Documentation

Get started with Neuledge

Install our tools and ground your AI agent in real data. Full documentation lives in each project's GitHub repo — this page gets you up and running fast.

@neuledge/context

GitHub

Local-first documentation for AI agents. Index library docs into SQLite, query via MCP.

1. Install

$ npm install -g @neuledge/context

Or use without installing: npx @neuledge/context

2. Add documentation

$ context add https://github.com/vercel/next.js

Point it at any Git repo with Markdown docs. It clones, parses, and indexes into a portable .db file.

3. Start the MCP server

$ context mcp

Your AI coding assistant now has sub-10ms access to accurate, version-specific docs.

4. Configure your editor

Add the MCP server to your AI coding assistant's configuration.

Claude Code

$ claude mcp add context -- npx @neuledge/context mcp

Cursor / VS Code (settings.json)

{
  "mcpServers": {
    "context": {
      "command": "npx",
      "args": ["@neuledge/context", "mcp"]
    }
  }
}

@neuledge/graph

GitHub

Semantic data layer for AI agents. Structured access to live data through a single lookup tool.

1. Install

$ npm install @neuledge/graph

2. Configure data sources

import { NeuledgeGraph } from "@neuledge/graph";

const graph = new NeuledgeGraph({
  sources: {
    products: { url: "https://api.internal/products" },
    pricing:  { url: "https://api.internal/pricing" },
  },
  cache: { ttl: 300 }, // 5-minute cache
});

3. Query with natural language

const result = await graph.lookup("current price for SKU-1234");
// Returns structured JSON — not free text

The agent describes what it needs; Graph routes the request to the right data source and returns structured JSON.