Owayo Gitlab MCP Server

🚀 GitLab-MCP-Server

This is a Model Context Protocol (MCP) server that provides integration capabilities with GitLab. It retrieves pipeline failure information and issues pointed out in merge requests from specific GitLab projects and provides them to the AI assistant.

🚀 Quick Start

This MCP server uses the GitLab API to provide the following information to the AI assistant:

1. Console output of jobs that failed in GitLab pipelines
2. Unresolved issues (comments) in GitLab merge requests (MRs)
3. Changes in GitLab MRs (differences from the current state of the local repository)

By using the MCP functionality, the AI assistant can directly obtain GitLab information and provide more accurate support.

📦 Installation

# Install uv
$ curl -LsSf https://astral.sh/uv/install.sh | sh

$ cd /path/to/this-mcp-server
# Install libraries
$ uv sync

⚙️ Preparation

You need a GitLab access token. Generate the access token in GitLab's settings → Access Tokens. When generating it, make sure to check read_api.

✨ Features

1. Retrieve and fix pipeline failure information (get_pipeline_failed_jobs)

Retrieve the console output of jobs that failed in GitLab pipelines. Based on the retrieved information, the AI assistant will perform fixes.

Output:

• Console output of failed jobs (including job name, status, and detailed logs)

2. Retrieve and address issues in MRs (get_review_comments)

Retrieve and address unresolved issues (comments) in GitLab MRs. Resolved comments and comments not associated with files are excluded.

Output:

• Unresolved issues associated with files in MRs (including commenter, time, comment content, file location information, etc.)

3. Retrieve and review changes in MRs (get_review_changes)

Retrieve the differences from the base commit (base_sha) of a GitLab MR to the current state of the local repository.

💻 Usage Examples

Basic Usage

1. Retrieve console output of failed jobs

from mcp_server import get_pipeline_failed_jobs

jobs = get_pipeline_failed_jobs()
for job in jobs:
    print(f"Job Name: {job['name']}")
    print(f"Status: {job['status']}")
    print(f"Log: {job['log']}\n")

2. Retrieve issues in MRs

from mcp_server import get_review_comments

comments = get_review_comments()
for comment in comments:
    print(f"Author: {comment['author']}")
    print(f"Time: {comment['time']}")
    print(f"Content: {comment['content']}\n")

3. Retrieve changes in MRs

from mcp_server import get_review_changes

changes = get_review_changes()
for change in changes:
    print(f"File: {change['file']}")
    print(f"Type: {change['type']}")
    print(f"Differences: {change['differences']}\n")

⚠️ Important Note

• In the above code examples, replace the following values appropriately:
  • your_gitlab_api_key: GitLab API access token
  • /path/to/git/repo: Absolute path of the local Git repository
  • /path/to/this-mcp-server: Absolute path of the directory of this MCP server