qfai 1.7.9 → 1.7.13

package/assets/init/.qfai/assistant/skills/qfai-discussion/templates/uiux/30_comparison.md

@@ -1,45 +0,0 @@
-# Option Comparison
-
-## Purpose
-
-Compare 2+ design options against scoring axes to support anchor screen selection.
-
-## Comparison Matrix
-
-### Invariant Axes
-
-| Axis               | Option A: [Name] | Option B: [Name] | Notes       |
-| ------------------ | ---------------- | ---------------- | ----------- |
-| Usability (20)     | [score 1-5]      | [score 1-5]      | [rationale] |
-| Consistency (21)   | [score 1-5]      | [score 1-5]      | [rationale] |
-| Accessibility (22) | [score 1-5]      | [score 1-5]      | [rationale] |
-| Delight (23)       | [score 1-5]      | [score 1-5]      | [rationale] |
-
-### Trend-derived Axes
-
-| Axis            | Option A: [Name] | Option B: [Name] | Source Translation |
-| --------------- | ---------------- | ---------------- | ------------------ |
-| (trend-derived) | [score 1-5]      | [score 1-5]      | [research source]  |
-
-### Product-specific Axes
-
-| Axis               | Option A: [Name] | Option B: [Name] | Notes       |
-| ------------------ | ---------------- | ---------------- | ----------- |
-| (product-specific) | [score 1-5]      | [score 1-5]      | [rationale] |
-
-## Aggregate Scoring
-
-| Option   | Weights                               | Normalized Total | Threshold             | Rank   |
-| -------- | ------------------------------------- | ---------------- | --------------------- | ------ |
-| Option A | Invariant: 60%, Trend: 25%, Prod: 15% | [total]          | [accept/refine/pivot] | [rank] |
-| Option B | Invariant: 60%, Trend: 25%, Prod: 15% | [total]          | [accept/refine/pivot] | [rank] |
-
-## Recommendation
-
-Selected: [Option X] — [rationale for selection based on scores and constraints]
-
-## Cross-references
-
-- Invariant axes: `20_eval_axis_usability.md`, `21_eval_axis_consistency.md`, `22_eval_axis_accessibility.md`, `23_eval_axis_delight.md`
-- 3-layer model & aggregate rules: `23_eval_axis_delight.md` (Aggregate Scoring Rules section)
-- Anchor selection: `31_anchor.md`

package/assets/init/.qfai/assistant/skills/qfai-discussion/templates/uiux/31_anchor.md

@@ -1,27 +0,0 @@
-# Anchor Screen Selection
-
-## Purpose
-
-Document the selected anchor screen and rationale for the design direction.
-
-## Selected Anchor
-
-- Screen: [Screen name/identifier]
-- Source option: [Option X from 30_comparison.md]
-- Rationale: [Why this screen was selected as the anchor]
-
-## Anchor Screen Description
-
-- Primary purpose: [What the user accomplishes on this screen]
-- Key interactions: [List of primary interactions]
-- Entry points: [How users reach this screen]
-- Exit points: [Where users go from this screen]
-
-## Design Constraints
-
-- [Constraint derived from comparison and scoring]
-
-## Cross-references
-
-- Option comparison: `30_comparison.md`
-- Screen contracts: `40_contracts.md`

package/assets/init/.qfai/assistant/skills/qfai-discussion/templates/uiux/40_contracts.md

@@ -1,51 +0,0 @@
-# Screen Contracts
-
-## Purpose
-
-Draft interaction contracts for the anchor screen and key screens.
-
-## Anchor Screen Contract
-
-### Screen: [Screen name]
-
-- Route: [/path-to-screen]
-- Actor: [primary user role]
-- Purpose: [what the user accomplishes on this screen]
-
-#### Primary Tasks
-
-| Task             | Trigger       | Success Criteria     |
-| ---------------- | ------------- | -------------------- |
-| [primary task]   | [user action] | [observable outcome] |
-| [secondary task] | [user action] | [observable outcome] |
-
-#### Required States
-
-| State     | Trigger                | Display                     |
-| --------- | ---------------------- | --------------------------- |
-| empty     | Initial load, no data  | [empty state description]   |
-| loading   | Data fetch in progress | [loading indicator]         |
-| error     | Fetch failure          | [error message + retry CTA] |
-| populated | Data available         | [primary content layout]    |
-
-#### Transitions
-
-| From      | To        | Trigger              |
-| --------- | --------- | -------------------- |
-| empty     | loading   | Data fetch initiated |
-| loading   | populated | Data received        |
-| loading   | error     | Fetch failure        |
-| error     | loading   | Retry action         |
-| populated | empty     | Clear action         |
-
-#### Observable Outcomes
-
-| Outcome                    | Verification Method |
-| -------------------------- | ------------------- |
-| [expected user outcome]    | [how to verify]     |
-| [expected system behavior] | [how to verify]     |
-
-## Cross-references
-
-- Anchor selection: `31_anchor.md`
-- State coverage: see `../03_Story-Workshop.md` Design Direction Summary

