qfai 1.8.1 → 1.8.3

package/README.md

@@ -52,13 +52,16 @@ npx qfai report
     (`.qfai/review/review-*/summary.json` + minimum schema), writes `.qfai/report/validate.json`,
     and appends run logs to `.qfai/report/run-*/`; use `--fail-on error` (or `--fail-on warning`) to turn it into a CI gate,
     and `--format github` to emit GitHub-friendly annotations.
-    Use `--phase refinement` only for local refinement checks; CI should use default/full validation.
+    Use `--profile discussion|sdd|prototyping|atdd|tdd|verify` for local skill-owned checks; CI should use default/full validation (or `verify` / `tdd` for the dedicated CI gates).
 - `npx qfai report`
   - Produces a human-readable report (`report.md` by default) or an internal JSON export (`report.json`) from `validate.json`; use `--base-url` to link file paths in Markdown to your repository viewer.
 - `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=full-harness` for supported UI surfaces only), not by a general-purpose end-user CLI flow.
+    (`/qfai-prototyping` — any mode; modes differ only in `maxCycles`, see spec-0012), not by a general-purpose end-user CLI flow.
+    Use `npx qfai prototyping round-start --round <r5|r3|r2|r1> --candidates <csv> --target-url <url> --mode <mode>`
+    to generate the round-scoped review bundle and command plans the AI evaluator sub-agent consumes, then use
+    `round-harvest`, `round-narrow`, `round-absorb`, and `round-reimplement-verify` to advance the candidate funnel.
     `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
@@ -110,9 +113,11 @@ QFAI includes a small set of custom skills (stored under `.qfai/assistant/skills
   Run this once right after `npx qfai init`, and re-run it when the repository structure changes.
 - **qfai-discussion**: Run a unified structured discussion that produces and maintains the latest discussion pack
   as 15 required markdown files under `.qfai/discussion/discussion-<ts>/`.
-  UI-bearing discussion packs require `prototyping.yaml`; non-ui discussion packs do not.
+  UI-bearing discussion packs may include `prototyping.yaml` as an optional recommendation artifact; non-ui discussion packs typically omit it.
 - **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 UI prototype under the `full-harness` only / UI-only contract, with calibration-pack SSOT and screen-level Browser QA evidence.
+- **qfai-prototyping**: Build a contract-aligned UI prototype using the Playwright CLI + AI
+  evaluator harness (spec-0012). Modes (`low-cost`/`standard`/`full-harness`) run the same
+  strictest review cycle; only `maxCycles` (1/3/20) differs.
 - **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/`.
@@ -286,7 +291,7 @@ npx qfai validate --fail-on error
 
 Recommended baseline.
 
-- Keep CI on default/full validation (`qfai validate --fail-on error`); do not use `--phase refinement` in CI.
+- Keep CI on default/full validation (`qfai validate --fail-on error` or `qfai validate --profile verify --fail-on error`); do not use partial profiles in CI.
 - Keep `pnpm check-types:future` as a separate mandatory gate so future TS compatibility runs once without duplicating `pnpm ci:gate`.
 - Add a report step (`npx qfai report`) when you need a human-readable artifact.
 - Tune traceability globs in `qfai.config.yaml` to match your test layout.

package/assets/init/.qfai/assistant/agents/frontend-engineer.md

@@ -2,12 +2,12 @@
 
 ## Mission
 
-- Implement frontend behavior aligned with selected anchor, strategy, screen contracts, and product experience decisions.
+- Implement frontend behavior aligned with the selected direction, finalized design system, screen contracts, and product experience decisions.
 
