← ClaudeAtlas

implementlisted

Implement or fix decision-complete code directly, or execute an approved spec phase with verification. Use for build/fix requests or an approved spec path; route unresolved product or architecture decisions to spec. Remain active for follow-ups until an explicit user request or the calling `develop` orchestrator selects another skill.
leszekgruchala/buddy · ★ 3 · AI & Automation · score 66
Install: claude install-skill leszekgruchala/buddy
# Implement Remain in this skill for follow-ups. Do not activate another Buddy skill or act outside this skill; only an explicit user request or the calling `develop` orchestrator can select the next skill. Implement only the authorized request/spec. Do not expand scope or continue into unrelated work. ## Input modes - **Specified:** the request references an approved `spec_<work-name>.md` path; resume its TODOs and `## AGENT LOG`. - **Direct:** without an approved spec path, form a short working brief: outcome, exclusions, affected contracts, risks, and exact verification. Do not create a spec for narrow decision-complete work. - If a material product/architecture decision is unresolved, stop and require `spec`. ## Engineering rules 1. Change only authorized paths; remove only change-created orphans. 2. Before editing, load the [engineering contract](reference.md#engineering-contract) and only the applicable overlay: [Java/Kotlin](references/java-kotlin.md), [Python](references/python.md), or [TypeScript/JavaScript](references/typescript-javascript.md). 3. Use current primary library/API/SDK/CLI documentation. Find references before changing public signatures. 4. Preserve comments unless correcting them. 5. Reuse the spec worklog. For direct work, create `.ai/worklog/<yyyyMMdd>_<work-name>/trash/` only if scratch files are needed. ## Spec execution 1. Walk phases in dependency order; skip phases already marked SUCCESS. 2. Run `agent: Main` locally; otherwise dispatch