GitHub

MCP ServerFree

The github mcp server connects an AI assistant to GitHub repositories, issues, pull requests, Actions, releases, and Projects. Its available tools depend on the user's permissions and the server configuration.

Last updated: November 3, 2026

Official. Built and maintained by the company or project behind the service, so usually kept in step with it. Not audited by us.
Community. Built and maintained by an independent developer or group, not by the company behind the service. The company has not reviewed it, and updates depend on its author.

About GitHub

GitHub MCP connects an AI assistant to GitHub, where people host and manage software repositories and collaborate through issues, pull requests, Actions, releases, and Projects. The github mcp server exposes tools that let an assistant inspect repositories, search files, read code, analyze commits, and work with project activity. Depending on the tools selected and the credentials provided, it can also create or update GitHub resources. The server's access follows the permissions of the user or token that authenticates it.

This GitHub MCP integration can be run locally or through GitHub's hosted remote endpoint. Local use supports communication over stdio, while the hosted endpoint uses HTTP. Which tools an assistant can use depends on the server configuration, host support, and GitHub permissions.

What can GitHub do?

The server groups tools around common repository and collaboration tasks. An assistant can browse repositories, search files, read code, and analyze commits. It can also work with issues and pull requests, including creating and updating them, and use tools related to GitHub Actions workflows and releases.

GitHub Projects tools can list projects, retrieve project details, and create or manage project items. Administrators can select toolsets to control which groups of capabilities the AI client receives. A read-only configuration leaves modification tools unavailable. Secret-scanning tools can scan code for exposed secrets, provided GitHub Secret Protection is enabled for the repository.

How GitHub connects and authenticates

The server can run locally, with stdio support, or use GitHub's hosted remote MCP endpoint over HTTP. Authentication options include personal access tokens and OAuth in supported configurations. OAuth support depends on the MCP host; other hosts that support remote servers may use personal access token authentication, with setup varying by host. The official README names VS Code, Claude Desktop, Cursor, and Windsurf as compatible-host examples. Support details differ between hosts, so these examples do not establish an exhaustive client list.

Who is GitHub best for?

This server suits people who want an AI assistant to work with repository code and GitHub collaboration tasks, such as reviewing files or commits, managing issues and pull requests, or handling project items. The official README names VS Code, Claude Desktop, Cursor, and Windsurf as compatible-host examples. You need a GitHub account and an appropriate access token or OAuth authorization. The account's permissions determine what the server can access. Specific GitHub capabilities may depend on the account's plan or enabled products.

What GitHub can access and change

The server can access only GitHub resources permitted to the authenticating user or token. Depending on the enabled tools and permissions, it can read repository information and code, and make changes to repositories, issues, pull requests, and other GitHub resources. Creating or updating these resources changes GitHub data and can affect repository work. Use read-only mode to remove modification tools, or limit access by selecting toolsets. Token permissions also constrain available actions; classic personal access token scopes can filter tools, while the GitHub API enforces fine-grained token permissions. Secret scanning requires GitHub Secret Protection to be enabled for the repository.

Alternatives to GitHub

For another GitHub-focused server, GitHub Official MCP Server is a related option. If your work centers on database projects rather than repository collaboration, Supabase MCP Server connects an assistant to Supabase. For local or managed database work, PostgreSQL MCP is another database-focused listing. These alternatives serve different parts of a development workflow, and their capabilities should be checked on their own listings.

Visit Website

Key Features

Browse repositories, search files, read code, and analyze commits.

Create and update issues and pull requests.

Use GitHub Actions workflow and release tools.

List GitHub Projects, retrieve project details, and manage project items.

Choose toolsets to limit which capability groups the AI client can access.

Use read-only mode to remove modification tools.

Scan repositories for exposed secrets when GitHub Secret Protection is enabled.

Before you installSecurity checklist for MCP servers

An MCP server runs code and gives your AI assistant access to files, accounts or data. A few minutes of checking is worth it.

Check the source. Make sure the repository belongs to the vendor or author you expect, that the vendor's own documentation links to it, and that it is actively maintained. Beware of lookalike names.
Read what it can do. Look at the tools the server exposes and the permissions or scopes it asks for. Start with read-only access where the server allows it.
Use the least access that works. Give it a dedicated API key or a narrowly scoped token, ideally on a test project or sandbox, not your production credentials.
Pin versions and review updates. Install a specific version, read changes before upgrading, and do not run copied commands you do not understand.
Treat what it returns as untrusted. Documents, web pages and tool results can contain text that tries to steer your AI assistant. Keep confirmation switched on for actions that write, delete, send or spend.
Know where your data goes. Find out whether the server runs on your machine or is hosted, and what it sends to the vendor or to third parties.
Keep a way back. Know how to revoke the key and remove the server from your client's configuration.

This page is general information, not security or legal advice, and listing a server is not an endorsement or a security review. We do not create, host or control MCP servers, and listing details such as commands, permissions or pricing can be out of date, so check the official source. You install and use MCP servers at your own risk. See the section on third-party products in our Terms of Service.

Pricing Plans

Free (open source)

$ 0 USD

The MCP server repository is MIT-licensed.
Requires a GitHub account and an appropriate access token or OAuth authorization.
Available actions depend on the authenticated user's access and token permissions.
GitHub plan charges are separate, and some capabilities depend on the plan or enabled products.

Pricing information last updated: November 3, 2026

Visit Website

FAQs

Does the GitHub MCP server have a separate usage fee?

The supplied facts list the server repository as MIT-licensed at $ 0 USD. GitHub plan pricing is separate, and the checked sources do not state a separate MCP server usage fee.

How does the GitHub MCP server authenticate?

It supports personal access tokens and OAuth in supported configurations. OAuth availability depends on the MCP host, and the host's setup varies.

Can I make the server read-only?

Yes. It can be configured in read-only mode so modification tools are not offered. You can also select toolsets to limit the groups of capabilities exposed to the AI client.

Which clients can connect?

The official README gives VS Code, Claude Desktop, Cursor, and Windsurf as compatible-host examples. Support and setup vary by host.

Can it scan repositories for exposed secrets?

Yes, when GitHub Secret Protection is enabled for the repository.