What is GitHub MCP Server?

GitHub MCP Server is a Model Context Protocol (MCP) server by GitHub, enabling AI models to interact with GitHub APIs for automation and data analysis.

What is the Model Context Protocol (MCP)?

MCP is an open protocol that standardizes how applications provide context to Large Language Models (LLMs), allowing seamless integration with external data sources and tools.

What are the primary use cases of the GitHub MCP Server?

The primary use cases include automating GitHub workflows, extracting data from repositories, and building AI-powered tools for the GitHub ecosystem.

How do I install the GitHub MCP Server?

You can install it via VS Code using one-click install buttons, manually configuring VS Code settings, using Docker, or building from source.

What prerequisites are required to run the GitHub MCP Server?

You need Docker installed and running, and a GitHub Personal Access Token with necessary permissions.

How can I configure the GitHub MCP Server with VS Code?

You can add a JSON block to your User Settings (JSON) file in VS Code or create a .vscode/mcp.json file in your workspace.

Can I use the GitHub MCP Server with Claude Desktop?

Yes, you can configure it by adding a JSON block to the Claude Desktop settings file.

What are toolsets in the GitHub MCP Server?

Toolsets are groups of functionalities that can be enabled or disabled to control which GitHub API capabilities are available to AI tools.

How do I specify which toolsets to use?

You can specify toolsets using command-line arguments or environment variables.

What is dynamic tool discovery?

Dynamic tool discovery allows the MCP host to list and enable toolsets in response to a user prompt, helping to avoid confusion caused by too many available tools.

Can I use GitHub Enterprise Server with the GitHub MCP Server?

Yes, you can use the --gh-host flag or the GITHUB_HOST environment variable to set the GitHub Enterprise Server hostname.

How can I override tool descriptions?

You can override tool descriptions by creating a github-mcp-server-config.json file or using environment variables prefixed with GITHUB_MCP_.

What is UBOS, and how does it relate to the GitHub MCP Server?

UBOS is a full-stack AI Agent Development Platform that helps orchestrate AI Agents, connect them with enterprise data, and build custom AI Agents. Integrating GitHub MCP Server with UBOS enhances AI-driven automation.

What toolsets are available in GitHub MCP Server?

Available toolsets include repos, issues, users, pull_requests, code_security, and experiments.

How do I use the special “all” toolset?

You can enable all available toolsets by providing the all value to the --toolsets flag or the GITHUB_TOOLSETS environment variable.

What is the purpose of the repo:// template?

The repo:// template is used to retrieve repository content at a specific path, branch, commit, tag, or pull request.

How do I create a GitHub Personal Access Token?

You can create a token in your GitHub settings under Developer settings -> Personal access tokens.

What types of tools are available for issues?

Tools include get_issue, get_issue_comments, create_issue, add_issue_comment, list_issues, update_issue, and search_issues.

Can I search for code using the GitHub MCP Server?

Yes, the search_code tool under the Repositories category allows you to search for code across GitHub repositories.