home / mcp / repomapper mcp server
A tool to produce a map of a codebase within a git repository. Based entirely on the "Repo Map" functionality in Aider.chat
Configuration
View docs{
"mcpServers": {
"pdavis68-repomapper": {
"command": "/usr/bin/python3",
"args": [
"/absolute/path/to/repomap_server.py"
]
}
}
}You can run RepoMap as an MCP server to expose continuous repository mapping capabilities to other applications. This enables automated code intelligence, prioritization of important files and symbols, and seamless integration with your delivery pipelines or large language models.
Set up an MCP client to communicate with the RepoMap MCP server via STDIO. The server listens for requests and returns structured repository maps that prioritize the most relevant files and code elements for your current project.
Prerequisites: Python 3.x installed on your system.
Install required Python packages.
pip install -r requirements.txtConfigure the MCP server as a standard STDIO server. Use the exact runtime command and path shown in the example configuration snippet.
{
"mcpServers": {
"RepoMapper": {
"disabled": false,
"timeout": 60,
"type": "stdio",
"command": "/usr/bin/python3",
"args": [
"/absolute/path/to/repomap_server.py"
]
}
}
}- The server uses STDIO for communication. No network port needs to be opened. - The MCP client should specify the absolute path to the repository you want to map when issuing requests. - Ensure the Python environment has access to the repository files you intend to map.
If the server does not start, verify that the path to repomap_server.py is correct and that Python 3 is installed. Check that the required dependencies from requirements.txt are installed in your environment.
Only expose the MCP server to trusted components and enforce appropriate authentication or access controls in your environment since repository maps may reveal sensitive code structure and dependencies.
Generates a structured map of the codebase showing files, definitions, and relationships to help you understand code organization.
Applies a graph-based ranking (PageRank-like) to highlight the most important files and symbols based on usage and references.
Uses persistent caching to speed up subsequent mappings and reduce recomputation.