 ## Domain Responsibilities
 
 - Build UI components, states, interactions, and user-facing flows.
-- Respect selected anchor, strategy, screen contracts, optional design tokens, optional fallback HTML/CSS mock, and screen flow constraints.
+- Respect the selected direction, finalized design system, screen contracts, optional design tokens, optional fallback HTML/CSS mock, and screen flow constraints.
 - Coordinate with backend changes without breaking surface contracts.
 - Implement with KISS and YAGNI: prefer the simplest component/state structure that satisfies the current contract, and do not add speculative hooks, props, flags, or abstraction layers.
 - Keep UI code cohesive and readable: isolate concerns, minimize hidden coupling, avoid duplication, and use existing patterns/utilities before inventing new ones.

package/assets/init/.qfai/assistant/agents/product-experience-architect.md

@@ -9,7 +9,7 @@
 - Define UX direction, user journeys, interaction patterns, and accessibility expectations.
 - Define visual design direction, tokens, typography, color, and layout hierarchy.
 - Design navigation structures, IA, and screen transition logic.
-- Integrate selected anchor, strategy, screen contracts, evaluation sidecars, and Mermaid flow consistency.
+- Integrate exploration artifacts, selected direction, finalized design system, screen contracts, and Mermaid flow consistency.
 - Resolve conflicts between specialist design perspectives as a unified product experience decision.
 - Apply KISS and YAGNI to product-surface design: only add screens, controls, states, and branching flows that are justified by the current user goal.
 - Keep product experience decisions cohesive and unsurprising across copy, IA, navigation, transitions, and visual hierarchy.
@@ -24,7 +24,8 @@
 - .github/instructions/principles.instructions.md
 - .instruction/00_universal/development-principles-checklist.md
 - .instruction/01_specialties/design.md
-- Selected anchor, strategy, screen contracts, evaluation sidecars, optional tokens, optional fallback HTML/CSS mock, and Mermaid flows
+- Exploration brief, reference pool, evaluation rubric, evaluator calibration, selected direction, finalized design system,
+  screen contracts, optional tokens, optional fallback HTML/CSS mock, and Mermaid flows
 - Runtime screenshots or rendered evidence when available
 
 ## Deliverables

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

@@ -111,7 +111,10 @@ Use the shared schema.
 - 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;
+  - validation evidence exists and `qfai validate --profile atdd --fail-on error` passes;
+  - Drift Protocol is enforced;
+  - test-layer policy is checked against `.qfai/assistant/steering/test-layers.md`;
+  - coverage floors and ratios are signals, not gates;
   - `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:
@@ -143,7 +146,7 @@ Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#delta-re
 - Do NOT declare completion based on unit/component tests.
 - `10_Plan.md` is the primary How SSOT for execution phases.
 - If `10_Plan.md` is missing, stop and run owner planning flow before proceeding.
-- Completion gate is validation with zero errors (`qfai validate --fail-on error`).
+- Completion gate is validation with zero errors (`qfai validate --profile atdd --fail-on error`).
 - Coverage obligations are mandatory:
   - `tests/e2e/**` must cover all required `US-*`.
   - `tests/integration/**` must cover all required `TC-*`.
@@ -216,7 +219,7 @@ Notes:
 - All required `US` are covered by E2E tests.
 - All required `TC` are covered by integration tests.
 - All required `CON-API` are covered by API tests.
-- Validation passes: `qfai validate --fail-on error`.
+- Validation passes: `qfai validate --profile atdd --fail-on error`.
 - Repository quality gates (format/lint/type/tests/pack) pass with evidence.
 - Evidence file exists and includes work orders + reviewer notes.
 - Completion is approved by a reviewer who did not implement tests.
@@ -315,7 +318,7 @@ Before declaring completion:
 2. Run:
 
    ```bash
-   qfai validate --fail-on error
+   qfai validate --profile atdd --fail-on error
    ```
 
 3. Run repository standard gates:

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

@@ -88,9 +88,10 @@ Use the shared schema.
 - 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`;
+  - doctor evidence exists: `qfai doctor --fail-on error` completed without failing checks;
   - Drift Protocol enforced;
-  - test-layer policy enforced where applicable.
+  - test-layer policy enforced against `.qfai/assistant/steering/test-layers.md`;
+  - tool-count heuristics are signals, not gates.
 - Route specialist reviewers from `.qfai/assistant/steering/agent-routing.yml`.
 - Default configure review set:
   - `completion-reviewer`

