Bring Git Intelligence to AI Chat with GitKraken MCP (new in GitLens 17.5)

[SpaceMerlin33]

GitKraken MCP: Git Intelligence for AI Chat

GitLens 17.5 bundles the new GitKraken MCP server, bringing Git and repository intelligence directly into AI chat interfaces such as VS Code, Cursor, Windsurf, and other compatible IDEs.

The GitKraken MCP transforms how you interact with Git through AI by surfacing repository information, issue tracking data, and pull request details. The same rich context GitLens has always provided is now accessible through conversational AI in your IDE.

⸻

Enhanced Developer Workflows

GitKraken MCP unlocks powerful new ways to work inside AI chat, including:

Issue and Branch Management

• “List all open Jira issues assigned to me” → MCP fetches and displays issue details from your connected trackers
• “Start work on Jira issue PROJ-123” → MCP pulls issue details, stashes changes if needed, and creates a feature branch

Pull Request Assistance

• “Create a pull request for my current branch” → MCP gathers branch diffs, generates a PR title and description, and opens the PR in your Git provider

Debug and Understand History

• “Who last modified the function calculateTotal in utils.js and why?” → MCP retrieves precise blame data, including commit details and linked issues & PRs

For more examples and use cases, visit help.gitkraken.com/mcp.

⸻

IDE Support and Setup

The MCP works seamlessly in VS Code with built-in configuration, while other IDEs like Cursor, Windsurf, and Trae require manual setup. The MCP leverages GitKraken CLI’s unified authentication system, so you authenticate once and gain access to all your connected Git providers (GitHub, GitLab, Azure DevOps, Bitbucket) and issue trackers (Jira, GitHub Issues, Linear).

Configuration setup and documentation can be found at help.gitkraken.com/mcp.

⸻

Share Your Feedback

We’re just scratching the surface of what’s possible with the GitKraken MCP. Future releases will expand its capabilities, integrations, and IDE support.

💬 We’d love to hear from you:

• What workflows are you trying with MCP?
• How’s the experience in your IDE setup?
• Have you run into any issues?

Please share your experiences, suggestions, and feedback here in this Discussion so we can continue improving MCP together.