--- title: "GitHub MCP server · Connect AI agents to GitHub" description: "Connect inference.eu agents to GitHub's official MCP server. Read code, triage issues, review pull requests and check Actions runs, with read-only mode available." canonical: "https://inference.eu/connect/github/" language: "en" --- MCP servers · Developer tools # GitHub MCP server: connect AI agents to GitHub from the EU GitHub maintains an official, open-source MCP server, available as a remote server hosted by GitHub or as a local binary. It gives agents tools for repositories, issues, pull requests, Actions, code security and more, grouped into toolsets you can switch on and off. inference.eu agents connect to it with OAuth or a personal access token. [Contact sales](https://inference.eu/contact/)[What agents can do with it](#tools) 1. 01Authentication: OAuth 2. 02Official server 3. 03EU data centres ## What agents can do with it - 1 ### Repositories, code and commits Read files with \`get\_file\_contents\`, search code and repositories, list branches, commits and releases, and create or update files and branches when write tools are enabled. - 2 ### Issues and pull requests Read, create and update issues and comments, search across repositories, open pull requests, read diffs and changed files, and add reviews, using the \`issues\` and \`pull\_requests\` toolsets. - 3 ### Actions, security and projects Inspect workflow runs and job logs, rerun failed jobs, read code scanning, secret scanning and Dependabot alerts, and work with Projects, Discussions and notifications through optional toolsets. ## What GitHub and its MCP server are GitHub, owned by Microsoft and based in San Francisco, hosts source code, issues and CI pipelines for most software teams. Its official MCP server, `github/github-mcp-server`, is written in Go and published under the MIT licence. It comes in two forms. The remote server at `https://api.githubcopilot.com/mcp/` is hosted by GitHub and uses Streamable HTTP. The local server runs from a Docker image or binary over stdio, and is the option for GitHub Enterprise Server. Toolsets such as `repos`, `issues`, `pull_requests`, `actions` and `code_security` control which tools the agent sees. ## What agents can do with GitHub A triage agent can read new issues each morning, label them, find duplicates and ask reporters for missing details. A review agent can summarise a pull request's diff, point out risky changes and leave a review comment for the author. When a build fails, an agent can read the failed job's logs, explain the error and propose a fix in a new branch. Security teams can ask for open code scanning and Dependabot alerts across repositories and get a prioritised list with links. ## Setup and authentication On inference.eu, add the GitHub connector to an agent and sign in with GitHub through OAuth; the agent then acts as you, within your repository permissions. For automation accounts, you can instead paste a fine-grained personal access token limited to selected repositories and permissions. Tokens are stored encrypted in the EU and only reach the agents you allow. Pick the toolsets each agent needs; fewer tools also help the model choose correctly. Read-only mode removes every write tool, even if a toolset includes one. ## Data protection and where the data goes With the remote server, requests go to GitHub's own MCP endpoint and from there to the GitHub API, so your code and issues stay with GitHub, a US company, as they do today. With the local server, the process calls the GitHub API directly. On inference.eu, the model that reads code and issues runs in EU data centres, and your OAuth token or access token is stored encrypted in the EU. Source code may contain secrets or personal data in comments and test fixtures, so scope access to the repositories an agent works on. ## Approvals and safe use Public issues and pull requests can contain text written to manipulate an agent. GitHub's lockdown mode filters content from authors without push access on public repositories, which reduces that risk. Read-only mode is the safest start for analysis agents. On inference.eu you can require a person to approve write calls, such as merging a pull request, pushing files, closing issues or rerunning workflows, while reads run freely. Use separate tokens per agent and rotate them, so access can be withdrawn without affecting other work. ## GitHub Vendor GitHub, Inc. Vendor based in United States Category Developer tools Authentication OAuth Transport Local or remote Server [Official server](https://github.com/github/github-mcp-server) Documentation [github.com](https://github.com/github/github-mcp-server/blob/main/docs/remote-server.md) ## GitHub: common questions The questions buyers, developers and legal teams ask us most. Something missing? Ask us directly. Yes. GitHub maintains github/github-mcp-server, available as a remote server hosted by GitHub at https://api.githubcopilot.com/mcp/ and as a local server via Docker or binary. Both support OAuth or a personal access token. The remote server is easiest: nothing to install, and GitHub keeps it updated. The local server suits GitHub Enterprise Server, which the remote server does not support, or setups that need full control over the process. Groups of tools you enable together, such as repos, issues, pull\_requests, actions, code\_security, dependabot or projects. The default set is context, repos, issues, pull\_requests and users. Enabling fewer toolsets reduces context size and helps tool choice. Yes. Read-only mode offers only read tools and skips every write tool, even ones requested explicitly. It is a good default for agents that analyse code, triage issues or summarise pull requests without changing anything. As few as the agent needs. Prefer a fine-grained token limited to selected repositories, or OAuth so the agent acts with your own permissions. Use separate tokens per agent and rotate them regularly. A filter that hides content on public repositories from authors without push access, to reduce prompt injection through issues or comments. It is not an authorisation boundary, so keep using narrow tokens and approvals for write actions. GitHub Enterprise Cloud, including data residency on ghe.com, can use the remote server with its own endpoint. GitHub Enterprise Server needs the local server with the GITHUB\_HOST setting pointing to your instance. Code stays on GitHub and is fetched through the GitHub API for each request. The model on inference.eu that reads it runs in EU data centres, and your GitHub credentials are stored encrypted in the EU. [Ask a question](https://inference.eu/contact/) ## Related [All MCP servers](https://inference.eu/connect/#directory) - [LinearConnect inference.eu agents to Linear's official MCP server to find, create and update issues, projects and comments, with a read-only option.Developer tools](https://inference.eu/connect/linear/) - [Atlassian (Jira and Confluence)Connect inference.eu agents to Atlassian's official MCP server for Jira, Confluence and more. Search, read and update issues and pages within user permissions.Developer tools](https://inference.eu/connect/atlassian/) - [SlackConnect inference.eu agents to Slack's official MCP server. Search messages and files, read channels and threads, post updates and edit canvases with admin approval.Communication](https://inference.eu/connect/slack/) - [n8nConnect inference.eu agents to n8n's built-in MCP server to search and run the workflows you expose, and build workflows and data tables, with approvals.Automation](https://inference.eu/connect/n8n/) ## Move your AI workloads to EU infrastructure Tell us which models you use and roughly how much traffic you have. We reply within one working day with pricing and next steps. - Reply within one working day - Pricing for your volume - DPA ready to sign