package/assets/init/.qfai/assistant/skills/qfai-discussion/templates/uiux/50_review_bundle.md

@@ -1,31 +0,0 @@
-# Review Input Bundle
-
-## Purpose
-
-Consolidate all sidecar artifacts into a review-ready bundle for design reviewers.
-
-## Bundle Contents
-
-| Artifact           | Path                                 | Status                    |
-| ------------------ | ------------------------------------ | ------------------------- |
-| Strategy           | `uiux/10_strategy.md`                | [draft/reviewed/approved] |
-| Usability axis     | `uiux/20_eval_axis_usability.md`     | [draft/reviewed/approved] |
-| Consistency axis   | `uiux/21_eval_axis_consistency.md`   | [draft/reviewed/approved] |
-| Accessibility axis | `uiux/22_eval_axis_accessibility.md` | [draft/reviewed/approved] |
-| Delight axis       | `uiux/23_eval_axis_delight.md`       | [draft/reviewed/approved] |
-| Comparison         | `uiux/30_comparison.md`              | [draft/reviewed/approved] |
-| Anchor selection   | `uiux/31_anchor.md`                  | [draft/reviewed/approved] |
-| Screen contracts   | `uiux/40_contracts.md`               | [draft/reviewed/approved] |
-
-## Review Checklist
-
-- [ ] Strategy aligns with surface type and project constraints
-- [ ] All 4 invariant scoring axes defined with measurable criteria
-- [ ] 2+ options compared against all axes
-- [ ] Anchor screen selected with rationale
-- [ ] Screen contracts cover all 4 required states (empty/loading/error/populated)
-- [ ] Critique loop entries addressed or deferred with rationale
-
-## Reviewer Notes
-
-<!-- Reviewer feedback is captured here during the review cycle. -->

package/assets/init/.qfai/assistant/skills/qfai-discussion/templates/uiux/60_critique_loop.md

@@ -1,23 +0,0 @@
-# Critique Loop Tracking
-
-## Purpose
-
-Track iterative design critique rounds and resolutions.
-
-## Critique Log
-
-| Round | Source                    | Finding               | Severity          | Resolution                             | Status                   |
-| ----- | ------------------------- | --------------------- | ----------------- | -------------------------------------- | ------------------------ |
-| 1     | [reviewer/self/automated] | [finding description] | [high/medium/low] | [resolution or rationale for deferral] | [open/resolved/deferred] |
-
-## Resolution Summary
-
-- Total findings: 0
-- Resolved: 0
-- Deferred: 0 (with rationale in Deferred column)
-- Open: 0
-
-## Cross-references
-
-- Review bundle: `50_review_bundle.md`
-- Design decisions: see `../03_Story-Workshop.md` Design Direction Summary

package/assets/init/.qfai/assistant/skills/qfai-prototyping-full-harness/SKILL.md

