mcp-server

Uttam-Mahata/mcp-server

3.2

If you are the rightful owner of mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The GitHub-Gemini MCP Server is an advanced integration of GitHub with the Google Gemini API, designed to provide intelligent code assistance and repository management.

Tools
2
Resources
0
Prompts
0

GitHub-Gemini MCP Server

An advanced Model Context Protocol (MCP) server that integrates GitHub with Google Gemini API for intelligent code assistance and repository management.

Features

  • Intelligent Code Analysis: Leverages Gemini's advanced reasoning capabilities to analyze code patterns and suggest improvements
  • Context-Aware Assistance: Uses GitHub repository context to provide better code suggestions and documentation
  • Function Calling: Intelligent tool selection for GitHub operations (issues, PRs, code search, etc.)
  • Structured Output: JSON responses for integration with other tools
  • Code Execution: Dynamic code analysis and execution for testing suggestions
  • Thinking Mode: Deep reasoning for complex coding problems
  • Context Caching: Efficient handling of large repositories with automatic caching

Installation

pip install -r requirements.txt

Configuration

Create a .env file with your API keys:

GEMINI_API_KEY=your_gemini_api_key
GITHUB_TOKEN=your_github_personal_access_token

Usage

Run the MCP server:

python -m github_gemini_mcp

Tools Available

GitHub Integration

  • Repository analysis and navigation
  • Issue management and automation
  • Pull request operations
  • Code search and discovery
  • Branch and commit operations

Gemini AI Features

  • Intelligent code suggestions
  • Automated documentation generation
  • Code review assistance
  • Bug detection and fixing
  • Architecture recommendations

Architecture

The server combines:

  • GitHub API integration for repository operations
  • Gemini 2.5 models for advanced reasoning
  • Function calling for intelligent tool selection
  • Context caching for performance optimization
  • Structured output for reliable integration

License

MIT License