execute-workflow

What execute-workflow does on Claude Flow

AI agents invoke execute-workflow 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 execute-workflow is rated High

Workflows can contain arbitrary operations (queries, API calls, state changes, integrations). Executing a workflow is functionally equivalent to running code or scripts. The blast radius depends on what the workflow contains, making this high severity. However, without evidence of irreversible deletion (Destructive) or financial transactions (Financial), Execute is the most precise category.

From the tool's definition Tool name is 'execute-workflow' and description states 'Execute a workflow'. The verb 'Execute' combined with context of running workflows in an orchestration system indicates this triggers external operations whose effects depend on arguments.

The rule that runs execute-workflow safely

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

{
  "version": "1",
  "default": "deny",
  "tools": {
    "execute-workflow": {
      "limits": [
        {
          "counter": "execute-workflow_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

Questions about execute-workflow