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

@@ -1,7 +1,7 @@
 ---
 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 classification-aware prototyping.yaml requiredness and OQ-driven exit."
+title: QFAI Discussion (Exploration Planner)
+description: "Run structured discussion that defines exploration conditions, evaluation rubric, and anti-goals for downstream prototyping."
 argument-hint: "<idea-or-problem> [--auto]"
 allowed-tools: [Read, Glob, Write, TodoWrite, Task, Bash]
 roles:
@@ -21,366 +21,106 @@ routing-profile: requirements-heavy
 mode: interactive-by-default
 ---
 
-<!--
-QFAI Skill Body (SSOT)
-- This file is intended to be referenced by tool-specific wrappers (e.g., GitHub/Claude/Codex skills).
-- Keep wrappers thin and route users to this skill body.
--->
-
-## /qfai-discussion - Unified Discuss + Require
+## /qfai-discussion - Exploration Planner
 
 [DRIFT-PROTOCOL:MANDATORY]
 
-## User Questions (AskUserQuestion Protocol)
-
-Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#user-questions-askuserquestion-protocol`.
-
-Skill-specific examples:
-
-- scope confirmation
-
-## FORMAT SSOT (Mandatory)
-
-- Follow `.qfai/assistant/instructions/shared-skill-operating-baseline.md#format-ssot-mandatory`.
-- Before writing artifacts, read:
-  - `.qfai/discussion/README.md`
-  - `.qfai/specs/README.md`
-  - `.qfai/evidence/README.md`
-  - `.qfai/assistant/steering/agent-catalog.yml`
-  - `.qfai/assistant/steering/agent-routing.yml`
-  - `.qfai/assistant/steering/review-profiles.yml`
-  - `.qfai/assistant/skills/qfai-discussion/references/review-cycle-playbook.md`
-- Keep templates as source of truth and preserve file naming/order.
-
-## Sub-agent Delegation (MANDATORY)
-
-Follow `.qfai/assistant/instructions/shared-skill-delegation-baseline.md`.
-
-### Orchestrator Protocol (MUST)
-
-- No additional overrides.
-
-### Capability Probe (MUST)
-
-1. Attempt the first required delegation at stage start.
-2. Treat that real delegation attempt as the capability check.
-3. If the delegation fails, stop the stage immediately and report remediation.
-
-### Delegation Failure (Hard Stop)
-
-- No additional overrides.
-- Do not simulate roles. If the first required delegation fails, stop the stage and report remediation.
-
-## Work Orders Summary
-
-Use the shared schema (per-row `Status (PASS/REVISE)` column, reviewer response `Result: PASS | REVISE`).
-
-### Reviewer Gate (MUST)
+## Goal
 
-- Delegate final completion gate to an independent `completion-reviewer`.
-- Reviewer response must include `Result: PASS | REVISE` (matching shared-skill-delegation-baseline.md#reviewer-response-template).
-- 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 or ratios are not gates; they are risk signals.
-- Do not declare DONE until all routed blocking reviewers return `PASS`; otherwise apply `REVISE`.
+Produce a unified 15-file discussion pack plus exploration-first UI sidecars so `/qfai-sdd` and `/qfai-prototyping` can operate without forcing an early visual direction decision.
 
