qfai 1.8.0 → 1.8.2

package/assets/uix-rev/scoring-review.md

@@ -1,58 +1,32 @@
 # UIX-REV: Scoring Review
 
-Review scoring-ready schema for canonical completeness.
-
-## Per-Axis Required Fields
-
-Each evaluation axis must include:
-
-1. `axis_id` — unique identifier
-2. `axis_name` — human-readable name
-3. `layer` — invariant / trend-derived / product-specific
-4. `origin` — derivation source (taste interview, trend scan, domain)
-5. `intent` — what this axis measures
-6. `why_it_matters` — rationale for inclusion
-7. `score_anchors` — concrete definitions for `low`, `mid`, `high`
-8. `positive_signals` — evidence that supports a high score
-9. `negative_signals` — evidence that supports a low score
-10. `anti_patterns` — patterns that should be flagged
-11. `evidence_required` — what evidence is needed to score
-12. `weight` — relative weight in aggregate scoring
-13. `minimum_floor` — minimum acceptable score (optional)
-14. `source_refs` — references to taste/trend sources
-15. `goal_refs` — references to project goals
-16. `review_questions` — questions the reviewer should ask
+Review exploration rubric and evaluator calibration for canonical completeness.
 
-## Evaluation Axes Quality
+## Required Rubric Dimensions
 
-- Axes must be measurable and verifiable
-- No generic/boilerplate axes without taste/trend derivation
-- Each axis must define `origin`, `intent`, `why_it_matters`, `score_scale`, `score_anchors.low/mid/high`
-- Each axis must define `positive_signals`, `negative_signals`, `anti_patterns`, `evidence_required`, `minimum_floor`
-- Each axis must define `source_refs`, `goal_refs`, and `review_questions`
+The rubric must include:
 
-## Aggregate Scoring Rules
+1. `Design Quality`
+2. `Originality`
+3. `Craft`
+4. `Functionality`
 
-Canonical field names (must match `23_design_eval_aggregate.md`):
+## Evaluation Axes Quality
 
-- `total_score_formula`: weighted composite formula
-- `layer_weights`: per-layer weight allocation (must sum to 1.0)
-- `accept_threshold`: minimum score for accept verdict (≥ 3.5)
-- `refine_band`: score range for refine verdict (2.5–3.4)
-- `pivot_band`: score range for pivot verdict (< 2.5)
-- `max_iterations`: maximum scoring iterations
-- `plateau_rule`: stagnation detection rule documented
-- `missing_score_policy`: how to handle missing axis scores
-- `disagreement_rule`: calibration disagreement handling documented
+- The rubric must be gradable and skeptical, not generic praise
+- Design Quality and Originality must create positive pressure against bland outputs
+- Craft and Functionality must operate as floors, not style-dominant constraints
+- Evaluator calibration must include good critique, too-lenient critique, blandness fail, and originality fail examples
+
+## Aggregate Scoring Rules
 
-### Trend-derived conversion check
+### Harness check
 
-- Trend scan results are converted to scoring axes
-- Trend scan results are reflected in selected anchor
-- Stale / overused AI slop avoidance is reflected in comparison and selection
+- Plateau / breakthrough conditions are explicitly documented downstream
+- Later iterations are not automatically preferred over stronger earlier ones
+- Breakthrough branches are judged against the incumbent with best-of-history handling
 
 ## Aggregate Review Focus
 
-- Review axis quality, overlap, and minimum floors as a single aggregate system
-- Review `source_refs` and `goal_refs` linkage for every scored axis
-- Remove old aggregate vocabulary and keep wording aligned with `23_design_eval_aggregate.md`
+- Review rubric pressure, floor conditions, and calibration quality as a single system
+- Remove old evaluation-axis vocabulary and keep wording aligned with `33_exploration_rubric.md` and `34_evaluator_calibration.md`

package/assets/uix-rev/strategy-review.md

@@ -1,27 +1,22 @@
 # UIX-REV: Strategy Review
 
-Review the UI/UX strategy document (`10_implementation_strategy.md`) for completeness and quality.
+Review the exploration brief (`30_exploration_brief.md`) for completeness and quality.
 
-This review is scoped to strategy completeness only. Selected anchor evaluation belongs to `comparison-review.md`.
+This review is scoped to planner-quality only. Winner selection belongs to the prototyping harness and `comparison-review.md`.
 
-## Required Fields
+## Required Sections
 
-- `surface`
-- `selection_required`
-- `decision`
-- `candidate_options`
-- `chosen_option`
-- `rationale`
-- `verification_expectations`
-- `notes_for_reviewer`
+- `Product Intent`
+- `Must-preserve Interactions`
+- `Brand Signals`
+- `Differentiation Targets`
 
 ## Alignment Check
 
-- `chosen_option` must exist in `candidate_options`
-- `decision` and `chosen_option` should use coherent vocabulary
-- Strategy `decision` must be consistent with the selected anchor in `31_selected_anchor_screen.md` (cross-reference only; SSOT judgment is in comparison-review)
-- Strategy must not contradict screen contracts in `40_screen_contracts.md`
-- `selection_required=false` with multiple `candidate_options` should be treated as a warning
+- The brief must be specific enough to support divergent exploration
+- The brief must not prematurely collapse to a single chosen direction
+- Brief language must remain consistent with `40_screen_contracts.md`
+- Differentiation targets must explicitly push away from generic defaults
 
 ## Verdict