claude or
-Cursor target must put `providers.claude.dispatchArgs.model` or
+per-phase review. A concrete managed Codex target must first send its exact
+registered reviewer as native `agent_type`; only a native role-selection
+rejection permits an explicitly pinned fresh child. Spawn acceptance plus the
+launcher payload supplies configured invocation evidence, so missing telemetry,
+missing self-report, or a later `BLOCKED` result cannot trigger fallback. Record
+the final review `target`, `model_axis`, and `effort_axis` from resolver output
+and the constructed launcher payload, never from reviewer self-report. A
+concrete managed Claude or Cursor target must put
+`providers.claude.dispatchArgs.model` or
 `providers.cursor.dispatchArgs.model` respectively into the actual provider
 invocation as the exact `model` argument; Cursor strings remain opaque. On
-timeout or retry, preserve the same exact role or complete invocation payload,
-including the model argument. If the host cannot apply the required role or
+timeout or retry, preserve the already-selected route as well as its complete
+invocation payload: an accepted native reviewer retries the same native
+`agent_type`, while a fresh pinned-child route is eligible only when the
+original native attempt received explicit pre-start role-selection rejection.
+Preserve the exact model argument. If the host cannot apply the required role or
 model argument, fail closed or block unless verified equivalent current-host
 controls permit inline execution. The preference below chooses only among
 routes that preserve that target; it cannot authorize generic inline or base

package/assets/skills/oat-project-review-provide/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: oat-project-review-provide
-version: 1.3.12
+version: 1.3.14
 description: Use when the user explicitly asks to review an OAT project — e.g. "review project", "review the project", "run project review", or confirms a previously offered review. Do NOT auto-invoke on completed work alone. Resolves a project review scope and offers before running.
 disable-model-invocation: false
 user-invocable: true
@@ -523,9 +523,25 @@ Before capability-tier selection, resolve the same reviewer contract used by
 plan artifact review and implementation phase/final review:
 
 ```bash
-oat project dispatch-ceiling resolve --provider "$ACTIVE_PROVIDER" --role reviewer --preflight --json
+oat project dispatch-ceiling resolve --provider "$ACTIVE_PROVIDER" --role reviewer --preflight --report-scope "$SCOPE_TOKEN" --report-action review --json
 ```
 
+Require `dispatchReport.schemaVersion: 1`. Render/consume the resolver's
+versioned report using `formatDispatchReport(dispatchReport)` semantics, and
+derive the formal compatibility line only with
+`formatDispatchStamp(dispatchReport)` / `toDispatchStampRecord(dispatchReport)`.
+Include that derived line in the review dispatch audit metadata; do not
+hand-assemble `Dispatch:` fields from a role name or model string.
+
+The exact managed provider target still comes from
+`providers.<provider>.dispatchArgs` plus
+`providers.<provider>.selection.target`, and the actual reviewer invocation
+must retain it byte-for-byte. The report is an audit/rendering surface, not a
+selection fallback. Configured invocation (including gate-owned immutable
+metadata), work-producer diversity, and independently observed reviewer runtime identity are distinct. Do not promote producer stamps, configured defaults, or
+reviewer self-report into `dispatchReport.runtimeIdentity`; leave it
+`not-reported` unless an independent observation exists.
+
 The reviewer resolver selects the final candidate of the configured review
 ceiling. Do not supply ephemeral implementer candidate requests for artifact,
 phase, project, or final review. A lower candidate is allowed only when a
@@ -539,13 +555,16 @@ fallbacks.
 A concrete managed Codex target takes precedence over tier availability.
 Build the actual provider invocation before reporting the target as enforced:
 
-- Codex uses the exact registered reviewer role or variant from
-  `providers.codex.dispatchArgs.variant` when selectable. If the current host
-  cannot select it, launch a fresh Codex child with the resolver target's
-  explicit model, reasoning effort, canonical role instructions from
-  `.agents/agents/oat-reviewer.md`, and the same Review Scope payload.
-  If that fresh child is unavailable, use only a verified-equivalent inline
-  route or block the review.
+- Codex must send the resolver-returned Codex variant first as the native
+  `agent_type`. Spawn acceptance plus the complete launcher payload is
+  configured invocation evidence; it does not require reviewer self-report or
+  separate runtime telemetry. A native role-selection rejection means an
+  explicit host rejection of that exact `agent_type` before any reviewer or
+  child starts. Only after that pre-start rejection may the workflow launch a
+  fresh Codex child with the resolver target's explicit model, reasoning
+  effort, canonical role instructions from `.agents/agents/oat-reviewer.md`,
+  and the same Review Scope payload. If neither exact route is available, use
+  only a verified-equivalent inline route or block the review.
 - Claude requires a non-empty `providers.claude.dispatchArgs.model`; the actual
   provider invocation must include that exact value as its `model` argument.
 - Cursor requires a non-empty `providers.cursor.dispatchArgs.model`; the actual
@@ -557,6 +576,18 @@ candidate ladder, fail closed before review. Route interactive repair through
 the planning workflow's `Complete Dispatch Ladder Adoption Contract`; do not
 invent a reviewer target.
 
+After constructing the complete provider payload, record the launcher-owned
+`target`, `model_axis`, and `effort_axis` with
+`launcher-selected/config-declared` provenance. These fields are immutable:
+missing telemetry, missing reviewer self-report, or contradictory self-report
+must not populate, replace, or overwrite them and must not trigger fallback.
+
+Once the native host accepts a reviewer, every terminal result is an
+authoritative review outcome. An accepted reviewer returning `BLOCKED` is a
+terminal blocking review outcome: it blocks the review and does not invoke or
+trigger the fresh-child fallback. Absent findings from that terminal must not
+be parsed, interpreted, or treated as a passing review.
+
 On timeout or retry, reuse the same exact role or complete invocation payload,
 including the Claude or Cursor model argument. If the host cannot apply a
 required role or model argument, fail closed or block unless inline execution
@@ -607,7 +638,7 @@ Then spawn the reviewer:
 - Use provider-appropriate dispatch:
   - Claude Code: Task tool with `subagent_type: "oat-reviewer"` (resolves from `.claude/agents/oat-reviewer.md`). For a concrete managed target, the payload must also contain `model: providers.claude.dispatchArgs.model` with the resolver-returned value.
   - Cursor: explicit invocation `/oat-reviewer` (or natural mention) with agent resolved from `.cursor/agents/oat-reviewer.md` or `.claude/agents/oat-reviewer.md` compatibility path. For a concrete managed target, the invocation must also contain `model: providers.cursor.dispatchArgs.model` with the exact opaque resolver-returned string.
-  - Codex style: for a concrete managed target, spawn the exact resolver-returned `agent_type`; if it cannot be selected, use the explicitly pinned fresh-child route from Step 6.0. Generic auto-selection is permitted only for the documented base-role exceptions.
+  - Codex style: for a concrete managed target, first spawn the exact resolver-returned native `agent_type`; only an explicit pre-start native role-selection rejection permits the explicitly pinned fresh-child route from Step 6.0. Generic auto-selection is permitted only for the documented base-role exceptions.
 - Pass the Review Scope metadata block from Step 5 as the prompt
 - Include the pre-computed artifact path for the subagent to write to
 - **If a worktree was resolved in Step 1.5:** include the worktree path in the prompt so the subagent writes the artifact to the worktree directory, not the current session's working directory
@@ -617,6 +648,9 @@ The `oat-reviewer` agent definition contains the full review process, mode contr
 
 After the subagent completes:
 
+- Treat its terminal status as authoritative. An accepted `BLOCKED` result
+  blocks the review; do not invoke fallback and do not infer a pass from a
+  missing review artifact or absent findings.
 - Verify the review artifact was written to the expected path
 - Continue with Step 9 (plan update) and Step 9.5 (commit)
 

package/assets/skills/oat-project-review-provide-remote/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: oat-project-review-provide-remote
-version: 1.0.2
+version: 1.0.3
 description: Use when reviewing a GitHub PR opened on another machine for an active OAT project and posting findings back as a single PR review. Resolves the project from the PR diff, reads project artifacts for mode-aware review, and posts via gh api.
 disable-model-invocation: true
 user-invocable: true
@@ -211,6 +211,33 @@ Guard outcomes:
 