 ## CRITICAL CONSTRAINTS (Read First)
 
 - Output path is fixed: `.qfai/discussion/discussion-YYYYMMDDhhmmssSSS/`.
-- Timestamp format is fixed to `YYYYMMDDhhmmssSSS`.
-- Required fixed files (all 15 are mandatory):
-  - `01_Context.md`
-  - `02_Inception-Deck.md`
-  - `03_Story-Workshop.md`
-  - `04_Sources.md`
-  - `05_Scope.md`
-  - `06_REQ.md`
-  - `07_NFR.md`
-  - `08_Glossary.md`
-  - `09_Constraints.md`
-  - `10_Policy.md`
-  - `11_OQ-Register.md`
-  - `12_OQ-Resolution-Log.md`
-  - `13_Deferred.md`
-  - `14_Review-Request.md`
-  - `99_delta.md`
+- Required fixed files (all 15 are mandatory) remain unchanged.
+- UI-bearing discussion packs may include `prototyping.yaml` as an optional recommendation artifact; non-ui discussion packs typically omit it.
 - Discussion completion requires `Disposition: open` count to be zero in `11_OQ-Register.md`.
-- `deferred` is allowed only when required metadata is complete in `13_Deferred.md`.
-- `02_Inception-Deck.md` and `03_Story-Workshop.md` MUST contain Mermaid diagrams in ` ```mermaid ` fences.
 - If UI requirements exist, behavior obligations are primary and HTML+CSS mock is optional fallback only.
-- UI-bearing discussion packs require `prototyping.yaml`; non-ui discussion packs do not.
-- `ui_bearing: true` requires `prototyping.yaml`.
-- `ui_bearing: false` means `prototyping.yaml` is not required.
-- `prototyping.yaml` must use the canonical namespaced schema under `prototyping:`.
-- The canonical namespaced block must include `recommended_mode`, `rationale`, `allowed_modes`, and `surface`.
-- Top-level recommendation keys are not supported, and coexistence with namespaced keys is invalid.
-- 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 and rationale; avoid duplicating finalized spec SSOT.
-- Example Mapping is mandatory and must be captured as `Example Seeds` sections in `03_Story-Workshop.md`.
-- OQ Register and Deferred tables must contain the canonical fields defined in the references.
-
-## UI-bearing Detection
-
-Use `.qfai/assistant/skills/qfai-discussion/references/ui-bearing-playbook.md` for the full playbook.
-
-### Surface Classification
-
-Classification is based on surface type, not interaction complexity.
-
-| Surface Type | UI-bearing | Sidecar Generation | Notes                        |
-| ------------ | ---------- | ------------------ | ---------------------------- |
-| web          | Yes        | Full uiux/ sidecar | UI-bearing                   |
-| mobile       | Yes        | Full uiux/ sidecar | UI-bearing                   |
-| desktop      | Yes        | Full uiux/ sidecar | UI-bearing                   |
-| cli          | Yes        | Full uiux/ sidecar | UI-bearing                   |
-| mixed        | Yes        | Full uiux/ sidecar | UI-bearing                   |
-| non-ui       | No         | No uiux/ directory | non-ui sidecar not generated |
-
-### Detection Signals
-
-- Prefer explicit surface declarations in `01_Context.md`.
-- HTML tags, Mermaid screen flows, and UI-related stories are supporting signals.
-- Classify by surface type, not by interaction complexity.
-
-### Sidecar Generation Flow
-
-Sidecar steps must execute in the order below. **(no parallel)** across Step 1c and Step 1d — Step 1c MUST complete before Step 1d starts.
-
-- Step 1a — Detect surface type from `01_Context.md`.
-- Step 1b — If non-UI, skip steps 1c–1d; no uiux/ directory is created.
-- Step 1c — Populate `04_Sources.md` Trend Scan entries for all visual categories (color, typography, visual motif, spacing, shape, imagery), each with an `evaluation_connection` field. **This step MUST finish before Step 1d.**
-- Step 1d — Derive trend-based evaluation axes in `uiux/21_design_eval_trend_derived.md`, drawing `source_refs` from the entries completed in Step 1c. Parallel execution with Step 1c is forbidden.
-- UI-bearing: generate the full sidecar family and apply UX intent cross-references.
-- non-ui: skip sidecar generation entirely; no uiux/ directory, no additional completion conditions.
-
-### UI-bearing Completion Conditions
-
-Use `.qfai/assistant/skills/qfai-discussion/references/discussion-completion-matrix.md` for the full matrix.
+- Discussion is planner-first: do not select a single visual winner and do not finalize the design system here.
+- Use artifact files, not conversational summaries, as the downstream handoff.
 
