Celery Flower MCP Server

Add to `~/Library/Application Support/Claude/claude_desktop_config.json`:

Local serverstdioPython

What is the Celery Flower MCP MCP server?

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:. That is what the celery flower mcp mcp server brings to an AI assistant: the same capability, reachable through the Model Context Protocol rather than a separate app or dashboard.

The short version

  • Full API coverage — — every Flower REST endpoint exposed as an MCP tool
  • Dependency injection — via dishka — clean, testable architecture
  • Pydantic Settings — — typed configuration with .env file support
  • Async — throughout — built on httpx + FastMCP
  • 65 tests — — 49 unit tests (99% coverage) + 16 integration tests against real Flower
  • Strict typing — — mypy strict mode, fully annotated

Getting it running

celery-flower-mcp on PyPI is all you need. Most clients run it directly, so configuration is a few lines and a restart.

The tools it exposes

The server publishes 14 tools. What each one is for:

  • list_workers — List all workers — optionally filter by name, refresh live stats, or get status only
  • shutdown_worker — Gracefully shut down a worker
  • restart_worker_pool — Restart a worker's process pool
  • grow_worker_pool — Add N processes to a worker's pool
  • shrink_worker_pool — Remove N processes from a worker's pool
  • autoscale_worker_pool — Configure autoscale min/max bounds
  • add_queue_consumer — Make a worker start consuming from a queue
  • cancel_queue_consumer — Make a worker stop consuming from a queue
  • list_tasks — List tasks with filters: state, worker, name, date range, search, pagination
  • list_task_types — List all registered task types across workers
  • get_task_info — Get full details for a task by UUID
  • get_task_result — Retrieve a task's result (with optional timeout)

What it needs from you

Configuration is passed through the environment: FLOWER_URL, FLOWER_API_TOKEN. Treat anything key-shaped as a real credential — scope it to the minimum the server needs, and rotate it if it ever lands in a shared config.

Things to watch

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • With 14 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Celery Flower MCP.
  • Missing credentials fail quietly in some clients — if no tools show up, check the environment block first.
  • Keep per-call confirmation enabled while you learn its behaviour; it is the cheapest safeguard you have.

How it compares

This sits in the planning and project tracking group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Celery Flower MCP's toolset — list_workers, shutdown_worker, restart_worker_pool and 11 more — is a fair guide to whether it matches your workflow. It is maintained by darius1223; worth a glance at recent repository activity before you build anything load-bearing on it.

We check each listing at SyncDev against the project's documentation before it goes live — if something here drifts out of date, it is a bug worth reporting.

Available tools

ToolWhat it does
list_workersList all workers — optionally filter by name, refresh live stats, or get status only
shutdown_workerGracefully shut down a worker
restart_worker_poolRestart a worker's process pool
grow_worker_poolAdd N processes to a worker's pool
shrink_worker_poolRemove N processes from a worker's pool
autoscale_worker_poolConfigure autoscale min/max bounds
add_queue_consumerMake a worker start consuming from a queue
cancel_queue_consumerMake a worker stop consuming from a queue
list_tasksList tasks with filters: state, worker, name, date range, search, pagination
list_task_typesList all registered task types across workers
get_task_infoGet full details for a task by UUID
get_task_resultRetrieve a task's result (with optional timeout)
apply_taskExecute a task synchronously and wait for the result
async_apply_taskDispatch a task asynchronously, returns task UUID

How to install the Celery Flower MCP MCP server

{
  "mcpServers": {
    "celery-flower": {
      "command": "uvx",
      "args": ["celery-flower-mcp"],
      "env": {
        "FLOWER_URL": "http://localhost:5555"
      }
    }
  }
}

Configuration as documented by the project. Restart the client after saving.

Configuration

VariableDescriptionRequired
FLOWER_URLEndpoint or connection string the server talks to.Yes
FLOWER_API_TOKENCredential the server authenticates with.Yes

Example prompts to try

  • Use Celery Flower MCP to list workers.
  • Use Celery Flower MCP to shutdown worker.
  • Use Celery Flower MCP to restart worker pool.

Frequently asked questions

It connects Celery Flower MCP to MCP-compatible AI assistants such as Claude and Cursor, exposing 14 tools (list_workers, shutdown_worker, restart_worker_pool, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Celery Flower MCP directly.