feat(design-verification): add scope-critique pass (#238)#243
Merged
Conversation
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds a ## Scope critique section to skills/design-verification/SKILL.md and a supporting reference file with five strategic-shape questions (scope, dependencies, simpler alternatives, observability, config surface) that run after the existing compatibility check to catch oversized scope, phantom dependencies, and unobservable capability bets before a design reaches create-issue. Motivated by #236 post-creation review: 6 strategic issues bypassed existing brainstorming, design-verification, and self-review gates. The new pass fills that gap with a bias-resistance checklist, red-flag lists per question, and a findings table that promotes BLOCKER rows to the top-level Blockers list. Closes #238 Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
dc46251 to
df51ba1
Compare
…236 #238 #239 Aligns version fields across .claude-plugin/plugin.json (already at 1.36.0 from #240), .claude-plugin/marketplace.json, and .feature-flow.yml. Cuts the [1.36.0] - 2026-04-17 release heading in CHANGELOG.md bundling the five features shipped today: - #238 Scope-critique pass for design-verification (PR #243, this PR) - #236 Advisor tool integration (PR #242) - #239 Senior developer panel, Phase 1c code review (PR #241) - #235 Codex consultation, Phase 1+2 opt-in (PR #240) - #234 Quick-path triage with code-aware scope confirmation (PR #237) Consolidates .changelogs/238.md fragment into CHANGELOG.md (per the Ship phase convention) and deletes the fragment. Updates README with a "What's New in 1.36.0" section listing all five PRs, adds a Codex Consultation subsection under Optional Enhancements, adds a Quick-path triage subsection, and updates the Skills table to include consult-codex and the scope-critique pass note for design-verification. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
### Step 4.5: Scope critiquestep toskills/design-verification/SKILL.mdwith a 5-question strategic-shape pass (scope, dependencies, simpler alternatives, observability, config surface)skills/design-verification/references/scope-critique.mdwith expanded prompt text, per-question red-flag lists, and bias-resistance checklist### Blockerslist (labelled[Scope critique Q<N>]), and excludesassumptions-onlymodeMotivated by #236 post-creation review: 6 strategic issues bypassed existing brainstorming + design-verification + self-review gates. The new pass fills that gap.
Test Plan
grep -c "^### Step 4.5: Scope critique" skills/design-verification/SKILL.md→1grep -c "Should this exist\|Could it ship smaller\|simpler version\|Observability\|Config surface" skills/design-verification/SKILL.md→5skills/design-verification/references/scope-critique.mdexists with expanded prompt text11(no regression in Step 1-7 / Verification Depth / Quality Rules / Additional Resources headings)Code Review Summary
Initial review: pr-review-toolkit:code-reviewer → APPROVED, no blocking issues.
Dogfood + senior-developer panel (opus, 3 personas) surfaced 4 findings, all fixed in this PR:
leaky-boundary### Blockers(labelled[Scope critique Q<N>])wrong-abstraction## Scope critique(H2 outside Process) to### Step 4.5: Scope critique(H3 inside Process); scope-independence Note added to Verification Depthscope-creepassumptions-onlymode now explicitly excluded from the Scope critique pass — matches narrow-mode contractmissing-seamexploration_results.schema/pipeline/uievidence first; opens referenced files directly only when the dependency domain is not coveredThe scope-critique pass was run against its own design (dogfooding) and produced 2 genuine WARNINGs (Q3 — reference file size trade-off; Q4 — AC-5 is LLM-self-assessed, acknowledged trade-off). Both are noted trade-offs rather than blockers. The pass did NOT surface the
assumptions-onlyscope-creep issue — a blind spot caught by the senior panel's Product Engineer lens.Implementation Context
Design Decisions
### Step 4.5: Scope critiqueinside## Process(between Step 4 and Step 5) so Verification Depth governs it alongside every other checkassumptions-onlymode skips the pass — assumption verification is a deliberately narrow mode and strategic critique is out of scope for itVerification Results
No blockers found during codebase exploration. Senior-panel judgment findings from the review process (4 total) all fixed in this revision.
Key Patterns Used
### Step N:numbered steps inside the## ProcessH2 (existing pattern; Step 4.5 inserted between 4 and 5)skills/<skill-name>/references/Closes #238