home / mcp / github chat mcp server
A Model Context Protocol (MCP) for analyzing and querying GitHub repositories using the GitHub Chat API.
Configuration
View docs{
"mcpServers": {
"asyncfuncai-github-chat-mcp": {
"command": "uvx",
"args": [
"github-chat-mcp"
],
"env": {
"GITHUB_API_KEY": "YOUR_GITHUB_API_KEY"
}
}
}
}You can run the GitHub Chat MCP server locally to index and query GitHub repositories. This MCP exposes two core capabilities: repository indexing to build a searchable model of a repo, and repository querying to ask questions about its content and architecture. Use an MCP client to start the server and interact with indexed data from your GitHub Repositories.
Start by launching the MCP server through your preferred MCP client. Use the standard command configuration below to run the server locally, then index a repository and begin asking questions about it. Typical workflows include: index a repository, then query for its tech stack, architecture, or dependencies. You can repeat the process for additional repositories as needed.
Prerequisites you need before installation: a GitHub Chat API key is required to access the service, and the uv runtime is used to run the MCP server locally.
Install the uv runtime (MacOS/Linux):
curl -LsSf https://astral.sh/uv/install.sh | shInstall the MCP server using the Cursor configuration (recommended for centralized setup):
{
"mcpServers": {
"github_chat": {
"command": "uvx",
"args": [
"github-chat-mcp"
]
}
}
}Alternatively, configure Claude Desktop to run the MCP server in a dedicated directory and environment (both require the uv runtime to be installed):
# claude_desktop_config.json
{
"mcpServers": {
"github_chat": {
"command": "uvx",
"args": ["github-chat-mcp"],
"env": {}
}
}
}If you want to debug or verify the MCP server, you can run the inspector tools to ensure the server starts and responds correctly. The inspector provides a local interface to monitor the MCP runtime and confirms your GitHub API key is configured properly.
Index and analyze a GitHub repository to build a searchable model of its content, dependencies, and structure.
Ask questions about the indexed repository to gain insights into architecture, tech stack, and key components.