qfai 1.7.14 → 1.8.0

package/README.md

@@ -17,8 +17,12 @@ The agent reads the repository, produces the required artifacts, and iterates un
 
 ## Release status
 
-- Current package version: `1.7.14`
-- Release posture: v1.7.14 converges init assets, validators, and docs on the canonical sidecar model.
+- Current package version: `1.7.15`
+- Release posture: v1.7.15 enforces runtime truthfulness.
+- Prototyping is UI-only; `full-harness` is measurement-driven iteration accumulation for UI-bearing surfaces only.
+- Runtime observation is observed-only (no synthetic 200 / API / DB prototyping coverage).
+- Browser QA is mandatory per screen in full-harness, and `actionsWired` reports action coverage rather than finding count.
+- Calibration SSOT is the calibration pack referenced by `calibrationRef.packPath`.
 - Current repo note: some repo-wide `qfai validate --fail-on error` blockers still come from historical review/evidence/ATDD/TDD artifacts and are being cleaned incrementally.
 
 ## Quick start
@@ -54,8 +58,27 @@ npx qfai report
 - `npx qfai doctor`
   - Diagnoses configuration discovery, path resolution, glob scanning, and `validate.json` inputs before running validate/report; use `--fail-on` to enforce failures in CI.
     Note: prototyping evidence (`.qfai/evidence/prototyping.json`) is produced by the AI workflow / skills
-    (`/qfai-prototyping` with `mode=low-cost|standard|full-harness`), not by a CLI command.
+    (`/qfai-prototyping` with `mode=full-harness` for supported UI surfaces only), not by a general-purpose end-user CLI flow.
     `qfai validate` consumes the resulting evidence files, including `mode.effective` and `fullHarness` metadata when present.
+    Traceability refs inside prototyping evidence must use repo-root-relative concrete artifact refs (for example `.qfai/specs/spec-0001/01_Spec.md#L3` or `.qfai/evidence/render.json#/screens/0`).
+    Absolute paths are invalid. The same strict ref grammar is enforced for top-level and leaf evidence-bearing fields, including
+    `runtimeGate.evidenceRefs`, `runtimeGate.ui[].declaredRef`, `runtimeGate.ui[].renderEvidenceRefs[]`,
+    `runtimeGate.ui[].browserQaEvidenceRefs[]`, `specs[].coverageRefs[].declaredRef`, `specs[].coverageRefs[].observedRefs[]`,
+    `fullHarness.iterations[].evidenceRefs.runtimeGate`, `fullHarness.iterations[].evidenceRefs.specCoverage`,
+    `fullHarness.iterations[].evidenceRefs.render`, `fullHarness.iterations[].evidenceRefs.browserQa`,
+    `fullHarness.iterations[].evidenceRefs.uiObservation`, `fullHarness.iterations[].evidenceRefs.discussion`,
+    `fullHarness.iterations[].evidenceRefs.screenContract`, `fullHarness.iterations[].evidenceRefs.trend`,
+    `fullHarness.iterations[].l1.axes[].evidenceRefs[]`, `fullHarness.iterations[].l2.axes[].evidenceRefs[]`, and
+    `fullHarness.reviewerLogs[].evidenceRefs[]`.
+    Semantic rules are also strict: `runtimeGate.ui[].declaredRef` and `fullHarness.iterations[].evidenceRefs.screenContract[]`
+    must use the canonical screen contract sourceRef `.qfai/discussion/<pack>/uiux/40_screen_contracts.md#<screenId>`,
+    and `specs[].coverageRefs[].declaredRef` must use the canonical spec declaration form
+    `.qfai/specs/<specId>/01_Spec.md#L<line>` (for example `.qfai/specs/spec-0001/01_Spec.md#L3`);
+    `notes.md`, `appendix.md`, anchor-fragment forms such as `#route-home`, discussion refs, and screen contract refs
+    are NOT valid `declaredRef` values.
+    `fullHarness` follows a terminal-first state machine: `status="in-progress"` requires `finalDecision="pending"`,
+    `reviewerSignoff.status="pending"`, and no `terminationReason`; `status="completed"` requires `terminationReason`,
+    a non-pending `finalDecision`, and a terminal `reviewerSignoff`.
 
