qfai 1.7.9 → 1.7.13

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

@@ -5,19 +5,16 @@ description: "Implement automated acceptance tests (E2E/API/Integration) aligned
 argument-hint: "<spec-id> [--auto]"
 allowed-tools: [Read, Glob, Write, TodoWrite, Task, Bash]
 roles:
-  - Orchestrator
-  - TestVolumeEstimator
-  - ATDDE2EImplementer
-  - ATDDAPIImplementer
-  - ATDDIntegrationImplementer
-  - QAEngineer
-  - TestEngineer
-  - BackendEngineer
-  - FrontendEngineer
-  - Reviewer
-  - RuntimeGatekeeper
-  - DevOpsCIEngineer
-  - CodeReviewer
+  - orchestrator
+  - delivery-planner
+  - test-design-analyst
+  - qa-strategist
+  - acceptance-test-engineer
+  - devops-ci-engineer
+  - completion-reviewer
+  - qa-gatekeeper
+  - implementation-reviewer
+routing-profile: runtime-heavy
 mode: execution-focused
 ---
 
@@ -114,31 +111,32 @@ Every major artifact in this stage MUST include this fixed table schema:
 
 ### Stage Minimum Roles (MUST)
 
-- Delegate: TestVolumeEstimator, ATDDE2EImplementer, ATDDAPIImplementer, ATDDIntegrationImplementer.
-- Integrate: Orchestrator consolidates delegated outputs and presents results.
-- Gate: Reviewer is delegated independently and returns only `PASS` or `REVISE`.
+- Delegate: `test-design-analyst` defines coverage and layer ownership.
+- Delegate: `acceptance-test-engineer` implements E2E, API, and integration acceptance tests.
+- Delegate: `devops-ci-engineer` captures execution evidence when CI/runtime proof is needed.
+- Integrate: `orchestrator` consolidates delegated outputs and presents results.
+- Gate: `completion-reviewer` is delegated independently and returns only `PASS` or `REVISE`.
 - Orchestrator must not draft the primary artifact body and must not self-approve.
 
 ### Reviewer Gate (MUST)
 
-- Final completion gate MUST be delegated to an independent Reviewer sub-agent.
+- 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.
-- Do not declare DONE until Reviewer returns `PASS`.
-- **All reviewers: alternative proposal obligation**:
-  - Every reviewer MUST provide a concrete alternative or fix proposal when returning FAIL. Feedback without a concrete alternative is invalid and triggers re-judgment.
-- **devils-advocate gate**:
-  - devils-advocate FAIL must include a concrete alternative proposal. Bare negation FAIL triggers re-judgment.
-  - 3 consecutive FAILs trigger advisory demotion and allow progression to the next phase.
-- **pattern-doubler gate**:
-  - Each pattern proposed by pattern-doubler must include rationale.
-  - Artifacts with no ID-bearing items (US/AC/BR/EX/TC) are marked N/A.
+- 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)
 
@@ -238,10 +236,11 @@ Turn specs/contracts obligations (`US` / `TC` / `CON-API`) into runnable accepta
 ## Mandatory Outputs
 
 1. Test Volume Estimate (signal table with evidence)
-2. Coverage obligations checklist (`US` / `TC` / `CON-API`)
-3. Implemented tests per layer (E2E/API/Integration)
-4. Reviewer notes (`PASS` or concrete rework list)
-5. Evidence file: `.qfai/evidence/atdd-<spec-id>.md`
+2. **Coverage Depth Matrix** (per spec; see `references/test-case-depth-checklist.md`)
+3. Coverage obligations checklist (`US` / `TC` / `CON-API`)
+4. Implemented tests per layer (E2E/API/Integration)
+5. Reviewer notes (`PASS` or concrete rework list)
+6. Evidence file: `.qfai/evidence/atdd-<spec-id>.md`
 
 ## Volume Signals (mandatory, not gates)
 
@@ -287,6 +286,7 @@ Notes:
 - Forbidden references remain.
 - Tests exist but were never executed.
 - Validation evidence is missing or failing.
+- Coverage Depth Matrix is missing or contains unjustified ❌ cells (normal-path-only coverage is incomplete).
 
