home / mcp / atlacp mcp server
Provides a Bitbucket MCP interface to manage pull requests, comments, tasks, and file content through automated prompts.
Configuration
View docs{
"mcpServers": {
"gemyago-atlacp": {
"url": "http://localhost:8080"
}
}
}Atlacp MCP Server provides an API to manage Bitbucket pull requests, comments, approvals, tasks, and related PR data through the Model Context Protocol (MCP). It exposes multiple endpoints and transports so you can integrate AI-assisted workflows directly with Bitbucket PR lifecycle actions.
You run Atlacp on your machine or in a container and connect your MCP client to it. The server exposes a HTTP transport by default, with an optional SSE transport at a dedicated URL. You can also run it locally as a stdio MCP process if you prefer to start it as a local, in-process server.
From your MCP client, point to the Atlacp server by using its root URL (for Streamable HTTP) or by appending /sse for the SSE transport. Example destinations are http://localhost:8080 for HTTP and http://localhost:8080/sse for SSE. When using the local, containerized setup, you will typically reach the service on port 8080 of the host.
In your MCP client configuration, define the Atlacp MCP server with the URL for HTTP transport and any local stdio configuration if applicable. You can prepare a configuration file under the mcpServers section to enable communication with Atlacp, then send prompts to interact with Bitbucket PRs, such as creating PRs, commenting on PRs, or updating tasks.
Prerequisites: you need Docker installed on your machine to run Atlacp as a container. If you prefer to run it directly, you can use a local MCP setup with a suitable runtime command.
Step 1: Create the accounts configuration file for Bitbucket access. Use the example as a starting point and replace the token value with your actual Bitbucket token. Save it as atlassian-accounts.json.
Step 2: Start the Atlacp MCP server using Docker. Run the command below to start the container, bind port 8080, and mount your accounts file into the container.
docker run -d --name atlacp-mcp \
--restart=always \
-p 8080:8080 \
-v $(pwd)/atlassian-accounts.json:/app/atlassian-accounts.json \
ghcr.io/gemyago/atlacp-mcp:latest \
-a /app/atlassian-accounts.json \
httpYou can connect to Atlacp via the default HTTP transport or use the SSE transport by appending /sse to the root URL. STDIO transport is supported as well, which runs Atlacp as a local process starting from a command such as docker run in a containerized environment.
A JSON example for configuring MCP servers shows how to declare an Atlacp server with an HTTP URL. You can add this to your MCP client configuration to enable communication with Atlacp.
Keep your Bitbucket personal access token secure. Store tokens and account details in a secure, access-controlled location. When running in containers, ensure network access is restricted to trusted clients and that port 8080 is exposed only to trusted hosts.
If you cannot reach the server at http://localhost:8080, verify that the Docker container is running and that port 8080 is mapped correctly. Check the accounts file path inside the container and ensure atlassian-accounts.json is readable by the container process.
json
{
"mcpServers": {
"Atlassian MCP": {
"url": "http://localhost:8080"
}
}
}Root URL serves the Streamable HTTP transport. Append /sse to use the SSE transport. STDIO is supported as well; it uses a local command to run the MCP server inside your environment.
Add a comment to a Bitbucket pull request, enabling you to annotate PR conversations via MCP.
Approve a pull request in Bitbucket through an MCP command.
Create a new pull request in a Bitbucket repository.
Create a task on a Bitbucket pull request to track follow-up work.
Retrieve the content of a file within a pull request for inspection or verification.
Get the diff for a specific pull request to review changes.
Obtain a diffstat for a pull request to summarize changes by file.
List tasks associated with a pull request to monitor progress.
Merge a pull request in Bitbucket after review.
Read and fetch details about a pull request.
Request changes on a pull request to trigger feedback and updates.
Update a pull request with new information or edits.
Update an existing task on a pull request to reflect current status.