jersonboydmilan/agent-skills-compendium
Developer tools이 페이지는 AI로 자동 번역되었습니다. 기술적인 세부 사항은 원문을 확인하세요. 원문 보기
An open registry of 76 fully specified AI agent skills — taxonomy, canonical schema, composer, governance model, and a fleet runner that deploys hundreds of agents against one objective and rehearses...
개요
A structured intelligence layer for building, understanding, and deploying AI agent capabilities. This is a , not a prompt library. The distinction it is built on: - 🔧 A is something an agent can use — a browser, a SQL client, a deployment system. - 🧠 A is something an agent knows how to accomplish — with a trigger, inputs, a procedure, validation criteria, failure modes and an escalation path. A skill may invoke several tools to produce a validated outcome. 76 skills 20 categories 5 architectural layers 49 normalised tools 15 public API routes MIT per definition A prompt produces a result once. A skill declares a trigger, typed inputs, a procedure, validation criteria, failure modes and an escalation path — and can therefore be tested, governed and composed. MCP and equivalents answer how an agent reaches a system. A skill answers what it is trying to accomplish and how it knows it succeeded. They compose; neither replaces the other. That is a packaging format.
README
원문 README의 일부입니다. 원문 보기.
A structured intelligence layer for building, understanding, and deploying AI agent capabilities.
This is a capability registry, not a prompt library. The distinction it is built on:
- 🔧 A tool is something an agent can use — a browser, a SQL client, a deployment system.
- 🧠 A skill is something an agent knows how to accomplish — with a trigger, inputs, a procedure, validation criteria, failure modes and an escalation path.
A skill may invoke several tools to produce a validated outcome.
추천 도구
다른 키워드를 입력하거나 필터를 제거해 보세요.
설치
npx skillfish add jersonboydmilan/agent-skills-compendium