 ## Failure handling (mandatory)
 
@@ -341,11 +341,12 @@ Template:
 
 ## ATDD Work Orders (mandatory)
 
+- **Test Case Depth Analyst**: `test-design-analyst` evaluates test cases using `references/test-case-depth-checklist.md`, produces Coverage Depth Matrix, flags gaps in boundary/error/edge coverage.
 - Test Volume Estimator: compute US/TC/CON signals with evidence.
 - ATDD E2E Implementer: implement required `US` coverage.
 - ATDD API Implementer: implement required `CON-API` coverage.
 - ATDD Integration Implementer: implement required `TC` coverage.
-- Reviewer: validate coverage obligations + gate results (non-edit).
+- Reviewer: validate coverage obligations + gate results + Coverage Depth Matrix (non-edit).
 - Runtime Gatekeeper: run suites and capture logs.
 
 ## Completion Separation (mandatory)

package/assets/init/.qfai/assistant/skills/qfai-atdd/references/test-case-depth-checklist.md

@@ -0,0 +1,80 @@
+# Test Case Depth Checklist
+
+This checklist ensures test cases cover not only happy paths but also boundary values, error paths, edge cases, and combinatorial scenarios.
+Reviewers MUST use this checklist when evaluating test case completeness during ATDD and SDD review gates.
+
+## 1. Equivalence Partitioning (同値分割)
+
+For each input parameter or condition:
+
+- [ ] Valid partitions identified with at least one representative test case each.
+- [ ] Invalid partitions identified with at least one representative test case each.
+- [ ] Special value partitions identified (null, empty, zero, default) with test cases.
+
+## 2. Boundary Value Analysis (境界値分析)
+
+For each numeric, date, string-length, or ordered domain:
+
+- [ ] Minimum valid value tested.
+- [ ] Maximum valid value tested.
+- [ ] Just below minimum (invalid) tested.
+- [ ] Just above maximum (invalid) tested.
+- [ ] Off-by-one boundaries tested where applicable.
+
+## 3. Normal / Error / Edge Path Coverage (正常系・異常系・エッジケース)
+
+For each US or TC:
+
+- [ ] At least one normal (happy) path test case exists.
+- [ ] At least one error/failure path test case exists (invalid input, missing data, unauthorized access, etc.).
+- [ ] Edge cases identified and tested (concurrent access, timing, empty collections, maximum payload, etc.).
+
+## 4. Special Values (特殊値)
+
+- [ ] Null / undefined / missing values handled.
+- [ ] Empty strings, empty arrays, empty objects handled.
+- [ ] Maximum-length strings or maximum-size payloads handled.
+- [ ] Special characters (Unicode, control characters, SQL injection patterns) considered where applicable.
+
+## 5. State Transitions (状態遷移)
+
+For business flows with state machines or multi-step processes:
+
+- [ ] All valid state transitions have test cases.
+- [ ] Invalid state transitions are tested and rejected.
+- [ ] Terminal / end states are reachable and verified.
+
+## 6. Combinatorial Coverage (組み合わせ)
+
+When multiple conditions interact:
+
+- [ ] Key condition combinations tested (at minimum pairwise for high-risk interactions).
+- [ ] Conflicting or contradictory input combinations tested.
+
+## 7. Business Rule Coverage (ビジネスルール網羅)
+
+- [ ] Every BR-\* referenced in 04_Business-Rules.md has at least one positive and one negative test case.
+- [ ] Conditional business rules have test cases for each branch.
+
+---
+
+## Coverage Depth Matrix (テンプレート)
+
+Reviewers and test-design-analysts MUST produce this matrix for each spec under review.
+Mark each cell: ✅ covered, ⚠️ partial, ❌ missing.
+
+| US/TC ID | Normal path | Error path | Boundary values | Special values | State transitions | Combinatorial | Status |
+| -------- | ----------- | ---------- | --------------- | -------------- | ----------------- | ------------- | ------ |
+| US-0001  | ✅/⚠️/❌    | ✅/⚠️/❌   | ✅/⚠️/❌        | ✅/⚠️/❌       | ✅/⚠️/❌          | ✅/⚠️/❌      | —      |
+| TC-0001  | ✅/⚠️/❌    | ✅/⚠️/❌   | ✅/⚠️/❌        | ✅/⚠️/❌       | ✅/⚠️/❌          | ✅/⚠️/❌      | —      |
+
+### Evaluation criteria
+
+- **PASS**: All cells are ✅ or ⚠️ with documented rationale for partial coverage.
+- **REVISE**: Any cell is ❌ without an explicit justification (e.g., Decision Record).
+
+### Usage
+
+- `test-design-analyst`: Produce this matrix when defining coverage obligations. Flag ❌ cells as gaps.
+- `qa-gatekeeper`: Verify the matrix exists and no unjustified ❌ cells remain.
+- `completion-reviewer`: Confirm the matrix was reviewed and any ⚠️ cells have rationale.

