1(self.webpackChunk_N_E=self.webpackChunk_N_E||[]).push([[254],{11032:(e,n,t)=>{"use strict";t.r(n),t.d(n,{default:()=>r,useTOC:()=>o});var i=t(23798),s=t(28293),a=t(67106),l=t(19935);function o(e){return[]}let r=(0,s.e)(function(e){let n={a:"a",h1:"h1",li:"li",p:"p",ul:"ul",...(0,l.R)(),...e.components};return(0,i.jsxs)(i.Fragment,{children:[(0,i.jsx)(n.h1,{children:"Agent Skills"}),"\n",(0,i.jsx)(n.p,{children:"Agent skills are modular components that define specific abilities or behaviors\nfor AI agents. Rather than hardcoded instructions, a skill is like a âjob\ndescriptionâ, it describes what the agent can do, the kind of input it expects,\nand the kinds of outputs or behaviors it can produce. Skills guide agents by\nproviding them with knowledge, context, and purpose, helping them to operate\neffectively in non-deterministic scenarios where flexibility and judgment are\nrequired."}),"\n",(0,i.jsx)(n.p,{children:"Key characteristics of agent skills:"}),"\n",(0,i.jsxs)(n.ul,{children:["\n",(0,i.jsx)(n.li,{children:"They enable agents to handle complex, open-ended, or ambiguous tasksâsuch as\nunderstanding documentation or synthesizing information from large volumes of\ntext."}),"\n",(0,i.jsx)(n.li,{children:"Skills are not tools themselves, but rather reusable competency modules that\nyou can compose, mix, and match."}),"\n",(0,i.jsx)(n.li,{children:"Skills provide structure and guidance, making agents more reliable and\nexplainable."}),"\n",(0,i.jsx)(n.li,{children:"They serve as excellent documentation, clarifying what the agent is capable of\nand how it should behave."}),"\n",(0,i.jsx)(n.li,{children:"Skills are defined in ways that LLMs (large language models) can understand."}),"\n"]}),"\n",(0,i.jsx)(n.p,{children:"A useful mental model for Agent Skills begins with the idea that whenever an LLM\nor agent is assigned a task, it starts from scratch with an entirely empty context.\nSkills provide initial guidance, highlighting what is important and how to begin."}),"\n",(0,i.jsxs)(n.p,{children:["Visit ",(0,i.jsx)(n.a,{href:"https://github.com/memgraph/skills",children:"Memgraph Agent Skills repo"}),", to start\nusing skills."]})]})},"/ai-ecosystem/skills",{filePath:"pages/ai-ecosystem/skills.mdx",timestamp:1789474418e3,pageMap:a.O,frontMatter:{title:"Agent Skills",description:"Overview of the Agent Skills within Memgraph's ecosystem"},title:"Agent Skills"},"undefined"==typeof RemoteContent?o:RemoteContent.useTOC)},24786:(e,n,t)=>{(window.__NEXT_P=window.__NEXT_P||[]).push(["/ai-ecosystem/skills",function(){return t(11032)}])}},e=>{e.O(0,[8293,7106,636,6593,8792],()=>e(e.s=24786)),_N_E=e.O()}]);
Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.