home / mcp / bitbucket mcp server

Bitbucket MCP Server

A Python-based MCP server to manage Bitbucket Pull Requests via commands and scripted actions.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "acendas-bitbucket-mcp": {
      "command": "python",
      "args": [
        "/path/to/bitbucket-mcp/server.py"
      ],
      "env": {
        "BITBUCKET_USERNAME": "[email protected]",
        "BITBUCKET_API_TOKEN": "YOUR_API_TOKEN",
        "BITBUCKET_WORKSPACE": "your-workspace"
      }
    }
  }
}

You run a FastMCP server to manage Bitbucket Cloud Pull Requests directly from Claude. It lets you create, view, update, and review PRs, fetch workspace members for reviewers, and add comments, all from a lightweight, scriptable interface.

How to use

You interact with the Bitbucket MCP server by running it locally and connecting to it from your MCP client. The server exposes a set of actions to create, list, view, update, and review pull requests, and to manage reviewers and comments. Use the functions to automate common PR workflows, such as creating a PR with default reviewers, approving or requesting changes on PRs, and adding notes.

How to install

Prerequisites: you need Python and git installed on your system.

git clone https://github.com/Acendas/bitbucket-mcp.git
cd bitbucket-mcp
pip install -r requirements.txt

Alternatively, you can run the installer with the UV runtime along with the Python environment.

git clone https://github.com/Acendas/bitbucket-mcp.git
cd bitbucket-mcp
uv pip install -r requirements.txt

Configuration and usage notes

You configure access to Bitbucket using either an interactive setup or environment variables. The interactive flow stores credentials securely on your machine for future use.

Interactive setup example, to configure credentials and workspace:

setup_bitbucket(
    workspace="your-workspace",
    username="[email protected]",
    api_token="your-api-token"
)

Claude Desktop configuration

Configure the MCP server entries in your Claude Desktop configuration so Claude can start and communicate with the local server.

{
  "mcpServers": {
    "bitbucket": {
      "command": "python",
      "args": ["/path/to/bitbucket-mcp/server.py"]
    }
  }
}

Or, if you run through UV, use this form to point Claude to the local server directory and start with Python inside the UV context.

{
  "mcpServers": {
    "bitbucket": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/bitbucket-mcp", "python", "server.py"]
    }
  }
}

Available tools

create_pull_request

Create a new PR in a repository, with optional default reviewers and settings.

list_pull_requests

List PRs for a repository with optional state filtering and pagination.

get_pull_request

Fetch details of a specific PR by its ID.

update_pull_request

Update PR metadata such as title, description, destination branch, and reviewers.

approve_pull_request

Approve a PR as the current user.

unapprove_pull_request

Remove your approval from a PR.

request_changes_pull_request

Request changes on a PR.

add_pull_request_comment

Add a markdown-formatted comment to a PR.

list_workspace_members

List members of a workspace to identify potential reviewers.

get_default_reviewers

Retrieve default reviewers configured for a repository.