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

@@ -4,7 +4,16 @@ title: QFAI Configure (Tune qfai.config.yaml)
 description: "Analyze the repository and tune qfai.config.yaml (testFileGlobs, exclude globs, optional specSections)."
 argument-hint: "[--auto]"
 allowed-tools: [Read, Glob, Write, TodoWrite, Task]
-roles: [DevOpsCIEngineer, QAEngineer, CodeReviewer, Planner]
+roles:
+  [
+    orchestrator,
+    delivery-planner,
+    qa-strategist,
+    devops-ci-engineer,
+    completion-reviewer,
+    qa-gatekeeper,
+  ]
+routing-profile: runtime-heavy
 mode: evidence-focused
 ---
 
@@ -32,7 +41,7 @@ QFAI Skill Body (SSOT)
 ## FORMAT SSOT (Mandatory)
 
 - **Before writing or editing any `.qfai/**` artifact\*\*, read and follow the relevant directory README template and sample:
-  - `.qfai/require/README.md`
+  - `.qfai/discussion/README.md`
   - `.qfai/specs/README.md`
   - `.qfai/contracts/**/README.md`
   - `.qfai/evidence/README.md`
@@ -91,7 +100,7 @@ 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 Reviewer sub-agent.
+- 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).
@@ -102,15 +111,12 @@ Every major artifact in this stage MUST include a `## Work Orders Summary` secti
   - **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.
-- Do not declare DONE or handoff until Reviewer returns `PASS`.
-- **All reviewers: alternative proposal obligation**:
-  - Every reviewer MUST provide a concrete alternative or fix proposal when returning FAIL. Feedback without a concrete alternative is invalid and triggers re-judgment.
-- **devils-advocate gate**:
-  - devils-advocate FAIL must include a concrete alternative proposal. Bare negation FAIL triggers re-judgment.
-  - 3 consecutive FAILs trigger advisory demotion and allow progression to the next phase.
-- **pattern-doubler gate**:
-  - Each pattern proposed by pattern-doubler must include rationale.
-  - Artifacts with no ID-bearing items (US/AC/BR/EX/TC) are marked N/A.
+- 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)
 
@@ -327,7 +333,7 @@ This workflow assumes the environment _may_ support subagents (e.g., Claude Code
 
 Delegate to multiple roles and then merge the results. Use a "real-world workflow" order:
 
-- Facilitator -> Interviewer -> Requirements Analyst -> Planner -> Architect -> (Contract Designer) -> Test Engineer -> QA Engineer -> Code Reviewer -> DevOps/CI Engineer
+- discovery-analyst -> requirements-analyst -> delivery-planner -> solution-architect -> test-design-analyst -> qa-strategist -> devops-ci-engineer -> completion-reviewer / qa-gatekeeper
 
 **Pseudo-invocation pattern** (adjust to your tool):
 
@@ -347,8 +353,8 @@ Only with explicit user approval (`Simulation mode allowed`), simulate roles by
 
 ## Completion Separation (mandatory)
 
-- Config changes (DevOpsCIEngineer) and completion approval (CodeReviewer) must be separate.
-- QAEngineer must confirm evidence sampling before approval.
+- Config changes (`devops-ci-engineer`) and completion approval (`completion-reviewer`) must be separate.
+- `qa-gatekeeper` must confirm evidence sampling before approval.
 