 ## ATDD annotation hard gate
 
@@ -89,7 +112,7 @@ QFAI includes a small set of custom skills (stored under `.qfai/assistant/skills
   as 15 required markdown files under `.qfai/discussion/discussion-<ts>/`.
   UI-bearing discussion packs require `prototyping.yaml`; non-ui discussion packs do not.
 - **qfai-sdd**: Unified SDD entrypoint with discussion-pack preflight guard (missing/incomplete/blocking OQ causes stop + next action guidance).
-- **qfai-prototyping**: Build a contract-aligned implementation skeleton with static-first evidence by default, and escalate to full-harness only when explicitly justified.
+- **qfai-prototyping**: Build a contract-aligned UI prototype under the `full-harness` only / UI-only contract, with calibration-pack SSOT and screen-level Browser QA evidence.
 - **qfai-atdd**: Implement acceptance tests driven by specs/scenarios.
 - **qfai-implement**: Unified TDD micro-cycle (Red/Green/Refactor) one test at a time using `test-list.md` as the execution ledger, including ledger status updates and exception closure.
 - **qfai-verify**: Run full-scan local quality gates (`validate --fail-on error`, `report`, repo gates) and produce reviewer-approved evidence under `.qfai/evidence/`.
@@ -191,7 +214,9 @@ Notes.
 
 - `validate.json`, `report.json`, `doctor.json`, and `run-*` JSON logs are internal exports and are not a stable external contract; prefer `report.md` for integrations that must survive tool upgrades.
 - Scenario files are expected to use the Gherkin extension `*.feature` (not `*.md`).
-- `prototyping.calibration` in `qfai.config.yaml` connects full-harness scoring thresholds to the report and validator.
+- `prototyping.calibration.packPath` points to the calibration pack SSOT; runtime and validator both resolve thresholds and iteration parameters from that pack.
+- `prototyping.calibration.thresholds`, `maxIterations`, `plateauDelta`, and `plateauLookback` are unsupported public config fields in v1.7.15.
+  Put calibration values in the referenced pack instead of `qfai.config.yaml`.
 - Observability modules (`src/core/observability/`) exist as foundation code but are **not integrated into blocking validation** in v1.7.14. They are reserved for future operational instrumentation.
 
 ## Specifications and contracts (SDD)

package/assets/init/.qfai/assistant/instructions/agent-selection.md

@@ -7,6 +7,8 @@ dependencies:
 version: 2.0.0
 ---
 
+<!-- markdownlint-disable MD041 -->
+
 > **言語指示（厳守）**
 >
 > - 報告・出力: 日本語（Plan も含む）

package/assets/init/.qfai/assistant/instructions/shared-skill-delegation-baseline.md

