invoke_openclaw_skill

What invoke_openclaw_skill does on Nodebench

AI agents invoke invoke_openclaw_skill to trigger actions in Nodebench. 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 invoke_openclaw_skill is rated High

This tool executes arbitrary OpenClaw skills, which are external operations. While security checks are present, execution of dynamic tools with argument-dependent side effects falls under Execute rather than Read or Write. The high severity reflects that misuse could trigger unintended external operations across the OpenClaw ecosystem.

From the tool's definition 'Run an OpenClaw tool' indicates execution of external tools/operations. 'through security checks' suggests sandboxing but does not eliminate the Execute category—the tool's core function is to trigger operations whose effects depend on which OpenClaw tool is…

The rule that runs invoke_openclaw_skill safely

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

{
  "version": "1",
  "default": "deny",
  "tools": {
    "invoke_openclaw_skill": {
      "limits": [
        {
          "counter": "invoke_openclaw_skill_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

Questions about invoke_openclaw_skill