-UI-bearing completion is blocked until all of the following are true:
+## UI-bearing Canonical Sidecar Family
 
-1. Strategy is selected in `uiux/10_implementation_strategy.md`.
-2. Scoring axes are defined through the 3-layer evaluation family (`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`, `uiux/24_design_eval_dynamic_overrides.md`).
-3. Comparison completed and selected anchor documented in `uiux/30_option_comparison.md` and `uiux/31_selected_anchor_screen.md`.
-4. Contracts drafted in `uiux/40_screen_contracts.md`.
+UI-bearing packs must produce the following sidecars as primary truth:
 
-### Non-UI Completion
-
-For non-UI projects, no additional UI/UX conditions apply, no sidecar artifacts are required, and `prototyping.yaml` is not required. uiux sidecar generation is skipped.
-
-## UI-bearing Authoring Requirements
-
-- Detection uses **surface type classification** as primary SSOT; HTML tags / Mermaid flows are **supplementary detection hints, not the primary SSOT**. **Non-UI packs are exempt** from sidecar validators (UIX-VAL series) — **zero new issues** when properly classified.
-- UI-bearing packs rely on the **canonical sidecar family** as primary truth: `uiux/10_implementation_strategy.md`, `uiux/11_design_taste_interview.md`, `uiux/12_design_system.md`, `uiux/20..24_design_eval_*.md`, `uiux/30_option_comparison.md` (**option comparison**), `uiux/31_selected_anchor_screen.md`, `uiux/40_screen_contracts.md`, `uiux/50_review_input_bundle.md`.
-- **Competitive Reference Registry** (`04_Sources.md`) entries must populate `adopted_points`, `rejected_points`, `local_translation`. **Placeholder-like values (TBD/N/A/TODO/empty) are treated as missing**. `99_delta.md` must include `## Rejected Visual Directions` with rationale and recurrence prevention. **Sidecar-family validators** (UIX-VAL series) are the primary quality gates for UI-bearing packs; non-ui packs stay exempt. Playbook detail: `references/ui-bearing-playbook.md` / `references/discussion-completion-matrix.md`.
-
-## Goal
-
-Produce a unified 15-file discussion pack with explicit decisions, requirements, OQ states, and rationale so `/qfai-sdd` starts without unresolved blockers.
-
-## Non-goals
-
-- Editing `.qfai/specs/**` directly
-- Writing implementation-level details
-- Leaving open blockers hidden in free text
-
-## Mandatory Outputs
-
-- `.qfai/discussion/discussion-*/01_Context.md`
-- `.qfai/discussion/discussion-*/02_Inception-Deck.md`
-- `.qfai/discussion/discussion-*/03_Story-Workshop.md`
-- `.qfai/discussion/discussion-*/04_Sources.md`
-- `.qfai/discussion/discussion-*/05_Scope.md`
-- `.qfai/discussion/discussion-*/06_REQ.md`
-- `.qfai/discussion/discussion-*/07_NFR.md`
-- `.qfai/discussion/discussion-*/08_Glossary.md`
-- `.qfai/discussion/discussion-*/09_Constraints.md`
-- `.qfai/discussion/discussion-*/10_Policy.md`
-- `.qfai/discussion/discussion-*/11_OQ-Register.md`
-- `.qfai/discussion/discussion-*/12_OQ-Resolution-Log.md`
-- `.qfai/discussion/discussion-*/13_Deferred.md`
-- `.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` when the latest discussion pack is `ui_bearing: true`
-- Evidence file: `.qfai/evidence/discussion-YYYYMMDDhhmmssSSS.md`
-- Reviewer notes (`PASS` or `REVISE`)
+- `uiux/30_exploration_brief.md`
+- `uiux/31_reference_pool.md`
+- `uiux/32_design_anti_goals.md`
+- `uiux/33_exploration_rubric.md`
+- `uiux/34_evaluator_calibration.md`
+- `uiux/40_screen_contracts.md`
+- `uiux/50_review_input_bundle.md`
 
