Skip to main content

universal-ai-skills

Compact plugin wrapper for automatic Universal AI Skills routing and on-demand skill loading through skill-router.

Ir a la instalación

Datos de origen

Repositorio
onfire7777/universal-ai-skills-library
Última actividad en el origen
22 de mayo de 2026 a las 01:58
Idioma detectado de SKILL.md
inglés
Estrellas
16
Forks
0

Opciones de instalación

De forma predeterminada está seleccionado el prompt que primero revisa el origen. Puedes cambiar a un comando directo o descargar una copia local.

Revisa los archivos de origen

Lee SKILL.md y los archivos complementarios que muestra SkillsMP antes de decidir si quieres instalarlo.

Mostrando SKILL.md

SKILL.md
Instrucciones de origen · Vista previa de solo lectura
name
universal-ai-skills
description
Compact plugin wrapper for automatic Universal AI Skills routing and on-demand skill loading through skill-router.
# Universal AI Skills Router - Canonical source: `%USERPROFILE%\universal-ai-skills-library`. - Binary: `%USERPROFILE%\go\bin\skill-router.exe` on PATH as `skill-router`. - For substantive user prompts, run `skill-router preflight --hook-event UserPromptSubmit --json "<latest user prompt>"` from prompt-submit hooks, or `skill-router preflight --json "<latest user prompt>"` from the host AI precheck. - Run automatic routing only for real user prompts. Do not route from tool output, assistant messages, startup, stop, compaction, status checks, or background jobs. - If `decision` is `route`, sanity-check that the selected skill clearly matches the user's core task, object, and action. Then load only that skill with `skill-router skill <name>`. - If `decision` is `ambiguous`, `host_ai_review.required` is true, or the match only hits generic words such as "issue", "install", "setup", "local", "AI", or "skill", continue without loading a skill unless one candidate is clearly correct. - Search unknown skills with `skill-router skill search <query> --limit 10`. - Keep this wrapper compact. Do not copy or paste the full skills corpus into plugin context.
Ver en GitHub