← ClaudeAtlas

mycomyco-skill-lifecyclelisted

Use this skill when running or debugging the Myco skill lifecycle end to end: identifying candidates from vault knowledge, curating them through approval, generating `SKILL.md` files, and evolving existing skills as the vault grows. It applies to `skill-survey`, `skill-generate`, `skill-evolve`, and any work on the Skills dashboard, including cases where candidates appear but no skills materialize, surveys return zero results, or generated skills need refreshes.
goondocks-co/myco · ★ 13 · AI & Automation · score 79
Install: claude install-skill goondocks-co/myco
# Myco Skill Lifecycle: Survey → Approve → Generate → Evolve The skill lifecycle is a four-phase pipeline that turns raw vault knowledge into structured, reusable SKILL.md files on disk. ## Pipeline Overview ``` vault spores → [Survey] → candidates → [Approve] → approved → [Generate] → SKILL.md files → [Evolve] → updated SKILL.md ``` - **Survey** (`skill-survey`): Scans vault spores for clusters of repeated knowledge worth encoding as skills. Produces `skill_candidates` with status `identified`. - **Approve** (Skills dashboard): Human review step. Candidates must reach `approved` status before skill-generate will act on them. - **Generate** (`skill-generate`): Writes SKILL.md files to `.agents/skills/<name>/` from approved candidates. - **Evolve** (`skill-evolve`): Identifies STALE skills and rewrites them with new vault knowledge. ## Scheduling Defaults (Asymmetric by Design) - `skill-survey`: `enabled: true` — auto-runs during idle. Passive discovery; no mutation. - `skill-generate`: `enabled: false` — requires explicit opt-in via Agent Tasks page. - `skill-evolve`: `enabled: false` — requires explicit opt-in via Agent Tasks page. This asymmetry is intentional: let the system build candidate evidence passively, but keep the generative and mutative steps user-driven until the user trusts the output quality. ## Phase 1: Survey **Task:** `skill-survey` | **Default:** auto-enabled Survey searches vault spores and sessions for recurring knowledge clusters. A good candi