@@ -1,185 +0,0 @@
-# /qfai-prototyping-full-harness
-
-[DRIFT-PROTOCOL:MANDATORY]
-
-Premium prototyping skill with planner/generator/evaluator iteration loop.
-Full-harness mode is an **explicit, non-default** path activated only via
-`qfai prototyping --mode full-harness` or discussion artifact recommendation.
-
-> This skill defines a real execution workflow — it is NOT a routing-only redirect.
-
-## When to Use
-
-- Projects requiring L3–L5 fidelity evidence (production-ready prototypes).
-- Evaluation needs: weighted multi-dimension scoring with decision gates.
-- Explicit opt-in only; never auto-activated from config, env, or standard mode.
-
-## Workflow Loop
-
-The full-harness iteration loop proceeds through four phases per cycle:
-
-### Planner Phase
-
-- Generate strategy from spec constraints, budget, and prior iteration feedback.
-- Produce a ranked plan with target dimensions, expected quality floor, and cost estimate.
-- Constraints propagated from discussion artifact recommendation.
-
-### Generator Phase
-
-- Execute prototyping output production based on planner strategy.
-- Incorporate refinement notes from prior evaluator feedback.
-- Produce render evidence, test results, and validator output as artifacts.
-
-### Evaluator Phase
-
-- Apply weighted scoring across configured dimensions (floor enforcement).
-- Decision gate: converge, refine, or pivot based on scoring-ready schema.
-- Record scoring trace with dimension-level breakdown for auditability.
-
-### Decision Gate
-
-- Convergence criteria: all dimension floors met AND aggregate score above threshold.
-- Refinement: loop back to generator with evaluator feedback.
-- Pivot: loop back to planner with strategic reassessment.
-- Termination: max iterations reached OR convergence achieved.
-
-## Evidence Collection
-
-Every full-harness run MUST produce:
-
-- Render evidence with `captured / skipped / failed` status (3-state vocabulary).
-- Iteration history (planner input → generator output → evaluator score per cycle).
-- Scoring trace with dimension-level breakdown.
-- Termination reason (converged / max-iterations / manual-stop).
-- Validator output from `qfai validate` at each iteration.
-
-## Reviewer Invocation
-
-- Generate review summary at convergence or termination.
-- Review findings must reference iteration history and scoring trace.
-- Reviewer sign-off is a required gate before evidence is finalized.
-
-### Reviewer Gate (MUST)
-
-- Final completion gate MUST be delegated to an independent Reviewer sub-agent.
-- Reviewer checks (minimum):
-  - evidence completeness and iteration history integrity,
-  - scoring trace auditability and calibration adherence,
-  - Drift Protocol was enforced,
-  - test-layer obligations match `test-layers.md` and plan,
-  - floors and ratios are **signals, not gates**.
-- Reviewer returns only `PASS` or `REVISE`.
-- Every `REVISE` must include a concrete alternative proposal.
-
-## Calibration Integration
-
-- Scoring-ready schema defines dimension weights and floor thresholds.
-- Calibration config is read from `qfai.config.yaml` under `prototyping.calibration`.
-- Threshold adjustments are logged in iteration history for traceability.
-
-## Three-Mode Positioning
-
-This skill covers **full-harness** mode only. See also:
-
-| Mode             | Scope                                   | Skill                                         |
-| ---------------- | --------------------------------------- | --------------------------------------------- |
-| **low-cost**     | Static checks only (L1/L2)              | `/qfai-prototyping`                           |
