delegate_to_agent · Unlimited MCP: Risk & Policy

WHAT IT DOES

What delegate_to_agent does on Unlimited

AI agents invoke delegate_to_agent to trigger actions in Unlimited. 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 delegate_to_agent needs a policy

Based on the server context, this tool delegates work (coding and sysops tasks) to background agents. This constitutes executing operations via external agents. The description is empty, lowering confidence, but the server description and sibling tools (cancel_job, await_job, add_queue) strongly imply this triggers job execution.

From the tool's definition Tool name 'delegate_to_agent' combined with server description: 'delegates coding and sysops work to cheaper agents via durable background queues, with git worktree isolation, safety policies, and clarification rounds'

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

delegate_to_agent Rate-limited

delegate_to_agent 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": {
    "delegate_to_agent": {
      "limits": [
        {
          "counter": "delegate_to_agent_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 Unlimited, so you can run agents against it safely.

FAQ

Questions about delegate_to_agent

What does the delegate_to_agent tool do? +

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

How do I enforce a policy on delegate_to_agent? +

Register the Unlimited MCP server in PolicyLayer and add a rule for delegate_to_agent: 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 Unlimited. Nothing to install.

What risk level is delegate_to_agent? +

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

Can I rate-limit delegate_to_agent? +

Yes. Add a rate_limit block to the delegate_to_agent 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 delegate_to_agent completely? +

Set action: deny in the PolicyLayer policy for delegate_to_agent. 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 delegate_to_agent? +

delegate_to_agent is provided by the Unlimited MCP server (triumsebas/unlimited-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.