zereight/gitlab-mcp

What it solves

This project provides a Model Context Protocol (MCP) server that allows AI clients (like Claude, Cursor, and VS Code) to interact directly with GitLab. It eliminates the need for AI agents to manually copy-paste data from GitLab, enabling them to manage projects, issues, merge requests, and pipelines through a standardized interface.

How it works

The server acts as a bridge between an AI client and the GitLab API. It supports multiple transport methods—stdio for local use, SSE for legacy clients, and Streamable HTTP for remote deployments. It offers flexible authentication options, including Personal Access Tokens (PAT), local OAuth2 browser flows, and a specialized MCP OAuth proxy for remote clients like Claude.ai.

Who it’s for

Developers and teams using GitLab for version control and CI/CD who want to integrate their GitLab workspace into AI-powered IDEs or agents to automate project management and repository browsing.

Highlights

• Broad GitLab Coverage: Supports projects, repository browsing, merge requests, issues, pipelines, wiki, releases, tags, labels, and milestones.
• Flexible Authentication: Offers PAT, local OAuth2, MCP OAuth proxy, and per-request remote authorization for multi-user deployments.
• Multi-Transport Support: Compatible with stdio, SSE, and Streamable HTTP.
• Client Compatibility: Ready-to-use setup guides for Claude Code, Cursor, VS Code, GitHub Copilot, and others.
• Granular Permissions: Includes a read-only mode and the ability to filter specific toolsets or individual tools to restrict AI access.