-| **standard**     | Static + optional light runtime (L2/L3) | `/qfai-prototyping`                           |
-| **full-harness** | Static + runtime-heavy (L3/L4/L5)       | `/qfai-prototyping-full-harness` (this skill) |
-
-Standard and low-cost modes are handled by the standard `/qfai-prototyping` skill.
-Full-harness adds runtime-heavy obligations: API non-404 checks, DB existence verification,
-and UI route reachability — these are NOT imposed on standard or low-cost modes.
-
-## Non-UI Projects
-
-For non-UI projects (`surface: non-ui`), this skill returns n/a.
-Full-harness obligations assume a UI-bearing surface.
-
-## Sub-agent Delegation (MANDATORY)
-
-This section is mandatory and overrides conflicting fallback text.
-
-### 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.
-
-### 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.
-
-### Simulation mode (Opt-in only)
-
-- Allowed only when the user explicitly states `Simulation mode allowed`.
-- Record both in evidence:
-  - `Subagents: simulated (reason: <why unavailable>)`
-  - `User approval: <quote or reference>`
-
-## Work Orders Summary
-
-Every major artifact in this stage MUST include this table:
-
-| Step | Role (sub-agent) | Task title | Input (refs) | Output (refs) | Status (PASS/REVISE) |
-| ---- | ---------------- | ---------- | ------------ | ------------- | -------------------- |
-| 1    | <role>           | <task>     | <refs>       | <refs>        | PASS/REVISE          |
-
-## CRITICAL CONSTRAINTS (Read First)
-
-- Scope is full-harness prototyping only — do not run standard/low-cost validation.
-- Contracts are strict inputs in this stage.
-- Runtime checks are mandatory for this mode (UI reachability, endpoint liveness, DB presence).
-- Evidence must capture iteration-level detail including scoring traces.
-- DONE is forbidden when iteration loop has not converged or max-iterations is not reached.
-
-## Completion Contract (Shared)
-
-Before declaring completion, you MUST:
-
-- resolve or explicitly defer ambiguous items,
-- verify every required artifact exists and is complete,
-- scan outputs for placeholders (TBD/TODO/OPEN QUESTION and equivalents),
-- run the smallest executable smoke proof and record outcomes.
-
-## Evidence (MANDATORY)
-
-Create/update both artifacts in `.qfai/evidence/`:
-
-1. Markdown evidence with sections:
-   - Iteration History (planner/generator/evaluator per cycle)
-   - Scoring Trace (dimension-level breakdown)
-   - Termination Reason
-   - Work Orders Summary
-2. JSON evidence with minimum fields:
-   - `iterations[]` with per-cycle scoring and decision
-   - `meta.generatedAt`, `meta.toolVersion`, `meta.commands[]`
-
-## FINAL CHECKLIST (Check Last)
-
-- [ ] All iteration cycles are recorded with scoring traces.
-- [ ] Convergence or termination condition was met.
-- [ ] Reviewer returned PASS.
-- [ ] Evidence artifacts are updated.
-- [ ] `qfai validate --fail-on error` passes.
-
-## Completion Checklist (MUST)
-
-- [ ] This skill's Definition of Done is satisfied.
-- [ ] Required artifacts were produced or updated.
-- [ ] Open questions were logged when needed.
-- [ ] Completion message was presented to the user.
-- [ ] Next actions were enumerated.
-
-## Completion Message & Next Actions (MUST)
-
-When complete, provide a final user-facing completion message and list actions.
-
-- Proceed (recommended): `/qfai-atdd`.
-  Action: implement acceptance tests against the full-harness prototype.
-- Quality gate run: `/qfai-verify`.
-  Action: run full validation/report flow and publish gate evidence.
-- Rework prototyping: rerun `/qfai-prototyping-full-harness`.
-  Action: fix convergence failures or scoring trace gaps.

package/assets/init/.qfai/assistant/steering/review-roster.yml

@@ -1,128 +0,0 @@
-schema_version: "1.0"
-
-feedback_policy:
-  alternative_required: true
-  description: "All reviewers MUST provide a concrete alternative or fix proposal when returning FAIL. Feedback without a concrete alternative is invalid and triggers re-judgment."
-
-roster:
-  - id: qa-lead
-    name: Quality Lead
-    scope: [discuss, require, sdd]
-    must_check:
-      - Verify scope, objectives, and requirement completeness.
-      - Verify risk, quality, and acceptance readiness.
-    can_be_na: false
-    na_rule: "N/A is not allowed for this reviewer."
-
-  - id: qa-gatekeeper
-    name: QA Gatekeeper
-    scope: [discuss, require, sdd]
-    must_check:
-      - Verify gate criteria and blocker handling rules.
-      - Verify review-cycle restart behavior on failure.
-    can_be_na: false
-    na_rule: "N/A is not allowed for this reviewer."
-
-  - id: reviewer
-    name: Independent Reviewer
-    scope: [discuss, require, sdd]
-    must_check:
-      - Verify consistency and independent pass/fail judgment.
-      - Verify evidence and rationale are reviewable.
-    can_be_na: false
-    na_rule: "N/A is not allowed for this reviewer."
-
-  - id: code-reviewer
-    name: Code Reviewer
-    scope: [discuss, require, sdd]
-    must_check:
-      - Verify maintainability and implementation-risk signals.
-      - Verify design intent is actionable for downstream coding.
-    can_be_na: true
-    na_rule: "Allowed only if no implementation-impacting decision exists."
-
-  - id: architect-reviewer
-    name: Architect Reviewer
-    scope: [discuss, require, sdd]
-    must_check:
-      - Verify architecture constraints and technical consistency.
-      - Verify decision trade-offs and rejected-option rationale.
-    can_be_na: true
-    na_rule: "Allowed only if no architecture-affecting decision exists."
-
-  - id: qa-reviewer
-    name: QA Reviewer
-    scope: [discuss, require, sdd]
-    must_check:
-      - Verify testability, edge cases, and failure-path coverage.
-      - Verify open/deferred items are explicit and actionable.
-    can_be_na: true
-    na_rule: "Allowed only if no quality-impacting change exists."
-
-  - id: frontend-reviewer
-    name: Frontend Reviewer
-    scope: [discuss, require, sdd]
-    must_check:
-      - Verify UI/UX, accessibility, and interaction implications.
-      - Verify user-facing flows and exception paths.
-    can_be_na: true
-    na_rule: "Allowed only if no frontend or UX impact exists."
-
-  - id: backend-reviewer
-    name: Backend Reviewer
-    scope: [discuss, require, sdd]
-    must_check:
-      - Verify backend/API/data consistency implications.
-      - Verify operational and reliability concerns.
-    can_be_na: true
-    na_rule: "Allowed only if no backend or data impact exists."
-
-  - id: design-review-lead
-    name: Design Review Lead
-    scope: [discuss, require, sdd]
-    must_check:
-      - Verify requirement/design coherence and structure quality.
-      - Verify information architecture and decision clarity.
-    can_be_na: true
-    na_rule: "Allowed only if no product/design decision changed."
-
-  - id: runtime-gatekeeper
-    name: Runtime Gatekeeper
-    scope: [discuss, require, sdd]
-    must_check:
-      - Verify operational readiness and runtime risk controls.
-      - Verify mitigation and rollback assumptions.
-    can_be_na: true
-    na_rule: "Allowed only if no runtime/operations impact exists."
-
-  - id: devils-advocate
-    name: Devil's Advocate
-    scope: [discuss, require, sdd]
-    must_check:
-      - Challenge every assumption, conclusion, and design decision as if it were fundamentally wrong.
-      - Provide a concrete desired-state alternative for every issue raised; bare negation without an alternative is invalid.
-      - Persist with objections using any rhetorical means (nitpicking, reductio ad absurdum, forced analogy) to surface hidden risks.
-    can_be_na: false
-    na_rule: "N/A is not allowed for this reviewer."
-    behavioral_principles:
-      premise: "Assume the current state is fundamentally incorrect."
-      stance: "Challenge assumptions aggressively and present objections clearly, even under uncertainty."
-      requirement: "When returning FAIL, always provide a concrete alternative proposal."
-    escalation:
-      consecutive_fail_limit: 3
-      demotion: advisory
-      demotion_scope: current_review_cycle
-
-  - id: pattern-doubler
-    name: Pattern Doubler
-    scope: [discuss, require, sdd]
-    must_check:
-      - Count all ID-bearing items (US, AC, BR, EX, TC) in the target spec and demand at least 2x the current count.
-      - Identify missing perspectives (boundary, negative, permission, state-transition, idempotency) and propose concrete additions with rationale.
-      - Persist with demands using any rhetorical means (edge-case invention, combinatorial explosion, analogy) to maximize pattern coverage.
-    can_be_na: true
-    na_rule: "Allowed only if the target artifact contains no ID-bearing spec items (US/AC/BR/EX/TC)."
-    behavioral_principles:
-      premise: "Current pattern coverage is insufficient."
-      stance: "Push clearly for at least 2x pattern coverage and challenge under-specified cases."
-      requirement: "Justify added patterns and set the doubling target from counts of ID-bearing items."

package/assets/init/.qfai/evidence/.gitignore

@@ -1,3 +0,0 @@
-*
-!.gitignore
-!README.md

package/assets/init/.qfai/report/.gitignore

@@ -1,3 +0,0 @@
-*
-!.gitignore
-!README.md

package/assets/init/.qfai/review/.gitignore

@@ -1,5 +0,0 @@
-*
-!.gitignore
-!README.md
-!review-*/
-!review-*/**