 ## Required Process
 
 1. Run the core interview for concept, scope, stakeholders, and constraints.
 2. Run Inception Deck and include at least one Mermaid diagram.
 3. Run Story Workshop, capture user stories and user flows, and keep HTML+CSS mock as optional fallback only.
-4. Register source traceability in `04_Sources.md`.
+4. Register source traceability and reference research in `04_Sources.md`.
 5. Capture scope, REQ, NFR, glossary, constraints, and policies.
 6. Run Example Mapping and capture `Example Seeds`.
 7. Update `11_OQ-Register.md`, resolve OQs until open count is zero, and move deferred items to `13_Deferred.md`.
-8. Update `12_OQ-Resolution-Log.md`, `14_Review-Request.md`, and `99_delta.md`.
-9. Generate `prototyping.yaml` only when the latest discussion pack is UI-bearing.
+8. Generate the exploration-first sidecar family for UI-bearing targets.
+9. Generate `prototyping.yaml` only when the latest discussion pack is UI-bearing and an explicit prototyping recommendation is useful.
 10. Request review and record the Reviewer result.
-11. **UI-bearing only — Design system initialization.**
-    - **Step 11.3 — Brand archetype selection and design system scaffold (autonomous; skip when `surface: non-ui`).**
-      - Phase A (brand autonomous selection): The agent MUST select one of the 8 canonical brand archetypes from `references/design-md-brand-catalog.md` by scoring each archetype against the taste-interview data in `uiux/11_design_taste_interview.md`. No human-confirmation prompt is issued; the agent selects autonomously using the highest aggregate score. When two archetypes tie, apply the tie-breaker: highest visual-theme weight, then alphabetical archetype name. The selected archetype name and rationale MUST be written into `uiux/12_design_system.md` under `## Visual Theme`.
-      - Phase B (customization): Apply the selected archetype's `aesthetic_properties` as defaults for Color Palette, Typography, Spacing & Layout, Component Style, and Animation & Motion. Override any default that conflicts with explicit constraints captured in `06_REQ.md`, `09_Constraints.md`, or taste-interview directives. Record each override and its rationale.
-      - Step 11.3 is idempotent: running it twice against the same `11_design_taste_interview.md` MUST produce byte-identical `12_design_system.md`.
-    - **Step 11.5 — Visual-axis derivation mandate (UI-bearing packs with visual-category Trend Scan entries).**
-      Before trend-derived axis finalization for any UI-bearing run, the agent MUST perform design guideline research against the applicable platform or library guidance and record the findings in `04_Sources.md` under `design_guideline_research`. The resulting TRD-XX visual axes MUST then be written in `uiux/21_design_eval_trend_derived.md`.
-      For every visual-category Trend Scan entry in `04_Sources.md` (color, typography, visual motif, spacing, shape, imagery), the agent MUST derive at least one corresponding TRD-XX axis in `uiux/21_design_eval_trend_derived.md` with `source_refs` pointing to that entry. If no visual axis is derived despite visual Trend Scan entries existing, UIX-VAL-T04 (WARNING per NFR-0007) will fire at validation time.
-
-Detailed execution guidance lives in:
-
-- `.qfai/assistant/skills/qfai-discussion/references/example-mapping-guide.md`
-- `.qfai/assistant/skills/qfai-discussion/references/oq-and-deferred-rules.md`
-- `.qfai/assistant/skills/qfai-discussion/references/discussion-coverage-checklist.md`
-- `.qfai/assistant/skills/qfai-discussion/references/review-cycle-playbook.md`
-
-## Example Mapping Perspectives (Mandatory)
 
-Use `.qfai/assistant/skills/qfai-discussion/references/example-mapping-guide.md`.
-
-Minimum perspectives that must remain visible:
-
-1. Happy path
-2. Negative path
-3. Edge or boundary
-4. Permission or role
-5. State transition when stateful
-6. Idempotency or retry when external I/O exists
-
-## OQ Data Model (Mandatory)
-
-`11_OQ-Register.md` must use the canonical fields from `.qfai/assistant/skills/qfai-discussion/references/oq-and-deferred-rules.md`, including:
-
-- `OQ-ID`
-- `Gate` (`discussion|sdd|atdd|tdd|ops`)
-- `Disposition`
-- `Owner`
-- `Recommendation`
-- `Next-Decision-Point`
-- `Evidence`
-
-## Deferred Metadata Rules (Mandatory)
+## UI-bearing Authoring Requirements
 
-`13_Deferred.md` must use the canonical deferred fields from `.qfai/assistant/skills/qfai-discussion/references/oq-and-deferred-rules.md`.
+- `30_exploration_brief.md` must define product intent, must-preserve interactions, brand signals, and differentiation targets.
+- `31_reference_pool.md` must define exploration references, adopted points, rejected points, and local translation.
+- `32_design_anti_goals.md` must contain explicit anti-goals and recurrence prevention notes.
+- `33_exploration_rubric.md` must define design quality, originality, craft, and functionality grading criteria.
+- `34_evaluator_calibration.md` must include good critique examples, too-lenient examples, blandness-fail examples, and originality-fail examples.
+- `50_review_input_bundle.md` must document best-of-history handling so later iterations are not automatically preferred.
 