 ## Context Refresh (mandatory for long tasks)
 

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

@@ -1,10 +1,23 @@
 ---
 name: qfai-discussion
 title: QFAI Discussion (Unified Discuss + Require)
-description: "Run structured discussion that merges discuss and require into a single 15-file discussion pack with OQ-driven exit."
+description: "Run structured discussion that merges discuss and require into a single 15-file discussion pack with required prototyping.yaml and OQ-driven exit."
 argument-hint: "<idea-or-problem> [--auto]"
 allowed-tools: [Read, Glob, Write, TodoWrite, Task, Bash]
-roles: [Researcher, Facilitator, Interviewer, RequirementsAnalyst, QAEngineer, Planner]
+roles:
+  [
+    orchestrator,
+    delivery-planner,
+    discovery-analyst,
+    requirements-analyst,
+    solution-architect,
+    product-experience-architect,
+    completion-reviewer,
+    requirements-reviewer,
+    architecture-reviewer,
+    product-surface-reviewer,
+  ]
+routing-profile: requirements-heavy
 mode: interactive-by-default
 ---
 
@@ -35,7 +48,9 @@ QFAI Skill Body (SSOT)
   - `.qfai/discussion/README.md`
   - `.qfai/specs/README.md`
   - `.qfai/evidence/README.md`
-  - `.qfai/assistant/steering/review-roster.yml`
+  - `.qfai/assistant/steering/agent-catalog.yml`
+  - `.qfai/assistant/steering/agent-routing.yml`
+  - `.qfai/assistant/steering/review-profiles.yml`
   - `.qfai/assistant/skills/qfai-discussion/references/rcp_footer.md`
 - Keep templates as source of truth and preserve file naming/order.
 
@@ -70,18 +85,18 @@ Every major artifact in this stage MUST include this table schema:
 
 ### Reviewer Gate (MUST)
 
-- Delegate final completion gate to an independent Reviewer.
-- Reviewer must check Drift Protocol compliance and alignment with `.qfai/assistant/steering/test-layers.md`.
+- Delegate final completion gate to an independent `completion-reviewer`.
+- Route specialist reviewers from `.qfai/assistant/steering/agent-routing.yml` and `.qfai/assistant/steering/review-profiles.yml`.
+- Default discussion review set:
+  - `completion-reviewer`
+  - `requirements-reviewer`
+- Conditional discussion reviewers:
+  - `architecture-reviewer` when architecture-affecting decisions exist
+  - `product-surface-reviewer` when the pack is UI-bearing
+- Reviewers must check Drift Protocol compliance and alignment with `.qfai/assistant/steering/test-layers.md`.
 - Test volume floors/ratios are not gates; they are risk signals.
-- Do not declare DONE until Reviewer returns `PASS`; otherwise apply `REVISE`.
-- **All reviewers: alternative proposal obligation**:
-  - Every reviewer MUST provide a concrete alternative or fix proposal when returning FAIL. Feedback without a concrete alternative is invalid and triggers re-judgment.
-- **devils-advocate gate**:
-  - devils-advocate FAIL must include a concrete alternative proposal. Bare negation FAIL triggers re-judgment.
-  - 3 consecutive FAILs trigger advisory demotion and allow progression to the next phase.
-- **pattern-doubler gate**:
-  - Each pattern proposed by pattern-doubler must include rationale.
-  - Artifacts with no ID-bearing items (US/AC/BR/EX/TC) are marked N/A.
+- Do not declare DONE until all routed blocking reviewers return `PASS`; otherwise apply `REVISE`.
+- Every reviewer MUST provide a concrete alternative or fix proposal when returning `FAIL`.
 
