メニュー
Run quality evaluation on the current solution. Grades every completed feature against acceptance criteria on Completeness, Correctness, Code Quality, Test Coverage, and Architecture Fit. Marks failures as needs-rework.
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
Run a goal-bounded autonomous loop over the harness — select, build, verify, and commit features one at a time with a maker/checker split, an explicit stop-contract, and a per-run token/commit budget. Best run in the orchestrator agent.
Sync FEATURES.json and PLAN.md after architecture or scope changes mid-project. Preserves completed feature status while updating not-started features to align with the latest ARCHITECTURE.md and PLAN.md.
Initialize a new harness-managed solution project from a brief idea. Use when starting a brand new project. Creates docs/IDEA.md, docs/PLAN.md, docs/FEATURES.json, docs/ARCHITECTURE.md, docs/PROGRESS.md, AGENTS.md, and .github/copilot-instructions.md.
Design a feature before building it — research latest patterns, evaluate approaches, design APIs/data models/UI, and produce a technical design document. Use after /add-features and before the builder starts implementing.
Provide feedback on a design document and request changes. Use after /design-feature to iterate on the technical design before implementation begins. Supports multiple rounds of review.
Plan new features to add to an existing onboarded project. Analyzes impact on existing modules, designs features that fit the current architecture, and ensures backward compatibility.
| name | evaluate |
| description | Run quality evaluation on the current solution. Grades every completed feature against acceptance criteria on Completeness, Correctness, Code Quality, Test Coverage, and Architecture Fit. Marks failures as needs-rework. |
Run the full evaluator protocol:
docs/PLAN.md, docs/FEATURES.json, docs/PROGRESS.md, and docs/ARCHITECTURE.md"complete":
docs/EVALUATION.md"needs-rework" in docs/FEATURES.jsonBe skeptical. The builder tends to approve its own work — your job is to catch what it missed.
| Criterion | 1 (Fail) | 3 (Adequate) | 5 (Excellent) |
|---|---|---|---|
| Completeness | Missing core functionality | All criteria met | Sensible additions |
| Correctness | Broken or buggy | Works for happy path | Handles edge cases |
| Code Quality | Messy, inconsistent | Clean and readable | Well-structured |
| Test Coverage | No tests or broken | Core paths tested | Edge cases tested |
| Architecture Fit | Violates boundaries | Follows architecture | Strengthens it |