mcp-name: io.github.csgear/bitbucket-agent

Bitbucket PR Agent (MCP Server)

This is an MCP (Model Context Protocol) server that allows AI agents to create Pull Requests in a self-hosted Bitbucket Server instance.

Setup

1. Install Dependencies:

   pip install -r requirements.txt

2. Configuration: Copy .env.example to .env and fill in your details:

   cp .env.example .env

   Edit .env with your Bitbucket URL, credentials, project key, and repository slug.

Related MCP server: Bitbucket MCP Server

running the Server

You can test the server locally by running:

python server.py

(Note: Standard MCP servers communicate over stdio, so running this directly in a terminal will just wait for input).

Integration with MCP Clients (e.g., Claude Desktop, VS Code with MCP)

To use this with an MCP-compliant client, you need to configure it to run this python script.

Example claude_desktop_config.json:

{
  "mcpServers": {
    "bitbucket-agent": {
      "command": "python",
      "args": ["/absolute/path/to/bitbucket/server.py"],
      "env": {
        "BITBUCKET_URL": "https://bitbucket.yourcompany.com",
        "BITBUCKET_USERNAME": "your_username",
        "BITBUCKET_PASSWORD": "your_password",
        "BITBUCKET_PROJECT_KEY": "PROJ",
        "BITBUCKET_REPO_SLUG": "repo"
      }
    }
  }
}

Note: You can either rely on the .env file loading (if the working directory is correct) or pass the environment variables directly in the config as shown above.

VS Code Configuration

To use this agent with GitHub Copilot (or other MCP-enabled VS Code extensions), you typically need to add the server configuration to your VS Code settings.

1. Open your Workspace Settings (.vscode/settings.json) or User Settings.

2. Add the definition for this MCP server. For example:

"github.copilot.mcpServers": {
    "bitbucket-agent": {
        "command": "python",
        "args": [
            "${workspaceFolder}/server.py"
        ],
        "env": {
             "BITBUCKET_URL": "https://bitbucket.yourcompany.com",
             "BITBUCKET_USERNAME": "your_username",
             "BITBUCKET_PASSWORD": "your_password",
             "BITBUCKET_PROJECT_KEY": "PROJ",
             "BITBUCKET_REPO_SLUG": "repo"
        }
    }
}

Note: Adjust the ${workspaceFolder}/server.py path if your script is located elsewhere. You may need to use an absolute path if ${workspaceFolder} is not resolved correctly by your specific extension version.