Skip to content
GitLab MCP Server

GitLab MCP Server

Featured Developer Tools
gitlabdevopsci-cdcode-reviewissues

Summary

GitLab's own MCP server — issues, merge requests, CI/CD pipelines, job logs, wikis and semantic code search over an OAuth-authorised HTTP endpoint.

Features

  • Create and read issues, work items and threaded notes
  • Full merge-request surface: create, list, diffs, commits, pipelines and discussion
  • Retrieve CI/CD job logs and manage pipelines (create, retry, cancel, delete)
  • Instance-wide search plus semantic_code_search for relevant snippets
  • Wiki page listing and label search
  • OAuth 2.0 Dynamic Client Registration — no personal access token to manage

Installation

Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.

Any MCP-compatible agent

Most agents (Claude, Cursor, Windsurf, VS Code, and more) read a standard mcpServers configuration.

  1. Open your agent's MCP configuration file.
  2. Merge the snippet below into it.
  3. Restart the agent — the "GitLab MCP Server" tools will be available.
{
  "mcpServers": {
    "gitlab": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://gitlab.com/api/v4/mcp"
      ]
    }
  }
}

Description

GitLab ships an MCP server as part of the platform itself rather than as a side project, exposed at https://<your-gitlab-host>/api/v4/mcp. Because it is built into the instance, there is no container to run and no personal access token to mint and rotate: the server advertises OAuth 2.0 Dynamic Client Registration, so a compatible client registers itself and the user authorises in a browser on first connect.

What agents can actually do with it. Twenty-three tools spanning the parts of GitLab that matter during day-to-day work. Issues and work items: create and read issues, add and read notes, and link work items with typed relationships. Merge requests: create, read, list and filter, plus commits, diffs, pipelines and threaded discussion notes. CI/CD: list pipeline jobs, pull a job's full trace when it fails, and manage pipelines (list, create, retry, cancel, delete). Discovery: instance-wide search with scope filters, label search, wiki page listing, and semantic_code_search for retrieving relevant snippets from a project. There is also attach_scan_profile for wiring a security scan profile to projects or groups.

The combination of job-log retrieval and pipeline management is the useful one in practice: an agent can notice a red pipeline, read the failing job's trace, propose a fix on the merge request, and retry the pipeline without a human relaying output between windows.

Requirements. GitLab 18.6 or later (the feature moved from experiment to beta in 18.6 and remains beta as of 19.2). Available on Free, Premium and Ultimate. The instance or group must have GitLab Duo availability set to always-on or on-by-default, beta features enabled, and MCP server access permitted. HTTP transport is recommended; mcp-remote bridges it to stdio-only clients on Node.js 20+.

Related MCP Servers

MCP: deja-vu

by Vladislav Shulcz

New

Indexes the coding-agent sessions already sitting on your disk — months of history from before you installed it — and serves them back over MCP. 3.5 GB searched in ~1.5 ms, no LLM, no embeddings.

Developer ToolsDatabases & Data
1 views
New

Microsoft's official MCP server for Azure DevOps - work items, repos, pipelines, wikis, test plans and iterations, reachable from any MCP client in plain language.

Developer Tools
1 views

MCP: Serena

by Oraios AI

New

Semantic code toolkit for coding agents — symbol-level search, editing and refactoring backed by language servers or the JetBrains platform.

Developer Tools

MCP: Hugging Face

by Hugging Face

New

Official Hugging Face MCP server: search models, datasets, Spaces and papers on the Hub, read repository files, and call Gradio apps as agent tools.

Developer ToolsWeb & Search
1 views
Browse all MCP servers →