Indexes multi-language codebases into a knowledge graph so you can query, search, and edit code in plain English via an AI agent or MCP server.
ix-infrastructure/Ix
Ix parses a repository into a persistent graph of symbols, calls and imports so developers and AI coding agents can query its structure.
- Stars
- 1,097
- Forks
- 82
- Open issues
- 25
- Last commit
- yesterday
What is it?
Ix is a command-line tool that maps a codebase into a graph of symbols, calls, imports and containment relationships. It uses tree-sitter to parse the repository and stores the result in a local backend, so the graph survives between sessions instead of being rebuilt from search results each time.
The CLI answers questions such as what a symbol is, what calls it, what it depends on, and what would be affected by a change. Output can be plain text, JSON, or a token-minimal LLM format. An MCP server exposes the same graph to MCP-capable clients such as Claude Code, Codex, Cursor and VS Code, and a visualizer called Compass opens with `ix view`.
The backend runs as a Docker image using ArangoDB with a memory layer. The installer sets up Node.js, ripgrep and Docker where they are missing. Pre-built CLI packages cover Apple Silicon macOS, Linux and Windows x86-64, and Intel Macs can build from source via Homebrew.
Key features
- 01Parses repositories with tree-sitter into symbols, calls and imports.
- 02Persists the code graph locally across sessions.
- 03Answers queries like explain, trace, impact and callers for a given symbol.
- 04Provides an MCP server and an installer that registers it with several AI clients.
- 05Supports 27 programming languages plus several config and data formats.
Use cases
- Getting oriented quickly in an unfamiliar codebase.
- Checking what a change to a function might break before editing.
- Giving a coding agent structural context without pasting whole files.
- Finding load-bearing modules by ranking dependents.
Quick start
- 1
Run the official install script from the README (macOS/Linux)
- 2
Map a repository
ix map . - 3
Register your AI clients with the MCP server
ix mcp install
Things to know
- Requires Docker and Docker Compose for the local ArangoDB backend.
- The install script pipes a remote script into the shell, and Intel Macs need a Homebrew source build.
Topics
FAQ
- What is Ix?
- Ix parses a repository into a persistent graph of symbols, calls and imports so developers and AI coding agents can query its structure. Ix is a command-line tool that maps a codebase into a graph of symbols, calls, imports and containment relationships. It uses tree-sitter to parse the repository and stores the result in a local backend, so the graph survives between sessions instead of being rebuilt from search results each time.
- Is Ix free to use? What license does it have?
- Ix is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
- Is Ix still maintained?
- Actively maintained. The last commit to Ix was on Oct 21, 2026.
- What can I use Ix for?
- Getting oriented quickly in an unfamiliar codebase. Checking what a change to a function might break before editing. Giving a coding agent structural context without pasting whole files. Finding load-bearing modules by ranking dependents.
- How do I install Ix?
- Run the official install script from the README (macOS/Linux) Map a repository: ix map . Register your AI clients with the MCP server: ix mcp install
- Where can I get Ix?
- The source code and installation instructions are on GitHub: https://github.com/ix-infrastructure/Ix
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A CLI that scans your codebase and generates compact AI context files (CLAUDE.md, .cursorrules, AGENTS.md) so coding assistants spend fewer tokens exploring the project.
A CLI and web tool that packs an entire code repository into one AI-friendly file for feeding to LLMs like Claude or ChatGPT.
Indexes a codebase into a knowledge graph of dependencies, call chains and execution flows, and serves it to AI coding agents via MCP tools.
An MCP server that indexes code, docs and infrastructure into a local knowledge graph so AI coding agents can query structure instead of re-reading files.
Turns a codebase, its docs, schemas and PDFs into a queryable knowledge graph for AI coding assistants, built locally with tree-sitter AST parsing.