This Bitbucket Server MCP implementation, developed for managing pull requests in Bitbucket Server environments, provides a TypeScript-based interface for AI assistants to interact with Bitbucket Server's API. Built using the Model Context Protocol SDK, it offers tools for creating, retrieving, merging, and declining pull requests, as well as adding comments and fetching diffs. The implementation focuses on streamlining PR workflows, making it easier for AI models to assist in code review processes and PR management. It's particularly useful for development teams using Bitbucket Server who want to integrate AI assistance into their PR workflows, enabling more efficient code reviews, automated PR creation, and status updates without requiring deep knowledge of Bitbucket Server's API intricacies.
Creates a new pull request. Parameters: project (string), repository (required string), title (required string), description (string), sourceBranch (required string), targetBranch (required string), reviewers (array of strings)
Retrieves detailed information about a specific pull request. Parameters: project (string), repository (required string), prId (required string)
Merges a pull request. Parameters: project (string), repository (required string), prId (required string), message (string), strategy (string: 'merge-commit', 'squash', 'fast-forward')
Declines a pull request. Parameters: project (string), repository (required string), prId (required string), message (string)
Adds a comment to a pull request. Parameters: project (string), repository (required string), prId (required string), text (required string), parentId (string)
Retrieves the diff for a pull request. Parameters: project (string), repository (required string), prId (required string), contextLines (number, default: 10)
Fetches the review history of a pull request, including approvals and reviews. Parameters: project (string), repository (required string), prId (required string)
No reviews yet. Be the first to review!
Sign in to join the conversation
Highlight AIStart the server with node to access it via any client or IDE.
node path/to/downloaded/file.mjs