seleznovivan-git-workflow

SKILLWorkflowcommunauté
v0.0.0SeleznovIvanMITMis à jour il y a 5 moisSource →

650-trial experiment proving that SKILL.md description wording has a 20× impact on Claude Code skill activation. Directive descriptions ("ALWAYS invoke...Do not X directly") achieve 100% vs 37% for passive ones.

Community-submitted skill. Not yet reviewed by the Forge team. Full prompt content may not be available.Request review →
1Étoiles du dépôt
1Clients
1Formats
il y a 5 moisDernière mise à jour
Skill
AuteurSeleznovIvan
Version0.0.0
LicenceMIT
CatégorieWorkflow
Formatsskill.md
PromptNon publié
Compatibilité
Claude✓ Pris en charge
Cursor—
Copilot—
ChatGPT—
Gemini—
À propos

650-trial experiment proving that SKILL.md description wording has a 20× impact on Claude Code skill activation. Directive descriptions ("ALWAYS invoke...Do not X directly") achieve 100% vs 37% for passive ones.

Mots-clés
skillclaude