GitHub Agentic Chat MCP Server

This is an MCP (Model Context Protocol) server implementation for GitHub agentic chat using Go. It provides tools for interacting with GitHub through natural language and includes vector search capabilities.

Features

• Search GitHub repositories
• Create issues
• Vector search functionality
  • Add documents to vector store
  • Semantic search across stored documents
• Extensible structure for adding more features

Prerequisites

• Go 1.21 or later
• PostgreSQL with pgvector extension
• GitHub Personal Access Token
• OpenAI API Key
• Claude Desktop or other MCP-compatible client

Setup

1. Clone the repository:

git clone https://github.com/akhidasTech/github-agentic-chat-mcp.git
cd github-agentic-chat-mcp

2. Set up environment variables:

export GITHUB_TOKEN=your_github_token_here
export DATABASE_URL=postgres://user:password@localhost:5432/dbname
export OPENAI_API_KEY=your_openai_api_key_here

3. Set up PostgreSQL with pgvector:

CREATE EXTENSION vector;

4. Build the server:

go build -o bin/github-agentic-chat-mcp

5. Configure Claude Desktop: Edit ~/Library/Application Support/Claude/claude_desktop_config.json:

{
    "mcpServers": {
        "github-chat": {
            "command": "/absolute/path/to/bin/github-agentic-chat-mcp"
        }
    }
}

6. Restart Claude Desktop

Available Tools

GitHub Tools

search_repositories

Search for GitHub repositories using a query string.

create_issue

Create a new issue in a GitHub repository.

Vector Search Tools

add_to_vector_store

Add a document to the vector store for semantic search.

Parameters:

• content: The text content to store
• metadata: JSON string containing metadata about the content

Example:

{
    "content": "This is a document about GitHub Actions",
    "metadata": "{\"type\": \"documentation\", \"tags\": [\"github\", \"ci-cd\"]}"
}

vector_search

Perform semantic search across stored documents.

Parameters:

• query: The search query text
• limit: Maximum number of results to return (default: 5)

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT License