Arlet
GitLab

Use GitLab from your AI

Search projects, read repositories, manage merge requests, issues and work items, and inspect CI pipelines and jobs on GitLab.com through GitLab's official MCP server. Connect by signing in.

Arlet makes GitLab's official MCP server available from the AI client you already use. Signing in to GitLab.com completes the connection.

The AI can read projects and repositories, review and comment on merge requests, create and update issues, and inspect or rerun CI pipelines and jobs. The available tools are those provided by GitLab's official server, which GitLab offers as a beta.

Choose your AI client

Pick the AI client you use to see how to connect it with GitLab.

What you can do

Tools the AI can call through the GitLab integration.

  • search

    Search across GitLab.com by keyword.

  • list_projects

    List projects.

  • get_project

    Get a project's details.

  • get_repository_file

    Read a file from a repository.

  • list_repository_tree

    List a repository's file tree.

  • list_commits

    List commits.

  • list_merge_requests

    List merge requests.

  • get_merge_request

    Get a merge request's details.

  • get_merge_request_diffs

    Get a merge request's diffs.

  • save_merge_request

    Create or update a merge request.

  • save_merge_request_review

    Post a review on a merge request.

  • save_note

    Comment on a merge request or issue.

  • list_work_items

    List issues, tasks and epics.

  • save_work_item

    Create or update an issue, task or epic.

  • list_pipelines

    List CI pipelines.

  • get_job

    Get a CI job's details and log.

  • save_pipeline

    Run, retry or cancel a pipeline.

The tool list follows what GitLab's official MCP server provides and may change.

Example uses

  • Review a merge request

    Read the diff, point out problems and post them as review comments.

  • Debug a failed pipeline

    Read the failing job's log and summarize the cause and a fix.

  • File issues

    Turn meeting notes into issues with labels and assignees.

  • Explore the code

    Search across repository files and explain where something is implemented.

Prompts you can use as-is

  • Review the diff of merge request !128 and post your concerns as comments
  • Read the job log of the last failed pipeline on main and explain the cause
  • Create issues in the backend project for the action items in these notes, with assignees
  • Find where payment retries are implemented in the repository and explain it