@@ -0,0 +1,88 @@
+# Shared Skill Delegation Baseline
+
+Use this document to keep SKILL bodies compact.
+Skill files should reference this baseline and only add role-, stage-, or gate-specific rules.
+
+## Sub-agent Delegation (MANDATORY)
+
+### Orchestrator Protocol (MUST)
+
+- The orchestrator may create work orders, delegate tasks, integrate outputs, and present results.
+- The orchestrator must not generate the primary artifact first draft.
+- The orchestrator must not self-approve or act as reviewer for convenience.
+
+### Capability Probe (MUST)
+
+1. Attempt the first required delegation at stage start using the platform's native delegation mechanism.
+2. Treat that first real delegation attempt as the capability check. Do not gate execution on preflight availability questions or synthetic probe-only checks.
+3. If the delegation fails, stop the stage immediately. Do not simulate roles and do not continue with self-execution.
+
+### Delegation Failure (Hard Stop)
+
+- Report all of:
+  - `Delegation failure: <raw reason or concise summary>`
+  - `Attempted role: <role>`
+  - `Attempted task: <task title>`
+  - `Why stopped: QFAI requires real sub-agent delegation in this environment.`
+  - `User action needed: <settings or tooling changes required>`
+  - `Retry condition: rerun after the required delegation succeeds`
+
+## Work Orders Summary
+
+Every major artifact in the stage should include this table schema:
+
+| Step | Role (sub-agent) | Task title | Input (refs) | Output (refs) | Status (PASS/REVISE) |
+| ---- | ---------------- | ---------- | ------------ | ------------- | -------------------- |
+| 1    | <role>           | <task>     | <refs>       | <refs>        | PASS/REVISE          |
+
+- `Output (refs)` should point to in-file anchors or relative evidence paths.
+
+## Reviewer Gate Baseline
+
+- Final completion gate must be delegated to an independent reviewer.
+- Reviewers must verify Drift Protocol enforcement.
+- Reviewers must verify test-layer policy enforcement when relevant.
+- Do not treat test volume ratios or floors as hard gates unless the skill explicitly says so.
+- Do not declare DONE until all routed blocking reviewers return `PASS`.
+- Every reviewer returning `FAIL` or `REVISE` must include a concrete fix proposal.
+
+## Work order template
+
+```text
+Task title: <short>
+Role: <sub-agent role>
+Goal: <what to decide/produce>
+Inputs (refs):
+- <file/section>
+Constraints:
+- must: enforce Drift Protocol
+- must: follow applicable test-layer or validation policy
+- must_not: patch upstream artifacts directly when owner rerun is required
+Output format:
+- <headings / bullet schema>
+Quality bar:
+- PASS if ...
+- REVISE if ...
+```
+
+## Reviewer response template
+
+```text
+Result: PASS | REVISE
+Findings:
+- <issue>
+Required fixes:
+- <action>
+Evidence checked:
+- <refs>
+```
+
+### Verdict vocabulary
+
+- Reviewer responses in-flight use `Result: PASS | REVISE` (this file).
+- `summary.json` archived into review packs historically uses
+  `status: "PASS|FAIL"` (validated by
+  `packages/qfai/src/core/validators/reviewArtifacts.ts`).
+- A `REVISE` verdict during iteration maps to `status: "FAIL"` when the
+  final `summary.json` is written; they represent the same outcome.
+  Review packs should not invent a third verdict.

package/assets/init/.qfai/assistant/instructions/shared-skill-operating-baseline.md

@@ -0,0 +1,49 @@
+# Shared Skill Operating Baseline
+
+Use this document to keep SKILL bodies compact.
+Skill files should reference this baseline and only restate skill-specific additions or overrides.
+
+## User Questions (AskUserQuestion Protocol)
+
+- When a question to the user is needed, use AskUserQuestion if the tool is available.
+- When AskUserQuestion supports structured choices, prefer structured choices over free-text input.
+- If AskUserQuestion is unavailable, ask the same question in a normal message with explicit numbered choices.
+- Preserve structured choice semantics when falling back.
+- State why AskUserQuestion was unavailable.
+
+## FORMAT SSOT (Mandatory)
+
+- Before writing or editing `.qfai/**`, read the relevant README/template/sample for the target artifact.
+- Do not copy templates or samples into prompt markdown.
+- Generated artifacts must match README-defined structure, headings, ordering, and table columns.
+- Completion requires a format self-check in evidence.
+
+## Stage 0 - Steering completion refresh (mandatory)
+
+Refresh these files before or during the stage when facts are missing or stale:
+
+- `.qfai/assistant/steering/manifest.md`
+- `.qfai/assistant/steering/product.md`
+- `.qfai/assistant/steering/structure.md`
+- `.qfai/assistant/steering/tech.md`
+
+Rules:
+
+- Detect incomplete content such as empty sections, placeholder-only text, `<...>`, `TBD`, or stale facts.
+- Fill only what is verifiable from repository evidence.
+- If something cannot be verified, record an Open Question and ask the user.
+- Update steering when new facts are discovered during the stage.
+
+## Delta Rejected Guard (Mandatory)
+
+- Do not reintroduce options marked as rejected in `09_delta.md`.
+- If a rejected option must be reconsidered, create a `[RE-OPEN]` decision record that references the prior DR-ID, states what changed, and includes explicit approval.
+
+## Completion Contract (Shared)
+
+Before declaring completion, you MUST:
+
+- resolve or explicitly defer undefined or ambiguous items with rationale;
+- verify every expected artifact exists and required sections are populated;
+- scan generated artifacts for unresolved placeholders such as `TBD`, `TODO`, `TBA`, `TBC`, `XXX`, `???`, `OQ`, `OPEN QUESTION`, `UNDEFINED`, and `PLACEHOLDER`;
+- run the smallest applicable smoke check, or state "not applicable" with a short rationale.

