stark-ai-de/agent-skills / incubator/skills/skill-maintenance/skill-repo-curator
skill-repo-curator
VerifiedMaintain a public Agent Skills repository. Use when updating skill catalogs, validating SKILL.md files, preparing a release, deprecating skills, reviewing install commands, checking README consistency, or keeping a skills repo ready for npx skills installation.
Skill Repo Curator
Goal
Keep a public Agent Skills repository coherent, installable, safe, and ready for users who install skills through npx skills.
When to use
- The user asks to maintain or release a skills repository.
- The README catalog may not match the
skills/tree. - Validation, deprecation, publishing, or install commands need review.
When not to use
- The user wants to write a single skill; use
skill-authoring-review. - The user wants install help only; use
skill-installation-support. - The repository is not a skills repository and needs a general audit; use
repo-health-audit.
Inputs
skills/**/SKILL.mdREADME.md,CHANGELOG.md,CONTRIBUTING.md,SECURITY.md, andLICENSEpackage.json, validation scripts, and CI workflows- Output from
npm run validate,npm run list, and the localnpx skillsdiscovery command.
Inputs to inspect
- Inspect the skill tree, catalog docs, validation scripts, ignored local helper installs, and install smoke-test output.
- Check that upstream helper skills are installed project-locally rather than vendored into the public catalog.
Process
- Inspect the skill tree and top-level repository files.
- Run the local validation and listing commands when possible.
- Compare README catalog entries against discovered skills.
- Check install commands for
<github-owner>/<repo>placeholders or final repository coordinates. - Review changelog and release checklist when preparing a release.
- Apply deprecation policy before removing or replacing a skill.
- Report the exact validation status, drift, and next action.
Workflow
Follow the process above, fix structural drift first, then update catalog, changelog, and release notes.
Decision points
- If a skill is obsolete, prefer deprecation over deletion.
- If README and skill tree disagree, update the catalog or explain why the skill is intentionally hidden.
- If validation fails, fix structural errors before content polish.
- If project-local helper skills exist under
.agents/skills/, usenpx skills@latest add ./skills --listfor local discovery so helper skills do not mask the publishable catalog.
Safety rules
- Do not publish, push, tag, install globally, or change remote settings without explicit approval.
- Do not add private paths, secrets, internal hostnames, or copied proprietary skill text.
- Do not vendor already-published upstream skills into
skills/orincubator/skills/; use ignored project-localnpx skillsinstalls. - Prefer read-only checks unless the user asks for repository edits.
References
Read only when needed:
references/agent-skills-release.mdwhen preparing a release for this repository.references/repo-release-checklist.mdbefore a release.references/deprecation-policy.mdbefore replacing or removing a skill.
Scripts
No bundled scripts.
Output format
Return:
- Repository status
- Catalog drift
- Validation result
- Release or deprecation notes
- Blocking issues
- Recommended next action
Failure modes
- If the
skillsCLI cannot discover local skills, report the exact command output. - If upstream helper skills are vendored into
skills/, block public release until they are removed or intentionally licensed for redistribution. - If validation scripts disagree, treat the stricter failure as release-blocking.
Completion criteria
- The repo validates.
- The README catalog matches the skill tree or intentional differences are documented.
- Install commands are correct for local and public use.
- Any release or deprecation path is explicit.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |