home / mcp / celery mcp server

Celery MCP Server

Exposes Celery task queues over MCP for easy integration with automated clients and LLMs.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "joeyrubas-celery-mcp": {
      "command": "celery-mcp-server",
      "args": []
    }
  }
}

You can run Celery tasks through MCP by exposing Celery functionality as an MCP server. This lets you manage queues, submit tasks, monitor status, and inspect workers from an automated client or an LLM, enabling scalable task orchestration and integration with other systems.

How to use

You interact with the Celery MCP server by starting the server locally and then issuing tool-based requests from your MCP client or automation layer. The server exposes a set of tools that let you initialize a Celery connection, list tasks, send tasks, check task status, manage active and scheduled tasks, revoke tasks, and retrieve worker statistics. Use these tools to integrate Celery queues into your workflows and to monitor and control long-running tasks from an automated agent.

How to install

Prerequisites you need on your machine are Python and a compatible package manager. You also need a running Celery broker (for example Redis) accessible from your environment.

pip install celery-mcp
```

```
git clone https://github.com/yourusername/celery-mcp.git
cd celery-mcp
pip install -e .

Starting and using the server

Start the MCP server to expose Celery functionality to your MCP clients. Once the server is running, you can call the available tools to interact with Celery.

celery-mcp-server

Available tools

initialize_celery_connection

Initialize a connection to the Celery broker from the MCP server to establish communication with the Celery backend.

list_registered_tasks

Return a list of all registered Celery task names available in the current broker.

send_task

Submit a Celery task to the queue with specified name and arguments.

get_task_status

Query the status of a Celery task by its task ID.

get_active_tasks

Retrieve information about Celery tasks that are currently running.

get_scheduled_tasks

Retrieve information about tasks that are scheduled for future execution.

revoke_task

Cancel or revoke a Celery task by its ID.

get_worker_stats

Get statistics about Celery workers, including active and total worker counts.