github-mcp-server

Configures GitHub MCP server with Docker or hosted modes and selectable tool sets.

4.9k|489|Updated Aug 12, 2025
One-click install
npx skills add https://github.com/githubnext/gh-aw --skill github-mcp-server
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: github-mcp-server
Source: https://github.com/githubnext/gh-aw/tree/main/skills/github-mcp-server
Command: npx skills add https://github.com/githubnext/gh-aw --skill github-mcp-server

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill simplifies the complex configuration of GitHub Model Context Protocol (MCP) servers, enabling AI agents to securely and efficiently interact with GitHub's API. It helps workflow authors configure AI agents to automate tasks on GitHub without deep API knowledge.

Core Features & Use Cases

  • Flexible Operation Modes: Configure the MCP server in local (Docker-based) or remote (hosted) modes.
  • Granular Toolset Control: Enable specific GitHub API toolsets (e.g., repos, issues, pull_requests) or use [default] for common operations.
  • Secure Authentication: Set up read-only mode and custom GitHub tokens for enhanced security.
  • Use Case: A workflow author needs to grant an AI agent access to create issues and manage pull requests. This skill guides them to configure the github tool with toolsets: [issues, pull_requests] in their workflow's frontmatter.

Quick Start

Consult the github-mcp-server skill to configure your AI agent to use the repos and issues toolsets for managing GitHub repositories and issues.

Frequently Asked Questions about github-mcp-server

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I configure an AI agent to access GitHub APIs?

Configure GitHub API access by setting up the MCP server with a GitHub personal access token (GITHUB_PERSONAL_ACCESS_TOKEN environment variable), selecting local Docker or remote hosted mode, and specifying which toolsets (repos, issues, pull_requests, etc.) your agent needs in the workflow frontmatter.

Can I restrict an AI agent to read-only access on GitHub?

Yes, enable read-only mode by setting the GITHUB_READ_ONLY environment variable and use the X-MCP-Readonly header. This restricts the agent to querying GitHub data without modifying repositories, issues, or pull requests.

What GitHub toolsets can I enable for AI agents?

Supported toolsets include repos, issues, pull_requests, contexts, gists, labels, and others. Configure specific toolsets in your workflow's YAML-based configuration, or use [default] for common operations like repository and issue management.

Do I need Docker to run the GitHub MCP server?

Docker is required for local mode operation. For remote hosted mode, the MCP server runs on a remote host and your agent communicates via Bearer token authentication, eliminating the local Docker requirement.

How do I authenticate the MCP server with GitHub?

Provide a GitHub personal access token via the GITHUB_PERSONAL_ACCESS_TOKEN environment variable. In local mode, use this token directly; in remote mode, pass it as a Bearer token in request headers for secure API authentication.

Can AI agents automate GitHub workflows like creating issues and managing pull requests?

Yes, configure the agent with the issues and pull_requests toolsets to automate issue creation, commenting, and pull request management without requiring the agent to understand GitHub's API directly.