freemcplab / 디렉터리 / 플레이그라운드 / skill-optimizer

● 커뮤니티 hqhq1025 ⚡ 바로 사용

skill-optimizer

Name: skill-optimizer (Claude Skill)
Author: hqhq1025

Audit a single SKILL.md using real session data plus static analysis — tells you exactly what to change.

skill-optimizer looks at a SKILL.md through two lenses: (1) session logs — did it activate when expected, was the output useful — and (2) static analysis — is the frontmatter clean, is the description properly intent-scoped, are examples calibrated. It outputs concrete edits you can apply. Pairs well with SkillCompass (which does bundle-level ranking).

▶ 데모 보기 📦 설치 💡 사용 사례

왜 쓰나요

핵심 기능

Combines runtime evidence with static checks
Works on Claude Code, Codex, and any skill-compatible agent
Outputs a SKILL.md patch, not vague advice
Flags over-eager descriptions that cause mis-activation
Calls out duplicated capability across multiple skills

라이브 데모

실제 사용 모습

skill-optimizer-skill.replay ▶ 준비됨

0/0

설치

작성자 저장소 · Claude Skills 공식 문서

사용 사례

실전 활용법: skill-optimizer

How to diagnose and fix one underperforming skill

👤 Skill authors iterating on a specific skill ⏱ ~30 min intermediate

언제 쓸까: You notice one skill never activates, or activates wrongly.

흐름

Review the patch

Show me the proposed SKILL.md diff.✓ 복사됨

→ Minimal, reviewable diff
Apply and validate

Apply the patch and tell me what to watch for this week.✓ 복사됨

→ Validation checklist

결과: A targeted improvement with an accountability plan.

함정

Optimizing for activation rate but breaking scope — Include a 'false positive rate' acceptance metric

함께 쓰기: skillcompass-skill

Polish your first SKILL.md before publishing

👤 First-time skill authors ⏱ ~20 min beginner

언제 쓸까: You wrote a SKILL.md and want expert review before shipping.

흐름

Ask for a review

Review my SKILL.md against skill-optimizer's static checks.✓ 복사됨

→ Checklist with pass/fail + rationale
Fix issues

Apply the fixes and show me the final version.✓ 복사됨

→ Clean SKILL.md

결과: A ship-ready SKILL.md.

함정

Over-promising in the description — Be specific about what the skill does; Claude uses it to decide activation

조합

다른 MCP와 조합해 10배 효율

skill-optimizer-skill + skillcompass-skill

SkillCompass ranks the bundle; skill-optimizer dives into the worst offender

Use SkillCompass to pick the weakest skill, then optimize its SKILL.md.✓ 복사됨

skill-optimizer-skill + filesystem

Apply SKILL.md patches directly

Apply the proposed diff to ~/.claude/skills/my-skill/SKILL.md.✓ 복사됨

도구

이 MCP가 노출하는 것

도구	입력	언제 호출
static-analysis	SKILL.md path	Before publishing or during iteration
log-based-analysis	logs + skill path	Post-use tuning
patch-generation	findings	After analysis

비용 및 제한

운영 비용

API 쿼터: none beyond LLM
호출당 토큰: 2–10k per skill analysis
금액: free
팁: Static analysis is nearly free; log analysis costs more — run it only when iterating

보안

권한, 시크릿, 파급범위

자격 증명 저장: none; logs may contain sensitive prompts — treat their path carefully

데이터 외부 송신: none

Session logs can contain secrets — redact before sharing

문제 해결

자주 발생하는 오류와 해결

No logs found

Confirm Claude Code trace path; pass the path explicitly if non-default.

Suggestions conflict with intended scope

Tell the optimizer your scope constraints; it can respect non-goals

대안

skill-optimizer 다른 것과 비교

대안	언제 쓰나	단점/장점
skillcompass-skill	You need bundle-level ranking, not single-skill depth	Breadth over depth