package/assets/init/.qfai/assistant/skills/qfai-atdd/SKILL.md

@@ -30,21 +30,16 @@ QFAI Skill Body (SSOT)
 
 ## User Questions (AskUserQuestion Protocol)
 
-- When a question to the user is needed (e.g., test scope decisions, runtime environment confirmation),
-  the agent MUST use AskUserQuestion if the tool is available.
-- When AskUserQuestion supports structured choices (radio/multi-select),
-  the agent MUST prefer structured choices over free-text input.
-- If AskUserQuestion is technically unavailable, present the same question as a normal message
-  with explicit numbered choices.
-  The agent SHOULD preserve structured choice semantics (enumerated options, selection constraints).
-  The reason for unavailability MUST be stated.
+Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#user-questions-askuserquestion-protocol`.
+
+Skill-specific examples:
+
+- test scope decisions
+- runtime environment confirmation
 
 ## FORMAT SSOT (Mandatory)
 
-- Before writing or editing any `.qfai/**` artifact, read and follow the relevant directory README template and sample.
-- Do not copy templates/samples into this prompt or other prompt markdown.
-- Generated artifacts must match README-defined structure (headings, ordering, table columns).
-- Completion requires a format self-check in evidence.
+Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#format-ssot-mandatory`.
 
 ## Inputs Priority (Preflight)
 
@@ -78,36 +73,26 @@ When unsure, read inputs in this order:
 
 ## Sub-agent Delegation (MANDATORY)
 
-This section is mandatory and overrides any conflicting fallback text in this file.
+Follow `.qfai/assistant/instructions/shared-skill-delegation-baseline.md`.
 
 ### Orchestrator Protocol (MUST)
 
-- Orchestrator may only create work orders, delegate tasks, integrate outputs, and present results.
+- Follow the shared baseline.
+- Orchestrator MUST NOT self-approve.
 - Orchestrator MUST NOT generate the primary artifact first draft.
-- Orchestrator MUST NOT serve as Reviewer or skip delegation for convenience.
 
 ### Capability Probe (MUST)
 
-1. Run one harmless Probe Task (for example: "reply with ok") once at stage start.
-2. If subagents are unavailable, explicitly ask the user for Simulation mode approval.
-3. Without explicit approval, stop the stage and do not continue.
+- No additional overrides.
 
-### Simulation mode (Opt-in only)
+### Delegation Failure (Hard Stop)
 
-- Allowed only when the user explicitly states `Simulation mode allowed`.
-- When used, record both in outputs/evidence:
-  - `Subagents: simulated (reason: <why unavailable>)`
-  - `User approval: <quote or reference>`
+- No additional overrides.
+- Do not simulate roles. If the first required delegation fails, stop the stage and report remediation.
 
 ## Work Orders Summary
 
-Every major artifact in this stage MUST include this fixed table schema:
-
-| Step | Role (sub-agent) | Task title | Input (refs) | Output (refs) | Status (PASS/REVISE) |
-| ---- | ---------------- | ---------- | ------------ | ------------- | -------------------- |
-| 1    | <role>           | <task>     | <refs>       | <refs>        | PASS/REVISE          |
-
-- `Output (refs)` must point to in-file anchors or relative evidence file paths.
+Use the shared schema.
 
 ### Stage Minimum Roles (MUST)
 
@@ -120,77 +105,37 @@ Every major artifact in this stage MUST include this fixed table schema:
 
 ### Reviewer Gate (MUST)
 
+- Follow `.qfai/assistant/instructions/shared-skill-delegation-baseline.md#reviewer-gate-baseline`.
 - Final completion gate MUST be delegated to an independent `completion-reviewer`.
-- Reviewer checks (minimum):
-  - Required roles were delegated (no orchestrator self-authoring).
-  - Drift Protocol enforced (no upstream edits without approval and owner rerun).
-  - Test-layer policy enforced via `test-layers.md`.
-  - Coverage obligations met: E2E covers `US`, Integration covers `TC`, API covers `CON-API`.
-  - **Test-case quality depth verified**: Coverage Depth Matrix reviewed; no unjustified ❌ cells remain (see `references/test-case-depth-checklist.md`).
-  - Validation evidence exists and `qfai validate --fail-on error` passes.
-  - Floors/ratios are signals, not gates.
-  - `scenario.feature` and coverage ledgers are optional legacy inputs, not completion gates.
+- ATDD-specific reviewer checks:
+  - coverage obligations met: E2E covers `US`, Integration covers `TC`, API covers `CON-API`;
+  - Coverage Depth Matrix is reviewed and no unjustified `X` cells remain;
+  - validation evidence exists and `qfai validate --fail-on error` passes;
+  - `scenario.feature` and coverage ledgers remain optional legacy inputs, not completion gates.
 - Route specialist reviewers from `.qfai/assistant/steering/agent-routing.yml`.
 - Default ATDD review set:
   - `completion-reviewer`
   - `qa-gatekeeper`
 - Add `implementation-reviewer` only when helper/runtime support code changed.
 - Do not declare DONE until all routed blocking reviewers return `PASS`.
-- Every reviewer MUST provide a concrete alternative or fix proposal when returning FAIL.
 
 ### Work order template (copy/paste)
 
-```text
-Task title: <short>
-Role: <sub-agent role>
-Goal: <what to decide/produce>
-Inputs (refs):
-- <file/section>
-Constraints:
-- must: enforce Drift Protocol (no upstream edits without user approval + CR)
-- must: verify test-layer obligations from `steering/test-layers.md`
-- must: provide validation evidence (`qfai validate --fail-on error`)
-- must_not: treat volume ratios/floors as hard gates
-- must_not: accept upstream edits made directly by downstream phase
-Output format:
-- <headings / bullet schema>
-Quality bar:
-- PASS if ...
-- REVISE if ...
-```
+Use the shared template.
 
 ### Reviewer response template
 
-```text
-Result: PASS | REVISE
-Findings:
-- <issue>
-Required fixes:
-- <action>
-Evidence checked:
-- <refs>
-```
+Use the shared template.
 
-## Stage 0 — Steering completion refresh (mandatory)
-
-Before moving forward in this stage, refresh:
-
-- `.qfai/assistant/steering/manifest.md`
-- `.qfai/assistant/steering/product.md`
-- `.qfai/assistant/steering/structure.md`
-- `.qfai/assistant/steering/tech.md`
+- Required field: `Status (PASS/REVISE)`.
 
-Rules:
+## Stage 0 — Steering completion refresh (mandatory)
 
-- Detect incomplete content (empty sections, placeholder-only lines, `<...>`, `TBD`, stale facts).
-- Fill what is verifiable from repository evidence.
-- If something cannot be verified, record an Open Question and ask the user.
-- Update steering when new facts are discovered during this stage.
+Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#stage-0---steering-completion-refresh-mandatory`.
 
 ## Delta Rejected Guard (Mandatory)
 
-- Do not reintroduce options marked as rejected in 09_delta.md.
-- If a rejected option must be reconsidered, create a `[RE-OPEN]` Decision Record that references prior DR-ID and explicit approval.
+Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#delta-rejected-guard-mandatory`.
 
