run_delegation · Delegations MCP: Risk & Policy

WHAT IT DOES

What run_delegation does on Delegations

AI agents invoke run_delegation to trigger actions in Delegations. 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 run_delegation needs a policy

This tool executes arbitrary delegations/tasks via the specified library:name argument. The effects are entirely dependent on the delegation's content and what sub-agents do during execution. While bounded, delegations can perform any action the sub-agents are capable of (API calls, file operations, external service interactions, etc.).

From the tool's definition Tool description states 'Run a delegation' and returns transcript_path, indicating execution of tasks. Server description emphasizes 'execution of self-contained, bounded tasks' and 'orchestrate tasks between a primary LLM and specialized sub-agents',…

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

run_delegation Rate-limited

run_delegation 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": {
    "run_delegation": {
      "limits": [
        {
          "counter": "run_delegation_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 Delegations, so you can run agents against it safely.

FAQ

Questions about run_delegation

What does the run_delegation tool do? +

Run a delegation by its library:name. Returns summary, prompt_path, transcript_path. It is categorised as a Execute tool in the Delegations MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on run_delegation? +

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

What risk level is run_delegation? +

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

Can I rate-limit run_delegation? +

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

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

run_delegation is provided by the Delegations MCP server (kaijfox/delegations-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.