 Mode-aware review against the project artifacts + the `oat-reviewer` checklist + severity model. Scope to the narrowing range when one was chosen; otherwise the full PR diff. Assign finding IDs per severity bucket (`C1`, `I1`, `M1`, `m1`), stable within the dispatch.
 
+**Step 5.0: Resolve and report the exact reviewer target.** Before capability
+probing, run:
+
+```bash
+oat project dispatch-ceiling resolve --provider "$ACTIVE_PROVIDER" --role reviewer --preflight --project-path "$PROJECT_PATH" --report-scope "$SCOPE_TOKEN" --report-action review --json
+```
+
+Require `dispatchReport.schemaVersion: 1`. Render/consume its versioned human
+block with `formatDispatchReport(dispatchReport)` semantics and derive the
+formal audit `Dispatch:` line only through
+`formatDispatchStamp(dispatchReport)` / `toDispatchStampRecord(dispatchReport)`.
+Never hand-build a parallel stamp schema.
+
+Retain the exact managed provider payload from
+`providers.<provider>.dispatchArgs` and
+`providers.<provider>.selection.target` in the actual Tier 1/2/3 invocation;
+the report does not authorize substitution or lowering. Configured gate
+invocation, work-producer diversity, and independently observed reviewer
+`dispatchReport.runtimeIdentity` remain separate. Do not treat PR markers, implementation
+producer stamps, configured defaults, or reviewer self-report as runtime
+confirmation. Leave report runtime identity not-reported unless independently
+observed.
+
+This resolver command is independent of lifecycle gates. Reusable `oat gate
+review` commands must remain provider-neutral and must not contain or add a
+provider/model `--target` argument.
+
 **Step 5a: Probe subagent availability** (mirrors `oat-project-review-provide` Step 6a):
 
 ```

package/dist/commands/config/index.d.ts

@@ -3,6 +3,7 @@ import { type PromptContext } from '../shared/shared.prompts.js';
 import { type OatConfig, type OatLocalConfig, type UserConfig } from '../../config/oat-config.js';
 import { type ResolvedConfig } from '../../config/resolve.js';
 import { type MatrixCellAvailabilityResponse, type ValidateMatrixCellOptions } from '../../providers/identity/availability.js';
+import { createDispatchValidationPassContext, validateDispatchMatrixRefs } from '../../providers/identity/dispatch-validation.js';