-## Drift Protocol (Mandatory)
+## Completion Contract (Shared)
 
-At any point during discussion:
+Before declaring completion, you MUST:
 
-1. Record the drift event in `99_delta.md`.
-2. Assess impact on all 15 files plus `prototyping.yaml`.
-3. Update affected files and re-validate OQ exit conditions.
-4. If drift contradicts a rejected option, record recurrence prevention.
+- verify all 15 mandatory output files exist and are populated;
+- ensure `Disposition: open` count is zero in `11_OQ-Register.md`;
+- ensure every deferred item has full metadata in `13_Deferred.md`;
+- ensure `02_Inception-Deck.md` and `03_Story-Workshop.md` include Mermaid diagrams;
+- ensure the UI-bearing sidecar family is complete;
+- run `qfai validate --profile discussion --fail-on error` and fix discussion-owned findings;
+- avoid selecting a single visual winner in discussion artifacts.
 
-## Review Gate Artifacts (RCP)
+### Reviewer Gate (MUST)
 
-For each review cycle, create:
+Reviewer checks must confirm:
 
-- `.qfai/review/review-<timestamp>/review_request.md`
-- `.qfai/review/review-<timestamp>/R01_<reviewer>.md`, `R02_<reviewer>.md`, ...
-- `.qfai/review/review-<timestamp>/summary.json`
+- the 15-file discussion pack is complete
+- `Disposition: open` count is zero in `11_OQ-Register.md`
+- the UI-bearing sidecar family is complete when the pack is UI-bearing
+- discussion stayed planner-first and did not choose a single visual winner
+- Drift Protocol is enforced
+- review policy is checked against `.qfai/assistant/steering/test-layers.md`
+- planning and coverage heuristics are signals, not gates
+- review findings end with `Status (PASS/REVISE)`
+- Reviewer result is explicit as `PASS` or `REVISE`
 
-Use `.qfai/assistant/skills/qfai-discussion/references/review-cycle-playbook.md` for append-only review pack rules, rerun rules, and `summary.json` requirements.
+## Sub-agent Delegation (MANDATORY)
 
-## RCP Footer Include (MUST)
+Follow `.qfai/assistant/instructions/shared-skill-delegation-baseline.md`.
 
-- Include and follow `.qfai/assistant/skills/qfai-discussion/references/rcp_footer.md`.
-- Routing and rerun rules must stay synchronized with `agent-routing.yml` and the footer SSOT.
+### Orchestrator Protocol (MUST)
 
-## Required Coverage Topics
+- do not self-approve
+- use artifact files as the handoff surface
+- integrate delegated outputs only after checking pack completeness
 
-Use `.qfai/assistant/skills/qfai-discussion/references/discussion-coverage-checklist.md`.
+### Capability Probe (MUST)
 
-Minimum topics that must remain covered:
+- No additional overrides.
 
-1. product concept and target users
-2. scope boundary and anti-goals
-3. user stories and user flows
-4. functional requirements with source traceability
-5. non-functional requirements with measurable targets
-6. security, performance, and operational constraints
-7. domain glossary consistency
-8. project policies
+### Delegation Failure (Hard Stop)
 
-## Completion Contract (Shared)
+- No additional overrides.
+- Do not simulate roles. If the first required delegation fails, stop the stage and report remediation.
 
