# Welcome to the GitLab MCP server tools (Beta) feedback issue The purpose of this feedback issue is to collect your experiences with the GitLab MCP server tools. Our goal is to understand what's working, what's broken, and what's missing so we can continue to mature MCP server tools toward GA. <details> <summary>What are MCP server tools?</summary> GitLab's MCP server exposes GitLab capabilities as tools that external AI assistants can use. When you connect an AI tool like Claude Desktop or Cursor to GitLab via MCP, these tools are what the AI uses under the hood to read issues, interact with merge requests, manage pipelines, and more. This is one direction of GitLab's MCP implementation — **external AI → GitLab**. The AI assistant is the client, and GitLab is the server. For the full list of current tools see [GitLab MCP server tools docs](https://docs.gitlab.com/user/gitlab_duo/model_context_protocol/mcp_server_tools/). </details> <details> <summary>Building agents or flows in DAP? MCP server tools affect you too!</summary> Did you know DAP uses the MCP server tools internally? This means that the same tools that external AI assistants use are also what power features like foundational agents too. Feedback from internal builders is just as valuable as feedback from external users. If you're hitting a gap while building an agent or flow, a missing tool, unexpected behavior, or data that isn't returning correctly, we want to hear about it here. </details> ## :dart: Feedback we're especially interested in We are actively working to expand tool coverage. We need your help identifying: 1. **Tool gaps:** What workflows are you trying to automate that we don't have tools for yet? 1. **Reliability issues:** Are tools returning errors, incomplete data, or behaving unexpectedly? 1. **Client compatibility:** Which AI tools are you using, and are there compatibility issues? ## :pencil: How to give feedback 1. **Check existing feedback:** Review threads below to see if your issue is already reported. Add a :thumbsup: or comment to show support. 2. **Start a new thread:** Use a descriptive title so others can quickly understand the focus of your thread. 3. **Include context:** Tell us what you were trying to do, which AI tool you were using, and what happened. Screenshots or error messages are especially helpful. ## :handshake: What you can expect from us 1. We **will read** all feedback while this issue remains open 2. We **will prioritize** fixes and new tools based on feedback patterns 3. We **will create issues** for reproducible bugs 4. We **may reach out** for clarification on complex issues or to invite you into design partnership conversations --- :clap: Thank you for helping us build a better MCP experience. Your feedback directly shapes what we build next.