 ## CRITICAL CONSTRAINTS (Read First)
 
@@ -212,12 +157,7 @@ Rules:
 
 ## Completion Contract (Shared)
 
-Before declaring completion, you MUST:
-
-- Resolve or explicitly defer undefined/ambiguous items with rationale.
-- Verify every expected artifact exists and required sections are populated.
-- Scan generated artifacts for unresolved placeholders (`TBD`, `TODO`, `???`, etc.).
-- Run the smallest smoke check proving runnable behavior (or state "not applicable" with rationale).
+Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#completion-contract-shared`.
 
 ## Goal
 

package/assets/init/.qfai/assistant/skills/qfai-configure/SKILL.md

@@ -29,25 +29,22 @@ QFAI Skill Body (SSOT)
 
 ## User Questions (AskUserQuestion Protocol)
 
-- When a question to the user is needed (e.g., configuration decisions, glob pattern confirmation),
-  the agent MUST use AskUserQuestion if the tool is available.
-- When AskUserQuestion supports structured choices (radio/multi-select),
-  the agent MUST prefer structured choices over free-text input.
-- If AskUserQuestion is technically unavailable, present the same question as a normal message
-  with explicit numbered choices.
-  The agent SHOULD preserve structured choice semantics (enumerated options, selection constraints).
-  The reason for unavailability MUST be stated.
+Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#user-questions-askuserquestion-protocol`.
+
+Skill-specific examples:
+
+- configuration decisions
+- glob pattern confirmation
 
 ## FORMAT SSOT (Mandatory)
 
-- **Before writing or editing any `.qfai/**` artifact\*\*, read and follow the relevant directory README template and sample:
+- Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#format-ssot-mandatory`.
+
+- Before writing or editing any `.qfai/**` artifact, read and follow the relevant directory README template and sample:
   - `.qfai/discussion/README.md`
   - `.qfai/specs/README.md`
   - `.qfai/contracts/**/README.md`
   - `.qfai/evidence/README.md`
-- **Do NOT copy** templates/samples into this prompt or into other prompt markdown.
-- The generated artifacts must match the README-defined structure (headings, ordering, table columns).
-- Completion requires a **Format Self-Check** in the evidence: list each artifact and confirm “matches README template”.
 
 ## Inputs Priority (Preflight)
 
@@ -60,36 +57,24 @@ When unsure, read inputs in this order:
 
 ## Sub-agent Delegation (MANDATORY)
 
-This section is mandatory and overrides any conflicting fallback text in this file.
+Follow `.qfai/assistant/instructions/shared-skill-delegation-baseline.md`.
 
 ### Orchestrator Protocol (MUST)
 
-- Orchestrator may only create work orders, delegate tasks, integrate outputs, and present results to the user.
-- Orchestrator MUST NOT generate the primary artifact first draft.
-- Orchestrator MUST NOT serve as Reviewer or skip delegation for convenience.
+- No additional overrides.
 
 ### Capability Probe (MUST)
 
-1. Run one harmless Probe Task (for example: "reply with ok") once at stage start.
-2. If subagents are unavailable, explicitly ask the user for Simulation mode approval.
-3. Without explicit approval, stop the stage and do not continue.
+- No additional overrides.
 
-### Simulation mode (Opt-in only)
+### Delegation Failure (Hard Stop)
 
-- Allowed only when the user explicitly states `Simulation mode allowed`.
-- When used, record both of the following in outputs/evidence:
-  - `Subagents: simulated (reason: <why unavailable>)`
-  - `User approval: <quote or reference>`
+- No additional overrides.
+- Do not simulate roles. If the first required delegation fails, stop the stage and report remediation.
 
