freemcplab / Annuaire / Playground / skill-optimizer

● Communauté hqhq1025 ⚡ Instantané

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).

▶ Voir la démo 📦 Installer 💡 Cas d'usage

Pourquoi l'utiliser

Fonctionnalités clés

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

Démo en direct

Aperçu en pratique

skill-optimizer-skill.replay ▶ prêt

0/0

Installation

Dépôt de l’auteur · Documentation officielle des Skills Claude

Cas d'usage

Usages concrets : skill-optimizer

How to diagnose and fix one underperforming skill

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

Quand l'utiliser : You notice one skill never activates, or activates wrongly.

Déroulement

Review the patch

Show me the proposed SKILL.md diff.✓ Copié

→ Minimal, reviewable diff
Apply and validate

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

→ Validation checklist

Résultat : A targeted improvement with an accountability plan.

Pièges

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

Combiner avec : skillcompass-skill

Polish your first SKILL.md before publishing

👤 First-time skill authors ⏱ ~20 min beginner

Quand l'utiliser : You wrote a SKILL.md and want expert review before shipping.

Déroulement

Ask for a review

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

→ Checklist with pass/fail + rationale
Fix issues

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

→ Clean SKILL.md

Résultat : A ship-ready SKILL.md.

Pièges

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

Combinaisons

Associez-le à d'autres MCPs pour un effet X10

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.✓ Copié

skill-optimizer-skill + filesystem

Apply SKILL.md patches directly

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

Outils

Ce que ce MCP expose

Outil	Entrées	Quand appeler
static-analysis	SKILL.md path	Before publishing or during iteration
log-based-analysis	logs + skill path	Post-use tuning
patch-generation	findings	After analysis

Coût et limites

Coût d'exécution

Quota d'API: none beyond LLM
Tokens par appel: 2–10k per skill analysis
Monétaire: free
Astuce: Static analysis is nearly free; log analysis costs more — run it only when iterating

Sécurité

Permissions, secrets, portée

Stockage des identifiants : none; logs may contain sensitive prompts — treat their path carefully

Sortie de données : none

Session logs can contain secrets — redact before sharing

Dépannage

Erreurs courantes et correctifs

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

Alternatives

skill-optimizer vs autres

Alternative	Quand l'utiliser	Compromis
skillcompass-skill	You need bundle-level ranking, not single-skill depth	Breadth over depth

Plus

Ressources

📖 Lire le README officiel sur GitHub

🐙 Voir les issues ouvertes

🔍 Parcourir les 400+ serveurs MCP et Skills