# CodeGraphX MCP server

> Provides a local, token-efficient codebase graph engine that lets AI agents perform structural reasoning and impact analysis via an MCP server.

- Repository: https://github.com/techcraze00/CodeGraphX
- Page: https://hostedmcp.org/servers/techcraze00/codegraphx
- Transport: —
- Stars: 3
- License: MIT

## Environment variables

| Name | Required | Secret | Description |
|---|---|---|---|
| `CGX_PROJECT_ROOT` | no | no |  |

## Tools (6)

- `get_graph_status`
- `list_files`
- `check_symbol_exists`
- `explain_impact`
- `verify_task`
- `get_session_diff`

## FAQ

### How does CodeGraphX save tokens for AI agents?

It uses a graph-based approach and a compact tabular encoding called TOON instead of JSON, allowing agents to get structural answers in a few hundred tokens instead of reading dozens of files.