Connect your agents to your software lifecycle with the GitLab MCP server
Blog post from GitLab
GitLab MCP server, generally available in GitLab 19.5, connects MCP-compatible AI agents to GitLab’s software development lifecycle workflows, allowing them to work with code, merge requests, work items, pipelines, documentation, and GitLab Duo Agent Platform while respecting existing user permissions. Built on the Model Context Protocol, it provides workflow-focused tools intended to reduce API endpoint guessing, unnecessary calls, and context consumption. New and consolidated capabilities include creating branches, reading repository and artifact files, semantic code search, pipeline diagnostics and history, group navigation, and optional wiki access, while configurable toolsets limit agents to relevant tool groups. Agents can also discover, start, monitor, and provide user-approved input to Duo Agent Platform flows, which run as audited CI jobs under GitLab policies. The server is available across GitLab Free, Premium, and Ultimate tiers on GitLab.com, Self-Managed, and Dedicated deployments, with setup support for clients including Claude Code, Cursor, GitHub Copilot, OpenAI Codex, Gemini, and Zed.
| Trend | Post Mentions | Total Month Mentions | Posts | Companies | MoM |
|---|---|---|---|---|---|
| MCP | 17 | 11,896 | 1,225 | 232 | +12% |
| AI Coding Assistant | 1 | 1,954 | 539 | 146 | +22% |
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.