merge_compose_output

What merge_compose_output does on Nodebench

AI agents use merge_compose_output to create or update resources in Nodebench, usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Nodebench environment.

Why merge_compose_output is rated Medium

This tool creates or modifies data by merging artifacts together. It is reversible (can be undone or re-run with different inputs) and does not delete data or execute arbitrary code, placing it in the Write category. Severity is medium because the output composition could impact downstream processes, but the judge-gating mechanism reduces risk of unchecked misuse.

From the tool's definition Tool performs a 'merge' operation that combines 'subtask artifacts into a composed output', which is a data modification/creation action. The 'judge-gated' qualifier indicates controls are in place but does not change the fundamental Write classification.

The rule that runs merge_compose_output 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 merge_compose_output, this is the rule to start with:

{
  "version": "1",
  "default": "deny",
  "tools": {
    "merge_compose_output": {
      "limits": [
        {
          "counter": "merge_compose_output_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

Questions about merge_compose_output