GitHub MCP Server
GitHub's official MCP server enables AI assistants to manage repositories, issues, pull requests, Actions workflows, and code security directly through natural language.
Overview
The GitHub MCP Server is GitHub's official Model Context Protocol server. It connects AI tools directly to GitHub's platform, enabling AI agents to read repositories, manage issues and PRs, analyze code, monitor CI/CD workflows, and automate team processes β all through natural language.
Official Server:
Developed and maintained by GitHub
Key Features
Repository Management
Browse and query code, search files, analyze commits, and understand project structure
Issue & PR Automation
Create, update, and manage issues and pull requests with AI-assisted triage and review
CI/CD Intelligence
Monitor Actions workflows, analyze build failures, manage releases
Code Security
Examine security findings, review Dependabot alerts, and manage secret scanning
Team Collaboration
Access discussions, manage notifications, analyze team activity and organizations
Remote & Local Modes
Use GitHub-hosted remote server or run locally via Docker for full control
Available Toolsets
The server organizes tools into configurable toolsets. Enable only what you need.
| Toolset | Description |
|---|---|
repos | Browse repositories, read files, search code, analyze commits |
issues | Create, read, update, search, and assign issues |
pull_requests | Create, review, merge, and manage PRs |
actions | Monitor workflows, trigger runs, view logs |
code_security | View code scanning alerts and security findings |
discussions | Read and list team discussions |
gists | Create, read, and manage gists |
git | Low-level Git operations via GitHub API |
orgs | Manage organization teams and members |
projects | Work with GitHub Projects boards |
dependabot | Review and manage dependency alerts |
notifications | Access and manage GitHub notifications |
Key Tool Examples
get_file_contentsβΆ
Retrieve file contents from any repository you have access to.
{
"owner": "vercel",
"repo": "next.js",
"path": "packages/next/README.md"
}
issue_writeβΆ
Create or update issues with labels, assignees, and milestones.
{
"method": "create",
"owner": "my-org",
"repo": "my-repo",
"title": "Add dark mode support",
"body": "Implement dark mode toggle in settings",
"labels": ["enhancement", "frontend"]
}
search_codeβΆ
Search code across repositories with query syntax and filtering.
{
"query": "useEffect cleanup function",
"owner": "facebook",
"repo": "react",
"language": "typescript"
}
actions_getβΆ
Get details of Actions workflows, runs, jobs, and artifacts.
{
"method": "get_workflow_run",
"owner": "my-org",
"repo": "my-repo",
"resource_id": "1234567890"
}
Installation
{
"servers": {
"github": {
"type": "http",
"url": "https://api.githubcopilot.com/mcp/"
}
}
}
Authentication:
Remote mode uses OAuth via your MCP host. Local Docker mode requires a GitHub Personal Access Token with appropriate scopes.
Toolset Configuration
Control which GitHub capabilities are available to your AI:
# Minimal setup β issues and PRs only
GITHUB_TOOLSETS="repos,issues,pull_requests"
# Read-only mode β no write operations
GITHUB_TOOLSETS="repos,issues,pull_requests,actions,code_security"
--read-only
# All tools
GITHUB_TOOLSETS="all"
Sources
Related Articles & Guides
Accessibility Scanner MCP Server
Accessibility Scanner MCP servers enable AI models to perform WCAG compliance checks, capture annotated screenshots, and generate detailed accessibility reports.
Auth0 MCP Server
Auth0 MCP server enables AI models to interact with Auth0 identity and access management, including applications, APIs, actions, logs, and forms.
AWS MCP Server
Interact with Amazon Web Services through your AI assistant using the AWS API MCP Server.