orchestrator

Route AI tasks to Haiku, Sonnet, Opus, or Codex by complexity.

147|10|Updated Nov 17, 2025
One-click install
npx skills add https://github.com/GGPrompts/TabzChrome --skill orchestrator-ggprompts
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: orchestrator
Source: https://github.com/GGPrompts/TabzChrome/tree/main/plugins/conductor/skills/orchestrator
Command: npx skills add https://github.com/GGPrompts/TabzChrome --skill orchestrator-ggprompts

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill orchestrates multi-model AI work across Haiku, Sonnet, Opus, and Codex based on task complexity.

Core Features & Use Cases

  • Task-based model routing: automatically selects Haiku for simple tasks, Sonnet for medium complexity, and Opus for advanced feature work.
  • Deterministic orchestration: uses structured spawn commands and profiles to ensure reproducible results.
  • Workflow integration: supports common tasks like code reviews, repository operations, and architectural planning.

Quick Start

To begin, describe the task and its complexity; the orchestrator will spawn the appropriate workers and manage the lifecycle using Task, tabz_spawn_profile, and Bash commands.

Frequently Asked Questions about orchestrator

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

FAQPage Schema
How do I orchestrate multi-model AI work based on task complexity?

Multi-model AI orchestration routes work across Haiku, Sonnet, Opus, and Codex based on task complexity. It uses tiered tasks and deterministic spawn commands like Task and Bash to structure work across agents for reproducible results.

Can I use AI orchestration for simple parallel exploration and complex feature development?

AI orchestration supports simple parallel exploration, code reviews, and complex feature development. It automatically selects Haiku for simple tasks, Sonnet for medium complexity, and Opus for advanced architectural planning workflows.

What is the best way to manage worker lifecycle during AI code reviews?

Managing worker lifecycle during AI code reviews requires deterministic spawn commands. The orchestrator uses structured profiles like tabz_spawn_profile and Bash commands to spawn appropriate workers and ensure reproducible results across agents.

How do I ensure reproducible results when orchestrating multi-model AI agents?

Reproducible multi-model AI orchestration relies on deterministic spawn commands and structured profiles. By applying tiered tasks and using commands like Task, tabz_spawn_profile, and Bash, the workflow guides model choice and ensures consistent agent execution.

Does multi-model AI orchestration require specific dependencies for repository operations?

Multi-model AI orchestration requires no external dependencies to manage repository operations. It integrates deterministic spawn commands and tiered task structures directly to guide model choice and spawn strategy across Haiku, Sonnet, Opus, and Codex.