hive-mind_leave · Claude Flow MCP: Risk & Policy

WHAT IT DOES

What hive-mind_leave does on Claude Flow

AI agents use hive-mind_leave to create or update resources in Claude Flow — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Claude Flow environment.

WHY IT NEEDS A POLICY

Why hive-mind_leave needs a policy

Removing an agent from the hive-mind alters the swarm topology and membership, which is a reversible configuration change (the agent is not destroyed, just disassociated). This falls under Write rather than Destructive since it does not irreversibly delete the agent itself. However, misuse could disrupt Byzantine-fault-tolerant consensus or degrade collective intelligence, giving it medium severity.

From the tool's definition 'Remove an agent from the hive-mind' — this modifies the composition of the swarm/collective by detaching an agent from it.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

hive-mind_leave Rate-limited

hive-mind_leave stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "hive-mind_leave": {
      "limits": [
        {
          "counter": "hive-mind_leave_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

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 hive-mind_leave

What does the hive-mind_leave tool do? +

Remove an agent from the hive-mind Use when native Task is wrong because you need queen-led collective intelligence — Byzantine-FT consensus, broadcast across many worker agents, shared memory with bounded conflict. For a single subagent, native Task is fine. Pair with swarm_init first to set topology. It is categorised as a Write tool in the Claude Flow MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on hive-mind_leave? +

Register the Claude Flow MCP server in PolicyLayer and add a rule for hive-mind_leave: 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 hive-mind_leave? +

hive-mind_leave is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit hive-mind_leave? +

Yes. Add a rate_limit block to the hive-mind_leave 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 hive-mind_leave completely? +

Set action: deny in the PolicyLayer policy for hive-mind_leave. 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 hive-mind_leave? +

hive-mind_leave 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.