execute_subagent · AGI- MCP: Risk & Policy

WHAT IT DOES

What execute_subagent does on AGI-MCP

AI agents invoke execute_subagent to trigger actions in AGI-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 execute_subagent needs a policy

This tool directly executes code/operations via subagents. While the description lacks detail on what specific subagents can do, the capability to execute isolated AI assistants with unspecified capabilities represents a significant risk: a misused subagent could perform harmful actions (data exfiltration, unauthorized modifications, etc.).

From the tool's definition Tool description states it will "Execute a specific subagent" and is part of a server designed for "sophisticated task management" with "agentic AI workflows." The term "Execute" combined with reference to subagents ("isolated AI assistants with specific…

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

execute_subagent Rate-limited

execute_subagent 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": {
    "execute_subagent": {
      "limits": [
        {
          "counter": "execute_subagent_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 AGI-MCP, so you can run agents against it safely.

FAQ

Questions about execute_subagent

What does the execute_subagent tool do? +

Execute a specific subagent for specialized tasks. Subagents are isolated AI assistants with specific capabilities. It is categorised as a Execute tool in the AGI-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 execute_subagent? +

Register the AGI- MCP server in PolicyLayer and add a rule for execute_subagent: 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 AGI-MCP. Nothing to install.

What risk level is execute_subagent? +

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

Can I rate-limit execute_subagent? +

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

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

execute_subagent is provided by the AGI- MCP server (muah1987/agi-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.