The Complete Production SOP: How to Deploy and Run Subagent Parallel Worker Pool Coordinator in Autonomous AI Agents

---
name: "subagent-worker-pool-coordinator"
description: "Distribute independent subtasks across a concurrent pool of specialized subagents."
metadata:
  author: "Microsoft"
  version: "1.0.0"
  runtime_compatibility:
    - "codex"
    - "claude-code"
    - "cursor"
    - "windsurf"
triggers:
  - "run parallel subagents"
  - "dispatch worker pool agents"
  - "aggregate subagent results"
permissions:
  fs_read: true
  fs_write: "workspace"
  shell_exec: false
  net_access: false
---

{
  "max_concurrent_workers": 4,
  "worker_timeout_ms": 45000,
  "fail_fast": false,
  "output_format": "json"
}

import asyncio
from typing import Dict, Any, List

async def execute_subagent_task(task_id: str, payload: Dict[str, Any]) -> Dict[str, Any]:
    # Subagent runs in isolated workspace context
    return {"task_id": task_id, "status": "success", "payload": payload}

async def dispatch_worker_pool(tasks: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
    results = await asyncio.gather(
        *[execute_subagent_task(t["id"], t) for t in tasks],
        return_exceptions=True
    )
    return results

interface WorkerResult {
  taskId: string;
  status: 'success' | 'failure';
  error?: string;
}

export function aggregateWorkerOutputs(results: WorkerResult[]): void {
  const failed = results.filter(r => r.status === 'failure');
  if (failed.length > 0) {
    throw new Error(`Worker pool failed on tasks: ${failed.map(f => f.taskId).join(', ')}`);
  }
  console.log(`Successfully completed ${results.length} parallel subagent tasks.`);
}