astroicers
UserCraft-first quality review for Agent Skill repos — an LLM judges how well a skill is WRITTEN (trigger design, writing quality, scope, anti-hallucination); lint only filters packaging and security. From a 97-repo star-gradient study.
Categories
Indexed Skills (3)
skill-reviewer
審查任意 Agent Skill repo **寫得好不好**,輸出分級式剖面診斷。Use when 使用者要求 review skill、審查 skill 品質、評估 SKILL.md、檢查 skill repo、或問「這個 skill 算不算高品質/距下一級差什麼」。**主判是 craft 質化判讀(L-001~004:trigger 設計/寫作風格/scope 清晰/anti-hallucination),lint 只是先跑的 packaging 與安全過濾器,其分數不是品質結論。** 輸出三段式:craft verdict + tier benchmark + gap list。
quoted
Use this skill when the user says "deck," "slides," or "presentation" — the escaped quotes are the point of this fixture. It also has a tab here and a literal backslash\in the middle.
asp
Use when working with AI-SOP-Protocol (ASP) framework workflows. Handles: planning new features, architecture decisions, ADR creation, SPEC writing, pre-commit checklists, code review, project health audits, autopilot execution, QA verification, security review, reality checks, impact analysis, pipeline gate evaluation, enforcement status. Triggers: asp-plan, asp-ship, asp-audit, asp-review, asp-autopilot, asp-dev-qa-loop, asp-reality-check, asp-impact, asp-gate, plan feature, new feature, create ADR, write SPEC, pre-commit check, ready to commit, code review, health audit, check project health, autopilot, run roadmap, verify, qa, security, reality check, impact analysis, 計劃功能, 新功能, 建立 ADR, 寫規格, 提交前, 準備提交, 程式碼審查, 審查, 健康審計, 健康檢查, 自動執行, 跑 roadmap, 審計, 驗證, 品質, 安全, 影響分析.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.