Codegenome documentation
Get started instantly
pip install codegenome
codegenome analyze .
Build an AST
Localized tree-sitter loops convert source files into Abstract Syntax Trees — fast and language-aware.
Graph Dependencies
Links imports, calls, and class relationships into an on-disk SQLite graph in your .genome/ folder.
Serve to AI Agents
Acts as an MCP Server, streaming ultra-low-token context windows straight into Cursor, Claude, or any MCP client.
Quick start
Install, analyze your repo, and open the TUI in five minutes.
Installation
Python 3.11+, pip, editable dev install, and first graph build.
CLI commands
analyze, export, mcp-start, evolve, tui, and more.
MCP server
Connect Cursor, Copilot, Claude, and other agents to your graph.
First steps
-
Install the package
pip install codegenome
-
Build a graph in your project
cd /path/to/your/project codegenome analyze .
-
Explore with the TUI (Recommended) or export
codegenome tui codegenome export --format html --path .
The Terminal UI is the most user-friendly way to manage your workspace, start servers, and analyze code. We highly recommend using it as your primary dashboard.
Documentation sections
Getting started
- Overview — what Codegenome is and how it fits your workflow
- Installation — requirements and setup
- Quick start — end-to-end tutorial
Topics
- Knowledge graphs — nodes, edges, and
.genome/artifacts - Watch & live graph — real-time updates and browser UI
- Export formats — JSON, HTML, Obsidian, Cypher, and more
- Terminal UI — interactive dashboard (Recommended)
- AI agents & MCP — context for LLM tools
How-to guides
Reference
- CLI overview
- Command reference
- MCP server · MCP tools
- Workspace artifacts
- Legacy flag CLI (
python -m codegenome) - Troubleshooting
Tip
New to the project? Read Overview, then follow Quick start. For AI editors, jump to Cursor + MCP setup.