Introduction
A local-first indexing engine that gives AI agents deep context over your codebase, docs, and knowledge via MCP.
Indexed
A local-first indexing engine that gives AI agents deep context over your codebase, documentation, Jira tickets, and Confluence pages — via MCP, running entirely on your machine.
Indexed parses your code with tree-sitter, your docs with Docling, embeds everything into a local hybrid index (BM25 + dense vectors), and exposes it to AI agents via the Model Context Protocol. Works with Claude Code, OpenAI Codex, Cursor, GitHub Copilot, and any MCP-compatible tool. No cloud. No API keys.
Index Your Knowledge
Indexing Overview
How collections work and what happens during indexing.
Local Files & Code
Index a folder of Markdown, PDFs, DOCX, and source code.
Jira
Make your Jira project semantically searchable.
Confluence
Index your team's Confluence wiki.
Search & AI Integration
Searching your Index
Search your collections directly from the CLI.
MCP Integration
Connect any MCP-compatible client in three steps.
Claude Plugin
Set up Indexed for Claude Desktop.
MCP Setup
Claude, Codex, Cursor, Copilot, and other MCP clients.