gitlab-mcp-server
@piatra-automation
About gitlab-mcp-server
Improved implementation of Anthropic's modelcontextprotocol/servers Gitlab MCP server
Basic information
Config
No standard config provided
This server doesn't expose a parseable MCP config block in its README. See the repository for install instructions.
RepositoryTools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is GitLab MCP Server?
A Model Context Protocol (MCP) server that allows AI assistants (like Claude) to interact with GitLab repositories. It provides comprehensive capabilities for managing projects, issues, merge requests, comments, labels, milestones, time tracking, and more. This enhanced implementation includes additional endpoints and improved error handling compared to the original MCP GitLab server.
How to use GitLab MCP Server?
Install globally (npm install -g @piatra-open-source/gitlab-mcp-server) or as a project dependency. Set the required environment variable GITLAB_PERSONAL_ACCESS_TOKEN (and optionally GITLAB_API_URL for self-managed instances). Run the server with gitlab-mcp-server or embed it in your application using the Node.js child process API.
Key features of GitLab MCP Server
- Project creation, update, deletion, forking, and search
- File operations: fetch, create/update, commit multiple files
- Issue management with filtering and time tracking
- Full CRUD for issue comments
- Label management: create, update, delete, attach/detach
- Milestone retrieval and creation
- User assignment and unassignment on issues
- Issue linking (relates_to, etc.) and link deletion
Use cases of GitLab MCP Server
- Create a project in a specific GitLab group with custom visibility
- Retrieve open issues with time tracking statistics
- Add spent time to an issue and update its attributes
- Manage comments (create, update, delete) on issues
- Create and assign labels and milestones to issues
- Link, unlink, and manage relationships between issues
- Automate repository administration (fork, delete, change visibility)
FAQ from GitLab MCP Server
What distinguishes this server from the original MCP GitLab server?
It adds comprehensive endpoints for delete/update projects, time tracking, notes, labels, milestones, user assignment, and issue linking, with fixed namespace_id handling and enhanced error messages.
What environment variables are required?
GITLAB_PERSONAL_ACCESS_TOKEN (mandatory) and GITLAB_API_URL (optional, defaults to https://gitlab.com/api/v4).
Why do I get a “file_path should be a valid file path” error?
GitLab expects properly formatted paths (forward slashes, no invalid characters, case-sensitive). Ensure the file path exists in the repository for updates and avoid special characters like #, ?, [, ].
How do I handle authentication or permission issues?
Verify your personal access token has the necessary scopes. Check you have adequate access rights to the repository and group. Be aware of GitLab API rate limits.
Does the server support self-managed GitLab instances?
Yes, by setting the GITLAB_API_URL environment variable to your custom GitLab API endpoint.
More Version Control MCP servers
Atlassian Bitbucket MCP Server
aashariNode.js/TypeScript MCP server for Atlassian Bitbucket. Enables AI systems (LLMs) to interact with workspaces, repositories, and pull requests via tools (list, get, comment, search). Connects AI directly to version control workflows through the standard MCP interface.
Code Review Server
crazyrabbitLTCA MCP server for code reviews
Atlassian Data Center MCP
b1ffMCP servers for the Atlassian products (Bitbucket, Confluence, JIRA) of the Data Center version
GitHub MCP Server Extension for Zed
LoamStudiosA GitHub MCP Server extension for Zed
Jira MCP
nguyenvanduocitA Go-based MCP (Model Control Protocol) connector for Jira that enables AI assistants like Claude to interact with Atlassian Jira. This tool provides a seamless interface for AI models to perform common Jira operations including issue management, sprint planning, and workflow tra
Comments