 ## CRITICAL CONSTRAINTS (Read First)
 
@@ -107,30 +122,35 @@ Every major artifact in this stage MUST include this table schema:
 - `deferred` is allowed only when required metadata is complete in `13_Deferred.md`.
 - `02_Inception-Deck.md` MUST contain at least one Mermaid diagram in ` ```mermaid ` fences.
 - `03_Story-Workshop.md` MUST contain at least one Mermaid diagram in ` ```mermaid ` fences.
-- If UI requirements exist, include an HTML+CSS visual mock in `03_Story-Workshop.md`.
+- If UI requirements exist, behavior obligations and sidecar artifacts are primary; HTML+CSS visual mock is optional fallback only.
 - **UI-bearing Authoring Requirements**:
   - UI-bearing detection is based on **surface type classification**
     (see `## UI-bearing Detection` below).
   - Content signals in `03_Story-Workshop.md`
     (HTML tags, Mermaid screen flows) serve as supplementary detection hints,
     not the primary SSOT.
-  - UI-bearing packs MUST include a `## Design Direction Summary` section in `03_Story-Workshop.md` with all 6 subsections:
-    1. `### Option Comparison` — 2+ distinct design options (QFAI-DDP-020)
-    2. `### Anchor Screen Selection` — explicit selection referencing a compared option (QFAI-DDP-021)
-    3. `### Competitive References` — summary referencing 04_Sources.md (QFAI-DDP-022 validates 04_Sources.md fields)
-    4. `### CTA Hierarchy` — must define at least a primary CTA (QFAI-DDP-023)
-    5. `### State Coverage` — must define empty, loading, error, populated states (QFAI-DDP-024)
-    6. `### Design Anti-goals` — 1+ patterns to intentionally avoid (QFAI-DDP-025)
-  - `04_Sources.md` must include a `## Competitive Reference Registry` with entries containing:
+  - UI-bearing packs rely on the **canonical sidecar family** as primary truth for design direction:
+    - `uiux/10_implementation_strategy.md` — implementation strategy with strict canonical schema
+    - `uiux/11_design_taste_interview.md` — design taste interview (10 sections)
+    - `04_Sources.md#Trend Scan` — trend scan with required category coverage and per-entry completeness
+    - `uiux/20-24` — 3-layer evaluation family (invariant, trend-derived, product-specific, aggregate, dynamic overrides)
+    - `uiux/30_option_comparison.md` — option comparison
+    - `uiux/31_selected_anchor_screen.md` — selected anchor screen (Selected Direction single source of truth)
+    - `uiux/40_screen_contracts.md` — screen contracts (strong schema)
+    - `uiux/50_review_input_bundle.md` — review input bundle
+  - `04_Sources.md` `## Trend Scan` must include all required categories, and each entry must include: `reference`, `observation`, `freshness_date`, `confidence`, `source_translation`, `local_implication`
+  - `04_Sources.md` `## Competitive Reference Registry` entries must include:
     - `adopted_points`: what was adopted and why
     - `rejected_points`: what was not adopted and why
     - `local_translation`: how adopted points were adapted
-    - Placeholder values (TBD, N/A, TODO, empty) are treated as missing (QFAI-DDP-022)
-  - `14_Review-Request.md` must include a `## Design Direction Decisions` section with anchor, rejections, and adopted refs.
+    - Placeholder-like values (TBD, N/A, TODO, empty) are treated as missing
+  - `14_Review-Request.md` must review selected direction from `uiux/31_selected_anchor_screen.md` and strategy alignment from `uiux/10_implementation_strategy.md`.
   - `99_delta.md` must include a `## Rejected Visual Directions` section with rationale and recurrence prevention.
-  - All 7 validators (QFAI-DDP-019..025) emit `severity: error` — violations block validation.
-  - Non-UI packs are exempt from all DDS validators (zero new issues).
-- Review roster is fixed by `.qfai/assistant/steering/review-roster.yml` and must be executed in full.
+  - `04_Sources.md` must include a `## Trend Scan` section where each required category has at least one complete entry.
+  - `04_Sources.md` must include a `## Competitive Reference Registry` where each entry has: `adopted_points`, `rejected_points`, `local_translation` fields populated.
+  - Sidecar-family validators (UIX-VAL series) are the primary quality gates for UI-bearing packs.
+  - Non-UI packs are exempt from all sidecar validators (zero new issues).
+- Reviewer routing is derived from `.qfai/assistant/steering/agent-routing.yml` and `.qfai/assistant/steering/review-profiles.yml`.
 - RCP wording must be sourced from `.qfai/assistant/skills/qfai-discussion/references/rcp_footer.md`.
 - Discussion artifacts are logs/rationale and must not duplicate spec SSOT.
 - If diagrams are written, Mermaid syntax must be in ` ```mermaid ` fences only.
@@ -148,11 +168,12 @@ Classification is based on **surface type only**, not interaction complexity (DR
 