 import { Command } from 'commander';
 interface ConfigCommandDependencies {
     buildCommandContext: (options: Parameters<typeof buildCommandContext>[0]) => CommandContext;
@@ -19,6 +20,8 @@ interface ConfigCommandDependencies {
     readFile: (path: string) => Promise<string>;
     confirmAction: (message: string, ctx: PromptContext) => Promise<boolean>;
     validateMatrixCell: (provider: string, value: string, options: ValidateMatrixCellOptions) => Promise<MatrixCellAvailabilityResponse>;
+    createDispatchValidationPassContext: typeof createDispatchValidationPassContext;
+    validateDispatchMatrixRefs: typeof validateDispatchMatrixRefs;
     processEnv: NodeJS.ProcessEnv;
 }
 export declare function createConfigCommand(overrides?: Partial<ConfigCommandDependencies>): Command;

package/dist/commands/config/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/commands/config/index.ts"],"names":[],"mappings":"AAGA,OAAO,EAAE,mBAAmB,EAAE,KAAK,cAAc,EAAE,MAAM,sBAAsB,CAAC;AAEhF,OAAO,EAEL,KAAK,aAAa,EACnB,MAAM,iCAAiC,CAAC;AAQzC,OAAO,EAML,KAAK,SAAS,EACd,KAAK,cAAc,EAGnB,KAAK,UAAU,EAgBhB,MAAM,oBAAoB,CAAC;AAC5B,OAAO,EAEL,KAAK,cAAc,EAEpB,MAAM,iBAAiB,CAAC;AAGzB,OAAO,EAGL,KAAK,8BAA8B,EACnC,KAAK,yBAAyB,EAC/B,MAAM,kCAAkC,CAAC;AAC1C,OAAO,EAAE,OAAO,EAAE,MAAM,WAAW,CAAC;AAkFpC,UAAU,yBAAyB;IACjC,mBAAmB,EAAE,CACnB,OAAO,EAAE,UAAU,CAAC,OAAO,mBAAmB,CAAC,CAAC,CAAC,CAAC,KAC/C,cAAc,CAAC;IACpB,kBAAkB,EAAE,CAAC,GAAG,EAAE,MAAM,KAAK,OAAO,CAAC,MAAM,CAAC,CAAC;IACrD,aAAa,EAAE,CAAC,QAAQ,EAAE,MAAM,KAAK,OAAO,CAAC,SAAS,CAAC,CAAC;IACxD,cAAc,EAAE,CAAC,QAAQ,EAAE,MAAM,EAAE,MAAM,EAAE,SAAS,KAAK,OAAO,CAAC,IAAI,CAAC,CAAC;IACvE,kBAAkB,EAAE,CAAC,QAAQ,EAAE,MAAM,KAAK,OAAO,CAAC,cAAc,CAAC,CAAC;IAClE,mBAAmB,EAAE,CACnB,QAAQ,EAAE,MAAM,EAChB,MAAM,EAAE,cAAc,KACnB,OAAO,CAAC,IAAI,CAAC,CAAC;IACnB,cAAc,EAAE,CAAC,aAAa,EAAE,MAAM,KAAK,OAAO,CAAC,UAAU,CAAC,CAAC;IAC/D,eAAe,EAAE,CAAC,aAAa,EAAE,MAAM,EAAE,MAAM,EAAE,UAAU,KAAK,OAAO,CAAC,IAAI,CAAC,CAAC;IAC9E,mBAAmB,EAAE,CACnB,QAAQ,EAAE,MAAM,EAChB,GAAG,EAAE,MAAM,CAAC,UAAU,KACnB,OAAO,CAAC,MAAM,CAAC,CAAC;IACrB,sBAAsB,EAAE,CACtB,QAAQ,EAAE,MAAM,EAChB,aAAa,EAAE,MAAM,EACrB,GAAG,EAAE,MAAM,CAAC,UAAU,KACnB,OAAO,CAAC,cAAc,CAAC,CAAC;IAC7B,iBAAiB,EAAE,MAAM,OAAO,CAAC,MAAM,CAAC,CAAC;IACzC,QAAQ,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