agent_spawn · Claude Flow MCP: Risk & Policy

WHAT IT DOES

What agent_spawn does on Claude Flow

AI agents invoke agent_spawn to trigger actions in Claude Flow. 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 agent_spawn needs a policy

Spawning a subagent initiates execution of a new autonomous process or agent instance. This is an Execute-category action because it triggers external operations (a new agent running code/tasks) whose effects depend on arguments passed. The blast radius is high because a misused spawn could launch uncontrolled agents performing arbitrary actions within the orchestration system.

From the tool's definition "Spawn a subagent" — creates and launches a new autonomous agent process

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

agent_spawn Rate-limited

agent_spawn 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": {
    "agent_spawn": {
      "limits": [
        {
          "counter": "agent_spawn_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 Claude Flow, so you can run agents against it safely.

FAQ

Questions about agent_spawn

What does the agent_spawn tool do? +

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

How do I enforce a policy on agent_spawn? +

Register the Claude Flow MCP server in PolicyLayer and add a rule for agent_spawn: 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 Claude Flow. Nothing to install.

What risk level is agent_spawn? +

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

Can I rate-limit agent_spawn? +

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

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

agent_spawn is provided by the Claude Flow MCP server (claude-flow). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.