メニュー
Use when the plan-review verdict is PROCEED (or plan-review is disabled) and code has not yet been written. Writes RED tests first, defining the contract Builder must satisfy. The mandatory hop before any implementation.
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 the user invokes /evo:loop or asks to run autonomous improvement cycles, self-evolving development, compound discovery, or multi-cycle code improvement with research, build, audit, and learning phases
Use after build has produced build-report.md. Validates the build via four parallel sub-auditors (eval-replay, lint, regression, build-quality) and produces ALL-PASS verdict. Adversarial mode default-on per CLAUDE.md.
Use when the user asks to explain how a feature/subsystem works and how the loops built it, or what the concurrent loops/lanes/cycles did and why. Default output = per-FEATURE HTML explainer pages (two movements — how it works today, then how it got built, with verified file:line claims and drift callouts) under docs/explain/; per-batch/per-lane story pages are the fallback framing. Derived from geoffreylitt/explain-diff-html, customized for evolve-loop.
Use when writing or changing any production code or tests — new features, bug fixes, refactors, or test authoring, in any language, by any model on any CLI.
Use when running any LLM agent below Fable 5 — Claude (Opus/Sonnet/Haiku), GPT via codex, Gemini via agy, or local models via ollama — to adopt Fable 5's operating discipline — evidence-first investigation, premise verification, root-cause-only fixes, adversarial self-review, calibrated autonomy, and honest failure reporting. Load at session start or as a persona overlay for phase agents on any CLI.
Use after ship completes. Three sub-reflectors (instinct, gene, failure) run in parallel to extract lessons, update gene pool, and analyze any failures. Off the latency-critical path.
| name | tdd |
| description | Use when the plan-review verdict is PROCEED (or plan-review is disabled) and code has not yet been written. Writes RED tests first, defining the contract Builder must satisfy. The mandatory hop before any implementation. |
Sprint 3 composable skill. Wraps the existing TDD-engineer phase with addyosmani-style structure. Mandatory hop before
build.
plan-review returns PROCEED, OR after Scout if plan-review disabled| Step | Action | Exit criteria |
|---|---|---|
| 1 | Read scout-report.md and plan-review.md | Inputs verified |
| 2 | Write RED tests (must fail on current code) | Tests exist; running them fails |
| 3 | Document the test contract in <workspace>/tdd-contract.md | Contract written |
| 4 | Append section to <workspace>/team-context.md | Bus updated for Builder |
Tests MUST fail when run against the current codebase. A test that passes immediately is NOT a TDD contract — it does not exercise the new behavior. Builder will refuse to start without confirmed RED state.
<workspace>/tdd-contract.md with sections: ## Tests Written, ## RED Verification, ## Contract for Builder<workspace>/team-context.md has populated ## TDD Contract section| Fact | Value |
|---|---|
| Phase | tdd (plan archetype, optional, gated by workflow.phase_enables.tdd=on) |
| Persona | agents/evolve-tdd-engineer.md |
| Profile | .evolve/profiles/tdd-engineer.json — CLI claude-tmux, tier balanced, single-writer |
| Inputs | scout-report.md · plan-review-report.md |
| Artifact | test-report.md (cycle workspace) |
test-report.md must declare:
## AC-Materialization (also accepted: ## Acceptance, ## Coverage Map)## RED Run Output (also accepted: ## RED Tests, ## Test Files Written)Invoked by:
/evo:tddloop macro after plan-review (or after discover when plan-review off)agents/evolve-tdd-engineer.md (persona definition).evolve/profiles/tdd-engineer.json