Rag query

Skill butchokoy25/lightrag-claude-skills/skills/rag-query

Query the personal LightRAG knowledge graph for persistent memory across sessionsFrom its SKILL.md

Install

npx -y skills add butchokoy25/lightrag-claude-skills --skill rag-query

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

4 things to look at

reads credentialsReads from 1 credential source: `LIGHTRAG_SERVER_URL`.
2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
runs commandsInstructs the agent to run 2 commands, including `node -e "const BASE = process.env.LIGHTRAG_SERVER_URL || 'http://YOUR_LIGHTRAG_HOST:YOUR_PERSONAL_PORT';\n(async () => {\n const auth = await fetch(BASE + '/auth-status').then(r => r.json());\n cons` and 1 more.
fetches URLsInstructs the agent to fetch 2 URLs, including BASE + '/auth-status' and 1 more.

SKILL.md

2.6 KB, 623 tokens by cl100k_base, as published. Nobody here has run it

RAG Query — Personal Knowledge Graph

Query your personal LightRAG knowledge graph to recall decisions, preferences, project context, and learnings from past sessions.

Query Modes

Mode	When to use	What it does
`hybrid` (default)	Most queries	Combines local entity relationships + global topic summaries
`local`	"What's related to X?"	Traverses entity relationships from specific nodes
`global`	"What do you know about topic Y?"	Summarizes across all documents on a topic
`naive`	Simple keyword lookup	Basic text search, fastest but least intelligent

How to query

Full query (returns answer)

node -e "
const BASE = process.env.LIGHTRAG_SERVER_URL || 'http://YOUR_LIGHTRAG_HOST:YOUR_PERSONAL_PORT';
(async () => {
  const auth = await fetch(BASE + '/auth-status').then(r => r.json());
  const token = auth.access_token;
  const res = await fetch(BASE + '/query', {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      'Authorization': 'Bearer ' + token
    },
    body: JSON.stringify({
      query: 'QUERY_HERE',
      mode: 'hybrid'
    })
  });
  const data = await res.json();
  console.log(JSON.stringify(data, null, 2));
})();
"

Context-only query (returns raw context for Claude to synthesize)

node -e "
const BASE = process.env.LIGHTRAG_SERVER_URL || 'http://YOUR_LIGHTRAG_HOST:YOUR_PERSONAL_PORT';
(async () => {
  const auth = await fetch(BASE + '/auth-status').then(r => r.json());
  const token = auth.access_token;
  const res = await fetch(BASE + '/query', {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      'Authorization': 'Bearer ' + token
    },
    body: JSON.stringify({
      query: 'QUERY_HERE',
      mode: 'hybrid',
      only_need_context: true
    })
  });
  const data = await res.json();
  console.log(typeof data === 'string' ? data : JSON.stringify(data, null, 2));
})();
"

Response format

When presenting results to the user:

Be concise — summarize, don't dump raw JSON
Cite the source type when relevant (e.g., "from a stored decision on 2026-03-15")
Flag potentially stale information (anything >30 days old)
If no results found, say so clearly — don't fabricate

Infrastructure

Server: YOUR_LIGHTRAG_HOST:YOUR_PERSONAL_PORT
Auth: Bearer token from /auth-status (guest/disabled mode)
Env var: LIGHTRAG_SERVER_URL

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.