# CodeGraphContext

CodeGraphContext turns local source repositories into queryable code graphs for AI assistants through an MCP server and CLI.

Canonical URL: https://a2a-mcp.org/entry/codegraphcontext

Language: en

Category: MCP Servers

Updated: 2026-09-14

## About

CodeGraphContext turns local source repositories into queryable code graphs for AI assistants through an MCP server and CLI.

## Key Features

- Index code symbols and relationships for structured lookup.
- Use a repository graph to investigate dependencies and code context.

## Use Cases

Provide repository context for code review, impact analysis or implementation planning.

## Limitations

Keep the index aligned with the current source tree. This entry refers to the CodeGraphContext organization’s maintained repository.

## Sources

- Project documentation: https://github.com/CodeGraphContext/CodeGraphContext

## MCP Setup

### Prerequisites

- Python 3.10+ and pip.
- Choose a graph backend compatible with your operating system.

### Client setup

#### Other

Transport: `stdio`

```json
{
  "mcpServers": {
    "CodeGraphContext": {
      "command": "codegraphcontext",
      "args": [
        "mcp",
        "start"
      ]
    }
  }
}
```

- Install: pip install codegraphcontext
- Use codegraphcontext mcp setup to configure the client; index the intended repository.

The README offers embedded and external database options; FalkorDB Lite is Unix-only, while Kuzu supports Windows, macOS and Linux.

Setup docs: https://github.com/CodeGraphContext/CodeGraphContext

## Tags

- MCP
- code-intelligence
- knowledge-graph

## Alternatives

- Srclight
- Repomix MCP

## Sources

- [Project website](https://github.com/CodeGraphContext/CodeGraphContext)
- [Project documentation](https://github.com/CodeGraphContext/CodeGraphContext)

Last checked: 2026-09-14