 | Surface Type | UI-bearing | Sidecar Generation                       | Example                                  |
 | ------------ | ---------- | ---------------------------------------- | ---------------------------------------- |
-| web-ui       | Yes        | Full 11-file uiux/ sidecar               | Web application with user-facing screens |
-| mobile-ui    | Yes        | Full 11-file uiux/ sidecar               | Mobile app with touch interactions       |
-| desktop-ui   | Yes        | Full 11-file uiux/ sidecar               | Desktop application with GUI             |
+| web          | Yes        | Full 11-file uiux/ sidecar               | Web application with user-facing screens |
+| mobile       | Yes        | Full 11-file uiux/ sidecar               | Mobile app with touch interactions       |
+| desktop      | Yes        | Full 11-file uiux/ sidecar               | Desktop application with GUI             |
+| cli          | No         | No uiux/ directory, no sidecar generated | CLI tool, terminal application           |
 | mixed        | Yes        | Full 11-file uiux/ sidecar               | Cross-platform with UI components        |
-| non-ui       | No         | No uiux/ directory, no sidecar generated | CLI tool, API service, library           |
+| non-ui       | No         | No uiux/ directory, no sidecar generated | API service, library                     |
 
 ### Detection Signals
 
@@ -171,24 +192,27 @@ When UI-bearing is detected:
 When non-ui is detected:
 
 - Skip uiux/ sidecar generation entirely — no uiux/ directory, no errors
-- Core 15-file pack is generated as before
+- Core 15-file pack plus required prototyping.yaml is generated as before
 - No additional UI/UX completion conditions apply
 
 ### UI-bearing Completion Conditions
 
 For UI-bearing projects, the following conditions must ALL be satisfied before discussion completion:
 
-1. **Strategy selected**: `uiux/10_strategy.md` is populated with a chosen implementation approach
-2. **Scoring axes defined**:
-   `uiux/20_eval_axis_usability.md`,
-   `uiux/21_eval_axis_consistency.md`,
-   `uiux/22_eval_axis_accessibility.md`,
-   `uiux/23_eval_axis_delight.md`
-   have evaluation criteria and measurement approaches
-3. **Anchor screen chosen**: `uiux/31_anchor.md` documents the selected anchor screen with rationale
-4. **Contracts drafted**: `uiux/40_contracts.md` contains interaction contracts for the anchor screen
-
-Completion is blocked until all 4 conditions are met. Skipping any condition prevents the discussion from being marked as complete.
+1. **Strategy selected**: `uiux/10_implementation_strategy.md` is populated with a canonical implementation approach
+2. **Taste interview completed**: `uiux/11_design_taste_interview.md` has all 10 sections populated
+3. **Scoring axes defined (3-layer)**:
+   `uiux/20_design_eval_invariant.md`,
+   `uiux/21_design_eval_trend_derived.md`,
+   `uiux/22_design_eval_product_specific.md`,
+   `uiux/23_design_eval_aggregate.md`
+   have invariant, trend-derived, and product-specific evaluation criteria
+4. **Dynamic overrides documented**: `uiux/24_design_eval_dynamic_overrides.md` lists any override rules
+5. **Comparison completed**: `uiux/30_option_comparison.md` documents option comparison against scoring axes
+6. **Anchor screen selected**: `uiux/31_selected_anchor_screen.md` documents the selected direction and anchor screen
+7. **Contracts drafted**: `uiux/40_screen_contracts.md` contains screen interaction contracts
+
+Completion is blocked until all 7 conditions are met. Skipping any condition prevents the discussion from being marked as complete.
 
