GitHub Chat MCP
Feedback
What features would you love to see? | どんな機能が欲しいですか？ | Join our Discord to share ideas!
Use GitHub Chat with Cursor, Claude, and other MCP-compatible assistants to understand codebases instantly.
Model Context Protocol (MCP) Support
Use GitHub Chat directly with Cursor, Claude, and other AI coding assistants.
Quick Installation
# Install with pip
pip install github-chat-mcp
# Or with uv
brew install uv
uv install github-chat-mcpCursor Setup
{
"mcpServers": {
"github-chat": {
"command": "uvx",
"args": ["github-chat-mcp"]
}
}
}GitHub Chat API
GitHub Chat API gives you programmatic access to chat with any public GitHub repository. Just two simple endpoints to integrate AI-powered code understanding into your workflow.
1
Repository Indexing
First, index a GitHub repository to analyze its codebase.
POST https://api.github-chat.com/verify
This endpoint processes the repository and makes it available for chat interactions.
2
Chat with Repository
Ask questions about the repository and receive detailed AI responses.
POST https://api.github-chat.com/chat/completions/sync
The messages array can contain conversation history for multi-turn conversations.
3. Response Format
Responses from the API include detailed information structured in a way that separates the reasoning, answer, and source context.
Rationale
The AI's explanation of how it approached the question, showing its reasoning process.
Answer
The formatted answer to the query, with Markdown support for rich text formatting.
Contexts
Repository content used by the AI to generate the response, with metadata about each file.