subagents

What subagents does on OpenClaw

AI agents call subagents to retrieve information from OpenClaw without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.

Why subagents is rated Low

The description is ambiguous: 'list' suggests a read operation, but 'manage' could imply write or execute capabilities. Given the ambiguity, the most conservative primary interpretation is Read (listing sub-agents), but confidence is reduced due to the potential for management (write) actions. Without more detail, Read is assigned as the base category since 'list' is mentioned first and is the clearest signal.

From the tool's definition 'List or manage sub-agents' — primary verb is 'List', suggesting read/query behavior, but 'manage' implies possible write operations

The rule that runs subagents safely

PolicyLayer is an MCP gateway: it sits between your AI agents and OpenClaw, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For subagents, this is the rule to start with:

{
  "version": "1",
  "default": "deny",
  "tools": {
    "subagents": {}
  }
}

Questions about subagents