-Before declaring completion, you MUST:
+## Work Orders Summary
 
-- verify all 15 mandatory output files exist and are populated;
-- ensure `Disposition: open` count is zero in `11_OQ-Register.md`;
-- ensure every deferred item has full metadata in `13_Deferred.md`;
-- ensure `02_Inception-Deck.md` and `03_Story-Workshop.md` include Mermaid diagrams;
-- ensure `Example Seeds` coverage is explicit in `03_Story-Workshop.md`;
-- ensure UI-related stories include behavior obligations in `03_Story-Workshop.md`;
-- ensure UI-bearing sidecars satisfy the completion matrix;
-- avoid duplicating finalized spec content in discussion outputs.
-
-## Evidence (MANDATORY)
-
-Create/update: `.qfai/evidence/discussion-YYYYMMDDhhmmssSSS.md`
-
-Required sections:
-
-- Objective
-- Inputs reviewed (files/paths)
-- Interview summary
-- Inception Deck summary
-- Story Workshop summary
-- Requirements summary (REQ count, NFR count)
-- OQ register summary
-- Deferred summary
-- Work Orders Summary
-- Reviewer result (`PASS`/`REVISE`)
-
-## DONE Declaration (Mandatory Output)
-
-When done, report:
-
-- generated discussion path (`.qfai/discussion/discussion-*/`)
-- open OQ count
-- deferred OQ count
-- REQ count
-- NFR count
-- reviewer result
-- ready-for-next command (`/qfai-sdd`)
-
-## FINAL CHECKLIST (Check Last)
-
-- [ ] CRITICAL CONSTRAINTS were followed.
-- [ ] All 15 required discussion files `01..14, 99` were produced.
-- [ ] OQ register fields follow the canonical data model.
-- [ ] Deferred table fields follow the canonical deferred data model.
-- [ ] `Disposition: open` count is zero at completion.
-- [ ] 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 behavior obligations in `03_Story-Workshop.md`.
-- [ ] Mermaid fence rules were satisfied when diagrams were used.
-- [ ] Evidence file exists and includes Work Orders Summary + Reviewer result.
-- [ ] Reviewer returned `PASS`.
-
-## Completion Checklist (MUST)
-
-- [ ] This skill's Definition of Done is satisfied.
-- [ ] Required artifacts were produced or updated.
-- [ ] Diagram artifacts follow Mermaid fence rules.
-- [ ] Open questions were logged to the proper OQ file.
-- [ ] The completion message was presented to the user.
-- [ ] Next actions were enumerated for all available options.
-
-## Review Cycle Checklist (MUST)
-
-- [ ] Review artifacts were generated for each required discussion review cycle.
-- [ ] All required reviewers completed their reviews for each review pack.
-- [ ] Any feedback triggered return/fix and a new review pack was appended.
-- [ ] `summary.json` satisfies the required schema with `target.kind: "discussion"`.
+Use the shared schema (per-row `Status (PASS/REVISE)` column, reviewer response `Result: PASS | REVISE`).
 
 ## Completion Message & Next Actions (MUST)
 
@@ -388,12 +128,3 @@ You MUST end the user-facing output with a handoff sentence to `/qfai-sdd` in th
 
 - Japanese output (use this exact sentence):
   ディスカッションが完了しました。他に要望などがあればご提示ください。問題なければ『/qfai-sdd』と入力してください。
-- Non-Japanese output:
-  Use the same meaning in the active user language, and keep `/qfai-sdd` as a literal command token.
-
-- Proceed (recommended): `/qfai-sdd`.
-  Action: run preflight on the latest discussion pack and generate shared/spec artifacts.
-- Upstream idea is still unclear: rerun `/qfai-discussion`.
-  Action: continue discussion loops until OQ states are explicit and complete.
-- Need additional risk analysis before SDD:
-  Action: update `03_Story-Workshop.md` and `11_OQ-Register.md` with additional findings.

package/assets/init/.qfai/assistant/skills/qfai-discussion/references/rcp_footer.md

@@ -28,7 +28,7 @@
 
 ## Validate Hard Gate（必須）
 
-- 各 review cycle で `qfai validate --fail-on error --format github` を実行していること
+- 各 review cycle で `qfai validate --profile discussion --fail-on error --format github` を実行していること
 - `.qfai/report/validate.log` が存在し、最新の成果物に対応していること
 
 ---