home / mcp / github mcp server

GitHub MCP Server

Provides AI-powered GitHub project management with end-to-end traceability and real-time task execution capabilities.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "harshkumarsharma-mcp": {
      "command": "mcp-github-project-manager",
      "args": [],
      "env": {
        "GITHUB_REPO": "your_repository_name",
        "GITHUB_OWNER": "your_github_username_or_organization",
        "GITHUB_TOKEN": "your_github_token",
        "GOOGLE_API_KEY": "your_google_api_key",
        "OPENAI_API_KEY": "your_openai_api_key",
        "ANTHROPIC_API_KEY": "your_anthropic_api_key",
        "PERPLEXITY_API_KEY": "your_perplexity_api_key"
      }
    }
  }
}

You can run and use the MCP GitHub Project Manager to transform project ideas into actionable tasks, manage GitHub Projects and issues with AI-assisted planning, and maintain complete traceability from business requirements to implementation. This guide shows practical steps to set up, run, and configure the server so you can start organizing work efficiently.

How to use

Install the MCP server and run it locally, then connect an MCP client to issue tools like PRD generation, task parsing, traceability creation, and AI-assisted task recommendations. You will generate PRDs from ideas, break them into tasks, link requirements to features and use cases, and continuously track changes and progress across GitHub Projects (v2). Use the server to compare business goals with implementation steps, estimate effort, and prioritize work using AI-powered guidance. You can also run the server in a Docker container for isolated environments or integrate with your preferred MCP client for automated workflows.

Key actions you can perform include generating PRDs from project ideas, parsing PRDs into development tasks, expanding complex tasks into subtasks, enhancing existing PRDs, and creating complete traceability matrices that link business requirements to features, use cases, and tasks. You can also analyze task complexity, get next-task recommendations, and manage sprints, milestones, and custom views within GitHub Projects.

How to install

# Option A: Install the MCP server globally (CLI)
npm install -g mcp-github-project-manager

# Start the MCP server locally (CLI)
mcp-github-project-manager

If you prefer running it directly from a project setup, clone the repository, install dependencies, and run the server. The steps below assume you will use a local development environment.

# Clone the repository
git clone https://github.com/kunwarVivek/mcp-github-project-manager.git
cd mcp-github-project-manager

# Install dependencies
npm install

# Build the project (if required by the setup)
npm run build

# Copy environment example and set values
cp .env.example .env
# Edit .env with your GitHub token and details

# Run the server (stdio transport by default)
npm run start

Configuration and prerequisites

Prerequisites you need before running the server: - Node.js installed (version compatible with the MCP project, typically Node 18+) - GitHub access token with repo, project, and org permissions - Optional: AI provider keys if you want AI-powered features (Anthropic, OpenAI, Google Gemini, Perplexity) - Docker (for containerized runs) if you prefer running the server in Docker

Notes on running with Docker

To run the MCP server in Docker, build the image and run it with the required environment variables. You can pass your GitHub credentials and optional AI keys when starting the container.

# Build the Docker image
docker build -t github-project-manager-mcp .

# Run the container with required environment variables
docker run -it \
  -e GITHUB_TOKEN=your_github_token \
  -e GITHUB_OWNER=your_github_username \
  -e GITHUB_REPO=your_repo_name \
  -e ANTHROPIC_API_KEY=your_anthropic_key \
  -e OPENAI_API_KEY=your_openai_key \
  -e GOOGLE_API_KEY=your_google_key \
  -e PERPLEXITY_API_KEY=your_perplexity_key \
  github-project-manager-mcp

Security and access

Ensure your GitHub token is kept secret and only exposed to trusted environments. Use the minimum required scopes for repository access and project management. If you enable AI-powered features, safeguard your AI keys and rotate them according to your organization’s security policy.

Troubleshooting

If you encounter module resolution errors, try using the containerized flow or ensure dependencies are installed correctly. Verify that your environment variables are set and that your GitHub token has the necessary permissions. For Windows-specific setups, use appropriate shell commands or scripts to export variables and run the server.

Examples of workflows

- Generate a PRD from a project idea, then parse it into a set of tasks linked to features and use cases. - Run AI-assisted complexity analysis on a critical task and receive recommended subtasks with estimated effort. - Create a traceability matrix that maps business requirements to implementation tasks and validates coverage.

Available tools

generate_prd

Generate a comprehensive Product Requirements Document from a project idea.

parse_prd

Parse a PRD into actionable development tasks with traceability links.

add_feature

Add a new feature with automatic impact analysis and task generation.

analyze_task_complexity

Analyze task complexity, effort, and risk using AI.

get_next_task

Provide AI-powered recommendations to prioritize the next task.

expand_task

Automatically break down a complex task into subtasks.

enhance_prd

Enhance an existing PRD with AI-powered gap analysis.

create_traceability_matrix

Create an end-to-end traceability matrix linking requirements, features, use cases, and tasks.