القائمة
Use when you have existing service design artifacts and want to stress-test them — finds contradictions, blind spots, missing coverage, and unsupported claims across the artifact suite
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Use when writing Rails controllers or implementing state changes - enforces resource extraction, thin controllers delegating to models, params.expect, and controller concerns for scoping
Use when designing database schema, writing migrations, or making data storage decisions - enforces UUIDs, account_id multi-tenancy, no foreign keys, and proper index patterns
Use when writing Hotwire (Turbo/Stimulus) code in Rails - enforces dom_id helpers, morph updates, focused Stimulus controllers, and JavaScript private methods
Use when writing background jobs or async operations - enforces thin job wrappers (3-5 lines) that delegate to models using _later/_now naming pattern
Use when writing Rails models - enforces state-as-records not booleans, concerns as adjectives namespaced under model, and concern extraction triggers
Use when naming classes, methods, routes in vanilla Rails codebases - enforces concern adjectives (-able/-ible only), no bangs for state, resource naming cascade, and commit message format
| name | service-design-critic |
| description | Use when you have existing service design artifacts and want to stress-test them — finds contradictions, blind spots, missing coverage, and unsupported claims across the artifact suite |
Core principle: Artifacts that agree aren't necessarily right. Artifacts that contradict reveal shallow understanding. Make implicit assumptions explicit. Find what's missing.
Read every artifact in the slice. For each pair:
| Artifact A | Artifact B | What to look for |
|---|---|---|
| Blueprint | Empathy map | Confidence claimed at a step with no backstage support? Contradiction. Arc peaks where no moment of truth exists? Misalignment. |
| Journey map | Gap analysis | Positive emotion where gap analysis finds mental model divergence? Investigate. Pain points in one but not the other? Incomplete. |
| Blueprint | Diagnosis | Moments of truth with no fail point analysis? Blind spot. Fail points with no blueprint lane? Orphaned. |
| Empathy map | Journey map | Emotional arcs align? Peaks and endings consistent? Says/thinks contradict journey thoughts? One is wrong. |
| Gap analysis | Diagnosis | Priority gaps with no PDSA test? No path to resolution. PDSA tests on non-priority gaps? Misallocated effort. |
| All artifacts | — | [gap] tags piling up in one area? That's where you need research, not inference. |
Flag any [hypothesis] treated as [confirmed] downstream. Flag sections with no confidence tag. Every [hypothesis] must state what would confirm it — if it doesn't, the hypothesis is unfalsifiable.
Map every stage and touchpoint across all artifacts. Which have thin coverage (one artifact only)? Which have none? Single-perspective findings can't be triangulated.
For each finding, prescribe a next action:
| Finding type | Action |
|---|---|
| Contradiction | Identify which artifact has weaker evidence. Gather data to resolve. |
| Blind spot | Specify which artifact section to populate. |
| Unsupported claim | Downgrade confidence tag. Note what would support it — interview, code audit, log query, ticket search. |
| Coverage gap | Run the missing skill against that stage. |
| Confidence drift | Correct the downstream tag. Trace the chain of inference. |
Produce a findings list, not a new artifact. Write to docs/service-design/<slice>/critic-findings.md or deliver conversationally. Order by severity: contradictions, blind spots, coverage gaps.
