メニュー
Use when starting or continuing service design analysis for a project — checks what artifacts exist, identifies gaps, recommends which analytical skill to run next
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
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-coordinator |
| description | Use when starting or continuing service design analysis for a project — checks what artifacts exist, identifies gaps, recommends which analytical skill to run next |
Core principle: A complete analysis for a service slice requires all 7 artifact types. The coordinator ensures coverage, not perfection.
When invoked, execute this workflow. Do not skip phases.
Scan docs/service-design/ for existing slices. For the target slice, list every artifact with its last_updated date. If no slice exists, ask the user to name one before proceeding.
Say: "Slice <name> has N/7 artifacts: [list present], missing: [list absent]."
Flag any artifact older than 30 days as stale — treat stale the same as missing for recommendation purposes.
Use this default order. Earlier artifacts feed later ones — but any order works if the user has context to supply.
| # | Skill | Rationale | Prerequisites |
|---|---|---|---|
| 1 | service-design:actor-profile | Shared foundation; every artifact references actors | None |
| 2 | service-design:stakeholder-map | How all actors connect; staff experience drives customer experience | Actor profiles |
| 3 | service-design:service-blueprint | Operational backbone; other artifacts reference it | Stakeholder map enriches it |
| 4 | service-design:empathy-analysis | User's inner world (emotions, information needs) | Actor profiles (blueprint enriches it) |
| 5 | service-design:journey-map | User's experience sequenced over time and channels | Actor profiles (blueprint + empathy enrich it) |
| 6 | service-design:gap-analysis | Where understanding diverges from reality | Empathy + journey recommended |
| 7 | service-design:service-diagnosis | Bottlenecks + improvement design | Blueprint + stakeholder map recommended |
| 8 | service-design:service-design-critic | Stress-test the whole suite | 3+ artifacts required |
Pick the first missing artifact in this order. If all 7 exist, recommend the critic.
Say: "Recommend running <skill> next because [rationale]. Invoke now?"
If the user confirms (or didn't object), invoke the recommended skill immediately. Pass the slice name as context.
After the skill completes, return to Phase 1 — re-inventory and recommend again until the suite is complete or the user stops.
Remember: This is a command. Execute each phase in order. Inventory before recommending, recommend before invoking.
