send_task · Celery MCP: Risk & Policy

WHAT IT DOES

What send_task does on Celery MCP

AI agents invoke send_task to trigger actions in Celery MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.

WHY IT NEEDS A POLICY

Why send_task needs a policy

Sending a task to Celery triggers execution of arbitrary registered tasks on remote workers. The effects depend entirely on what task is dispatched and its arguments, making this an Execute-category tool. Misuse could trigger destructive, financial, or otherwise harmful operations depending on what tasks are registered, hence high severity.

From the tool's definition "Send a task to the Celery queue" — dispatches work to a distributed task queue for asynchronous execution

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

send_task Rate-limited

send_task stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "send_task": {
      "limits": [
        {
          "counter": "send_task_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

RATE-LIMIT THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches Celery MCP, so you can run agents against it safely.

FAQ

Questions about send_task

What does the send_task tool do? +

Send a task to the Celery queue. It is categorised as a Execute tool in the Celery MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on send_task? +

Register the Celery MCP server in PolicyLayer and add a rule for send_task: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Celery MCP. Nothing to install.

What risk level is send_task? +

send_task is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit send_task? +

Yes. Add a rate_limit block to the send_task rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block send_task completely? +

Set action: deny in the PolicyLayer policy for send_task. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides send_task? +

send_task is provided by the Celery MCP server (joeyrubas/celery-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.