 ### Non-UI Completion
 
@@ -196,7 +220,7 @@ For non-ui projects, completion conditions remain unchanged from prior versions.
 
 ## Goal
 
-Produce a unified 15-file discussion pack with explicit decisions, requirements, OQ states, and rationale so `/qfai-sdd` starts without unresolved blockers.
+Produce a unified 15-file discussion pack plus required prototyping.yaml with explicit decisions, requirements, OQ states, and rationale so `/qfai-sdd` starts without unresolved blockers.
 
 ## Non-goals
 
@@ -222,6 +246,7 @@ Produce a unified 15-file discussion pack with explicit decisions, requirements,
 - `.qfai/discussion/discussion-*/14_Review-Request.md`
 - `.qfai/discussion/discussion-*/99_delta.md`
 - review artifacts under `.qfai/review/review-YYYYMMDDhhmmssSSS/`
+- `.qfai/discussion/discussion-*/prototyping.yaml`
 - Evidence file: `.qfai/evidence/discussion-YYYYMMDDhhmmssSSS.md`
 - Reviewer notes (`PASS` or `REVISE`)
 
@@ -229,7 +254,7 @@ Produce a unified 15-file discussion pack with explicit decisions, requirements,
 
 1. Run the core interview for product concept, scope, stakeholders, and constraints (`01_Context.md`).
 2. Run Inception Deck (10 questions) for ambiguity removal and project alignment, and include at least one Mermaid diagram (`02_Inception-Deck.md`).
-3. Run Story Workshop to capture user stories, user flows, and at least one Mermaid diagram; add HTML+CSS screen mock when UI requirements exist (`03_Story-Workshop.md`).
+3. Run Story Workshop to capture user stories, user flows, and at least one Mermaid diagram; HTML+CSS visual mock is optional fallback only when it materially clarifies the selected direction (`03_Story-Workshop.md`).
 4. Register source traceability in `04_Sources.md` with stable `SRC-XXXX` identifiers.
 5. Define scope boundaries and success criteria in `05_Scope.md`.
 6. Capture functional requirements in `06_REQ.md` with `REQ-0001` format.
@@ -242,7 +267,17 @@ Produce a unified 15-file discussion pack with explicit decisions, requirements,
 13. Run OQ resolution hearing repeatedly until open count is zero.
 14. Move deferred items to `13_Deferred.md` with all mandatory metadata columns.
 15. Update `12_OQ-Resolution-Log.md`, `14_Review-Request.md`, and `99_delta.md`.
-16. Request review and record Reviewer result.
+16. Generate `prototyping.yaml` at the discussion pack top level for downstream prototyping recommendation.
+    - MUST use the **namespaced canonical schema** with `prototyping.recommended_mode` (not top-level `recommended_mode`).
+    - Top-level `recommended_mode` is legacy compatibility only and MUST NOT be emitted in new artifacts.
+    - All fields (`recommended_mode`, `rationale`, `allowed_modes`, `surface`) are required under the `prototyping:` key.
+17. Choose `recommended_mode` with these defaults:
+
+- `low-cost`: rough draft or strong cost priority
+- `standard`: normal customer-presentable path
+- `full-harness`: only when the premium runtime loop is justified
+
+18. Request review and record Reviewer result.
 
