Launch Sub Agent

This is a structured approach to spawning focused sub-agents with automatic model selection. It walks through task analysis, picks between Opus/Sonnet/Haiku based on complexity and output size, optionally loads specialized agent prompts for domains like development or research, then wraps everything in mandatory chain-of-thought reasoning at the start and self-critique verification at the end. The whole point is context isolation so each sub-agent gets a clean window instead of your accumulated conversation history. It's thorough to the point of being prescriptive, with decision trees and verification tables baked in. If you're orchestrating multi-agent workflows and want guardrails around model choice and quality checks, this gives you a repeatable pattern.

skills.sh installs: 1.3K
GitHub stars: 1.8K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add neolabhq/context-engineering-kit --skill launch-sub-agent --agent claude-code

Installs into .claude/skills of the current project.

Files

launch-sub-agent

Launch a focused sub-agent to execute the provided task. Analyze the task to intelligently select the optimal model and agent configuration, then dispatch a sub-agent with Zero-shot Chain-of-Thought reasoning at the beginning and mandatory self-critique verification at the end. This command implements the **Supervisor/Orchestrator pattern** from multi-agent architectures where you (the orchestrator) dispatch focused sub-agents with isolated context. The primary benefit is **context isolation** - each sub-agent operates in a clean context window focused on its specific task without accumulated context pollution.

Process

Phase 1: Task Analysis with Zero-shot CoT

Before dispatching, analyze the task systematically. Think through step by step:

Let me analyze this task step by step to determine the optimal configuration:

SKILL.md on GitHub

Details

Repository: github.com/neolabhq/context-engineering-kit
Category: AI & Agent Building
First seen: Jun 3, 2026