gitlab-mcp-server and gitlab-mcp

The second project, Alosies/gitlab-mcp-server, appears to be a reimplementation or evolution of the first, structured-world/gitlab-mcp, given they share the core "gitlab-mcp-server" concept, with the former boasting "fully-typed TypeScript" and "advanced features" for comprehensive GitLab API integration.

gitlab-mcp-server
56
Established
gitlab-mcp
45
Emerging
Maintenance 10/25
Adoption 5/25
Maturity 24/25
Community 17/25
Maintenance 13/25
Adoption 2/25
Maturity 18/25
Community 12/25
Stars: 9
Forks: 8
Downloads:
Commits (30d): 0
Language: TypeScript
License: MIT
Stars: 2
Forks: 1
Downloads:
Commits (30d): 0
Language: TypeScript
License: Apache-2.0
No risk flags
No risk flags

About gitlab-mcp-server

Alosies/gitlab-mcp-server

A fully-typed TypeScript MCP server for comprehensive GitLab API integration with Claude Desktop. Manage projects, issues, merge requests, CI/CD pipelines, and job logs with advanced features for large-scale DevOps workflows.

Implements the Model Context Protocol (MCP) as a stdio-based server, enabling Claude Desktop to execute GitLab operations through natural language queries while maintaining full TypeScript type safety across all API interactions. Supports advanced merge request workflows including threaded discussions, inline code comments, and review state management, plus self-hosted GitLab instances via configurable base URLs. Architecture uses environment-based configuration for token authentication and optional configuration files for extended setup.

About gitlab-mcp

structured-world/gitlab-mcp

Advanced GitLab MCP server

Implements 44 tools across 18 GitLab entity types (projects, MRs, pipelines, work items, wiki) using CQRS architecture with `browse_*` query and `manage_*` command patterns. Supports multi-instance deployments with per-instance OAuth 2.1, multiple transports (stdio, SSE, StreamableHTTP), and fine-grained feature flags for selective tool enablement. Integrates directly with Claude Desktop, VS Code, and other MCP clients via npm or Docker.

Scores updated daily from GitHub, PyPI, and npm data. How scores work