メニュー
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.
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 when the user invokes /evo:publish or asks to release a new version, ship a release, or publish a tag. Wraps the go-native self-healing release pipeline (`evolve release`) — pre-flight checks, auto-changelog, atomic ship, marketplace propagation poll, auto-rollback on failure.
| name | retro |
| description | 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. |
Sprint 1.3 fan-out + Sprint 3 composable skill (v8.16+). Sub-reflectors merge via dedup-by-title.
ship completes (PASS) or after audit FAIL/WARN (capture lessons either way)ship or audit phase, transitioning to learn| Step | Action | Exit criteria |
|---|---|---|
| 1 | Read build-report.md, audit-report.md, state.json | Inputs loaded |
| 2 | Dispatch 3 sub-reflectors in parallel | 3 worker artifacts |
| 3 | Aggregator dedups ## Lesson: blocks across workers | <workspace>/retrospective-report.md produced |
| 4 | Optionally write to .evolve/instincts/lessons/<slug>.yaml | Lessons persisted |
| Sub-reflector | Focus | Output style |
|---|---|---|
retro-instinct | Reusable patterns across cycles | ## Lesson: <pattern title> |
retro-gene | Gene-pool updates based on cycle outcome | ## Lesson: <gene insight> |
retro-failure | Failed approaches with reproduction steps | ## Lesson: <failure pattern> |
retrospective-report.md carries deduplicated ## Lesson: blocks. Optionally one or more YAML files in .evolve/instincts/lessons/.
| Fact | Value |
|---|---|
| Phase | retrospective (control archetype, optional) |
| Persona | agents/evolve-retrospective.md |
| Profile | .evolve/profiles/retrospective.json — CLI agy-tmux, tier deep, fan-out ×3 |
| Inputs | audit-report.md · build-report.md |
| Artifact | retrospective-report.md (cycle workspace) |
retrospective-report.md — no required sections (classified by the phase runner).
Invoked by:
/evo:retroloop macro after /evo:shipFan-out prompts live in .evolve/profiles/retrospective.json:parallel_subtasks (count projected into Phase facts above).
.evolve/profiles/retrospective.jsonlegacy/scripts/dispatch/aggregator.sh (phase=learn → lessons mode with dedup)agents/evolve-retrospective.mdskills/loop/phase6-learn.md (legacy detailed workflow)