cao-supervisor-protocols

Orchestrate CAO worker agents using assign, handoff, and send_message primitives.

1.0k|202|Updated Jul 29, 2025
One-click install
npx skills add https://github.com/awslabs/cli-agent-orchestrator --skill cao-supervisor-protocols
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cao-supervisor-protocols
Source: https://github.com/awslabs/cli-agent-orchestrator/tree/main/skills/cao-supervisor-protocols
Command: npx skills add https://github.com/awslabs/cli-agent-orchestrator --skill cao-supervisor-protocols

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires CAO, MCP, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the challenges of orchestrating worker agents in the CAO framework, ensuring efficient task distribution and coordination.

Core Features & Use Cases

  • Assign vs. Handoff Decision Making: Guides on choosing between asynchronous assign and synchronous handoff for worker tasks.
  • