spawn-child-agent

Spawn independent Claude Code agent processes for isolated sub-tasks.

Updated Jan 19, 2026
One-click install
npx skills add https://github.com/mrjuguy/kitchen_assistant --skill spawn-child-agent
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spawn-child-agent
Source: https://github.com/mrjuguy/kitchen_assistant/tree/main/.agent/skills/spawn-child-agent
Command: npx skills add https://github.com/mrjuguy/kitchen_assistant --skill spawn-child-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Spawns independent, self-contained Claude Code agent processes to handle complex sub-tasks without interrupting the main workflow.

Core Features & Use Cases

  • Spawns a new main-session agent to tackle a complex sub-task with isolation.
  • Enables delegated sub-agents (e.g., QA bot, Researcher) while allowing parent supervision and result integration.
  • Provides headless operation with strict tool permissions and a takeover/merger workflow for returned outputs.

Quick Start

Spawn a headless child agent to perform the sub-task and return its result to the parent.

Frequently Asked Questions about spawn-child-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I delegate a complex sub-task to an isolated agent process?

To delegate a complex sub-task to an isolated agent, you can spawn a new, independent Claude Code agent process. This allows the sub-task to run headlessly without disrupting your main workflow.

Can I run a headless sub-agent with a predefined tool set for QA automation?

Yes, you can run a headless sub-agent for QA automation by spawning an independent process with a predefined, pre-authorized tool set. This ensures strict isolation and controlled execution for specialized tasks.

What is the best way to integrate results from a delegated sub-agent into a parent session?

The best way to integrate sub-agent results is through a predefined handoff mechanism that allows parent supervision. The spawned agent completes its task and merges the returned outputs back into the parent session.

When do I need to spawn a child agent for task delegation instead of processing inline?

You need to spawn a child agent when task delegation requires isolation or when running specialized sub-tasks like QA or research. This prevents the sub-task from interrupting or disrupting your main workflow.

Does spawning a headless child agent disrupt the main workflow execution?

No, spawning a headless child agent does not disrupt the main workflow. The process operates independently and in isolation, allowing the parent session to continue execution while the sub-task is handled.

What are the limitations of using isolated sub-agents for task delegation?

Isolated sub-agents require strict tool permissions and headless execution, meaning they cannot interactively prompt the user. You must pre-authorize the tool set and define a clear handoff mechanism before execution.