 ## Example Mapping Perspectives (Mandatory)
 
@@ -298,7 +333,7 @@ Rules:
 At any point during discussion, if the user changes direction or scope:
 
 1. Record the drift event in `99_delta.md` with Change Type = `Drift`.
-2. Assess impact on all 15 files.
+2. Assess impact on all 15 files plus prototyping.yaml.
 3. Update affected files and re-validate OQ register exit condition.
 4. If drift contradicts a previously rejected option, record in `99_delta.md` Rejected section with `Recurrence Prevention`.
 
@@ -314,22 +349,15 @@ RCP rules:
 
 - Append-only: create a new review pack for each cycle.
 - Apply `.qfai/assistant/skills/qfai-discussion/references/rcp_footer.md` as the common footer rule set.
-- Any `FAIL` requires return/fix/full rerun from the first reviewer.
-- Mark fixed only when all reviewers are `PASS` or valid `N/A`.
+- Run only the reviewers routed by `agent-routing.yml` for the current phase and conditions.
+- On `FAIL`, rerun only the failed reviewer and any reviewers whose scope changed because of the fix.
+- Mark fixed only when all routed blocking reviewers are `PASS`.
 - `summary.json` `target.kind` must be `"discussion"`.
-- Execution order: existing 10 reviewers (1-10) → devils-advocate (11) → pattern-doubler (12).
-- devils-advocate (11th):
-  - `can_be_na: false` — N/A is not allowed.
-  - FAIL must include a concrete alternative. Bare negation FAIL is invalid.
-  - 3 consecutive FAILs trigger advisory demotion (current review cycle only).
-- pattern-doubler (12th):
-  - `can_be_na: true` — N/A is default in discussion phase as ID-bearing items are sparse.
-  - Evaluates Example Seeds count and perspective coverage.
 
 ## RCP Footer Include (MUST)
 
 - Include and follow `.qfai/assistant/skills/qfai-discussion/references/rcp_footer.md` without rewriting it per skill.
-- Roster and loop rules must stay synchronized with the footer SSOT.
+- Routing and rerun rules must stay synchronized with `agent-routing.yml` and the footer SSOT.
 
 ## Required Coverage Topics
 
@@ -356,7 +384,13 @@ Before declaring completion, you MUST:
 - ensure `02_Inception-Deck.md` includes at least one Mermaid diagram;
 - ensure `Example Seeds` sections are present and perspective coverage is explicit in `03_Story-Workshop.md`;
 - ensure `03_Story-Workshop.md` includes at least one Mermaid diagram;
-- ensure UI-related stories include an HTML+CSS screen mock section in `03_Story-Workshop.md`;
+- ensure UI-related stories include behavior obligations in `03_Story-Workshop.md` (HTML+CSS mock is optional fallback);
+- ensure `uiux/11_design_taste_interview.md` is complete (all 10 sections) when UI-bearing;
+- ensure `04_Sources.md` includes a `## Trend Scan` section with freshness metadata when UI-bearing;
+- ensure 3-layer evaluation family files (20-24) are populated when UI-bearing;
+- ensure `uiux/10_implementation_strategy.md` uses the strict canonical schema when UI-bearing;
+- ensure `uiux/40_screen_contracts.md` uses the strong screen contract schema when UI-bearing;
+- ensure `uiux/50_review_input_bundle.md` is review-ready when UI-bearing;
 - avoid duplicating finalized spec content in discussion outputs.
 
 ## Evidence (MANDATORY)
@@ -398,7 +432,7 @@ When done, report:
 - [ ] Deferred items include required metadata.
 - [ ] `02_Inception-Deck.md` includes at least one Mermaid diagram.
 - [ ] `03_Story-Workshop.md` includes at least one Mermaid diagram.
-- [ ] UI-related stories include HTML+CSS screen mock details in `03_Story-Workshop.md`.
+- [ ] UI-related stories include behavior obligations in `03_Story-Workshop.md` (HTML+CSS mock is optional fallback only).
 - [ ] Mermaid fence rules were satisfied when diagrams were used.
 - [ ] Evidence file exists and includes Work Orders Summary + Reviewer result.
 - [ ] Reviewer returned `PASS`.