 ### Work Orders Summary (MANDATORY evidence)
 
-Every major artifact in this stage MUST include a `## Work Orders Summary` section with this fixed table schema:
-
-| Step | Role (sub-agent) | Task title | Input (refs) | Output (refs) | Status (PASS/REVISE) |
-| ---- | ---------------- | ---------- | ------------ | ------------- | -------------------- |
-| 1    | <role>           | <task>     | <refs>       | <refs>        | PASS/REVISE          |
-
-- `Output (refs)` must point to in-file anchors or relative evidence file paths.
+Use the shared schema.
 
 ### Stage Minimum Roles (MUST)
 
@@ -100,80 +85,37 @@ Every major artifact in this stage MUST include a `## Work Orders Summary` secti
 
 ### Reviewer Gate (MUST)
 
-- Final completion gate MUST be delegated to an independent `completion-reviewer`.
-- Reviewer checks (minimum):
-  - Required roles were delegated (no orchestrator self-authoring).
-  - DoD satisfied (validate gate, test-layer hard gate, evidence, DR-IDs).
-  - Validate gate evidence exists: `qfai validate --fail-on error` completed with `error=0`.
-  - **Drift Protocol enforced**:
-    - No upstream artifact edits were made without an explicit user-approved Change Request.
-    - If upstream changes exist, the correct owner skill was re-run after approval; downstream did not patch upstream directly.
-  - **Test-layer policy enforced**:
-    - E2E/API/Integration coverage aligns with `steering/test-layers.md` and the project’s plan.
-    - Do not use pyramid ratios as a gate; use floors/ratios only as signals. Coverage obligations are the gate.
+- Follow `.qfai/assistant/instructions/shared-skill-delegation-baseline.md#reviewer-gate-baseline`.
+- Reviewer checks:
+  - required roles were delegated;
+  - validate evidence exists: `qfai validate --fail-on error` completed with `error=0`;
+  - Drift Protocol enforced;
+  - test-layer policy enforced where applicable.
 - Route specialist reviewers from `.qfai/assistant/steering/agent-routing.yml`.
 - Default configure review set:
   - `completion-reviewer`
   - `qa-gatekeeper`
 - Do not declare DONE or handoff until all routed blocking reviewers return `PASS`.
-- Every reviewer MUST provide a concrete alternative or fix proposal when returning FAIL.
 
 ### Work order template (copy/paste)
 
-```text
-Task title: <short>
-Role: <sub-agent role>
-Goal: <what to decide/produce>
-Inputs (refs):
-- <file/section>
-Constraints:
-- must: enforce Drift Protocol (no upstream edits without user approval + CR)
-- must: verify plan/test-layer adherence (`steering/test-layers.md` + plan)
-- must: check `qfai validate --fail-on error` passes with evidence (`error=0`)
-- must: enforce `.qfai/assistant/steering/test-layers.md` hard gates
-- must_not: accept test-volume ratios/floors as a hard gate
-- must_not: accept upstream edits made directly by downstream phase
-Output format:
-- <headings / bullet schema>
-Quality bar:
-- PASS if ...
-- REVISE if ...
-```
+Use the shared template.
 
 ### Reviewer response template
 
-```text
-Result: PASS | REVISE
-Findings:
-- <issue>
-Required fixes:
-- <action>
-Evidence checked:
-- <refs>
-```
-
-## Stage 0 — Steering completion refresh (mandatory)
+Use the shared template.
 
-Before moving forward in this stage, refresh these files:
+- Required field: `Status (PASS/REVISE)`.
 
-- `.qfai/assistant/steering/manifest.md`
-- `.qfai/assistant/steering/product.md`
-- `.qfai/assistant/steering/structure.md`
-- `.qfai/assistant/steering/tech.md`
+## Stage 0 — Steering completion refresh (mandatory)
 
-Rules:
+Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#stage-0---steering-completion-refresh-mandatory`.
 
-- Detect incomplete content (empty sections, placeholder-only lines, `<...>`, `TBD`, stale facts).
-- Fill what is verifiable from repository evidence (tree, docs, require/spec artifacts, package.json, CI definitions).
-- If something cannot be verified, record it as an Open Question and ask the user.
-- Even if steering is already complete, update it when new facts are discovered in this stage.
+- Fill steering from verifiable repository evidence first; when evidence is missing, mark the field `TBD` and record the gap in the evidence file.
 
 ## Delta Rejected Guard (Mandatory)
 
-- Do NOT reintroduce options marked as rejected in 09_delta.md.
-- If a rejected option must be reconsidered, create a **[RE-OPEN]** Decision
-  Record in 09_delta.md that references the prior DR-ID, states what changed +
-  new criteria, and includes explicit approval (user or instructions/steering).
+Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#delta-rejected-guard-mandatory`.
 
 ## CRITICAL CONSTRAINTS (Read First)
 
@@ -187,19 +129,7 @@ Rules:
 
 ## Completion Contract (Shared)
 
-Before declaring completion, you MUST:
-
-- OQ / undefined resolution: detect undefined or ambiguous items; resolve them or explicitly defer them with documented rationale and (when required by this prompt) user approval.
-- Deliverable completeness: verify every expected artifact listed in this prompt (and required README templates) exists and is fully populated; no missing required sections.
-- OQ / placeholder scan: scan all generated artifacts (including evidence) for
-  placeholders such as "TBD", "TODO", "TBA", "TBC", "XXX", "???", "OQ",
-  "OPEN QUESTION", "UNDEFINED", "PLACEHOLDER", and localized equivalents in
-  the user's language. Resolve or explicitly defer; do not leave silent
-  placeholders.
-- Smoke check (if applicable): when the prompt produces runnable code, tests,
-  or configs, execute the smallest command that proves basic run/start/operate
-  and record evidence. If not applicable, state "not applicable" with a short
-  rationale.
+Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#completion-contract-shared`.
 
 ## Goal
 
@@ -327,29 +257,35 @@ Do not edit any `.qfai/**/README.md` file; raise an Open Question instead.
 
 ## Multi-Role Orchestration (Subagents)
 
-This workflow assumes the environment _may_ support subagents (e.g., Claude Code "Task" tool) or may not.
+Use the platform's native sub-agent delegation mechanism for Claude Code, GitHub Copilot, and Codex.
 
-### If subagents are supported
+### Delegation order
 
-Delegate to multiple roles and then merge the results. Use a "real-world workflow" order:
+Use `.qfai/assistant/steering/agent-routing.yml` as the routing SSOT.
 
-- discovery-analyst -> requirements-analyst -> delivery-planner -> solution-architect -> test-design-analyst -> qa-strategist -> devops-ci-engineer -> completion-reviewer / qa-gatekeeper
+- First required delegation / Capability Probe: `delivery-planner` in the `analysis` phase.
+- Then follow routed phases in order: `analysis` (`delivery-planner`, `qa-strategist`) -> `config` (`devops-ci-engineer`) -> `review` (`completion-reviewer`, `qa-gatekeeper`).
+- Do not prepend non-routed roles before the first required delegation attempt.
 
-**Pseudo-invocation pattern** (adjust to your tool):
+### Delegation contract (tool-neutral)
 
 ```text
-Task(
-  subagent_type="planner",
-  description="Analyze repo and propose testFileGlobs",
-  prompt="Context: ...\nGoal: Tune qfai.config.yaml\nConstraints: minimal diff\nReturn: globs + evidence"
-)
+Role: delivery-planner
+Task title: Analyze repo and propose testFileGlobs
+Goal: Tune qfai.config.yaml with a minimal diff
+Inputs:
+- relevant steering and repo layout
+Constraints:
+- minimal diff
+- evidence-first
+Return:
+- proposed globs + rationale + evidence refs
 ```
 
-### If subagents are NOT supported
-
-Only with explicit user approval (`Simulation mode allowed`), simulate roles by running the same sequence yourself:
+### Failure rule
 
-- Write a short "role output" section per role, then consolidate into the final deliverable(s).
+- The first required delegation attempt doubles as the capability check.
+- If that delegation fails, stop immediately. Do not simulate roles or continue with self-execution.
 
 ## Completion Separation (mandatory)