,OAAO,CAAC,MAAM,CAAC,CAAC;IAC5C,aAAa,EAAE,CAAC,OAAO,EAAE,MAAM,EAAE,GAAG,EAAE,aAAa,KAAK,OAAO,CAAC,OAAO,CAAC,CAAC;IACzE,kBAAkB,EAAE,CAClB,QAAQ,EAAE,MAAM,EAChB,KAAK,EAAE,MAAM,EACb,OAAO,EAAE,yBAAyB,KAC/B,OAAO,CAAC,8BAA8B,CAAC,CAAC;IAC7C,UAAU,EAAE,MAAM,CAAC,UAAU,CAAC;CAC/B;AA45DD,wBAAgB,mBAAmB,CACjC,SAAS,GAAE,OAAO,CAAC,yBAAyB,CAAM,GACjD,OAAO,CAsKT"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/commands/config/index.ts"],"names":[],"mappings":"AAGA,OAAO,EAAE,mBAAmB,EAAE,KAAK,cAAc,EAAE,MAAM,sBAAsB,CAAC;AAEhF,OAAO,EAEL,KAAK,aAAa,EACnB,MAAM,iCAAiC,CAAC;AAezC,OAAO,EAGL,KAAK,SAAS,EACd,KAAK,cAAc,EAGnB,KAAK,UAAU,EAahB,MAAM,oBAAoB,CAAC;AAC5B,OAAO,EAEL,KAAK,cAAc,EAEpB,MAAM,iBAAiB,CAAC;AAGzB,OAAO,EAGL,KAAK,8BAA8B,EACnC,KAAK,yBAAyB,EAC/B,MAAM,kCAAkC,CAAC;AAC1C,OAAO,EACL,mCAAmC,EACnC,0BAA0B,EAC3B,MAAM,yCAAyC,CAAC;AACjD,OAAO,EAAE,OAAO,EAAE,MAAM,WAAW,CAAC;AAkFpC,UAAU,yBAAyB;IACjC,mBAAmB,EAAE,CACnB,OAAO,EAAE,UAAU,CAAC,OAAO,mBAAmB,CAAC,CAAC,CAAC,CAAC,KAC/C,cAAc,CAAC;IACpB,kBAAkB,EAAE,CAAC,GAAG,EAAE,MAAM,KAAK,OAAO,CAAC,MAAM,CAAC,CAAC;IACrD,aAAa,EAAE,CAAC,QAAQ,EAAE,MAAM,KAAK,OAAO,CAAC,SAAS,CAAC,CAAC;IACxD,cAAc,EAAE,CAAC,QAAQ,EAAE,MAAM,EAAE,MAAM,EAAE,SAAS,KAAK,OAAO,CAAC,IAAI,CAAC,CAAC;IACvE,kBAAkB,EAAE,CAAC,QAAQ,EAAE,MAAM,KAAK,OAAO,CAAC,cAAc,CAAC,CAAC;IAClE,mBAAmB,EAAE,CACnB,QAAQ,EAAE,MAAM,EAChB,MAAM,EAAE,cAAc,KACnB,OAAO,CAAC,IAAI,CAAC,CAAC;IACnB,cAAc,EAAE,CAAC,aAAa,EAAE,MAAM,KAAK,OAAO,CAAC,UAAU,CAAC,CAAC;IAC/D,eAAe,EAAE,CAAC,aAAa,EAAE,MAAM,EAAE,MAAM,EAAE,UAAU,KAAK,OAAO,CAAC,IAAI,CAAC,CAAC;IAC9E,mBAAmB,EAAE,CACnB,QAAQ,EAAE,MAAM,EAChB,GAAG,EAAE,MAAM,CAAC,UAAU,KACnB,OAAO,CAAC,MAAM,CAAC,CAAC;IACrB,sBAAsB,EAAE,CACtB,QAAQ,EAAE,MAAM,EAChB,aAAa,EAAE,MAAM,EACrB,GAAG,EAAE,MAAM,CAAC,UAAU,KACnB,OAAO,CAAC,cAAc,CAAC,CAAC;IAC7B,iBAAiB,EAAE,MAAM,OAAO,CAAC,MAAM,CAAC,CAAC;IACzC,QAAQ,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,OAAO,CAAC,MAAM,CAAC,CAAC;IAC5C,aAAa,EAAE,CAAC,OAAO,EAAE,MAAM,EAAE,GAAG,EAAE,aAAa,KAAK,OAAO,CAAC,OAAO,CAAC,CAAC;IACzE,kBAAkB,EAAE,CAClB,QAAQ,EAAE,MAAM,EAChB,KAAK,EAAE,MAAM,EACb,OAAO,EAAE,yBAAyB,KAC/B,OAAO,CAAC,8BAA8B,CAAC,CAAC;IAC7C,mCAAmC,EAAE,OAAO,mCAAmC,CAAC;IAChF,0BAA0B,EAAE,OAAO,0BAA0B,CAAC;IAC9D,UAAU,EAAE,MAAM,CAAC,UAAU,CAAC;CAC/B;AAo0DD,wBAAgB,mBAAmB,CACjC,SAAS,GAAE,OAAO,CAAC,yBAAyB,CAAM,GACjD,OAAO,CAsKT"}