home / mcp / omnigit github & local git mcp server

Omnigit GitHub & Local Git MCP Server

Provides an MCP interface to query GitHub and local Git data, enabling AI tools to read repos, manage issues/PRs, analyze code, and automate workflows.

Installation
Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "aifity-omnigit-mcp": {
      "url": "https://api.githubcopilot.com/mcp/"
    }
  }
}

The Omnigit MCP Server connects AI tools to GitHub and local Git contexts, enabling natural language interactions to read repositories, manage issues and PRs, analyze code, and automate workflows across GitHub and local environments.

How to use

To leverage this MCP server, configure your MCP client to connect to either the remote GitHub MCP server or run the local MCP server in a container or on your machine. Once connected, you can perform repository queries, issue and PR management, code analysis, and workflow automation using natural language prompts.

How to install

Prerequisites: You need Docker installed and running if you plan to run the local MCP server in a container. If you plan to use the remote GitHub MCP server, you only need network access to the remote endpoint.

1) Run the remote server (recommended if supported by your environment). 2) Run the local server in Docker or build from source as described below.

Local installation using Docker (recommended for quick starts):

Additional sections

Configuration and usage details are organized to help you tailor the MCP server to your workflow. You can enable read-only mode to prevent write operations, or lockdown mode to restrict content surfaced from public repositories. Familiarize yourself with environment variables and how to pass tokens securely.

Security best practices include using minimal GitHub token scopes, separate tokens per project, and rotating credentials regularly. Store tokens in environment variables where supported, protect your .env files, and avoid committing credentials.

If you want to customize descriptions or translations, you can override tool descriptions via a JSON configuration alongside the binary. You can export translations to preserve your overrides while incorporating new ones.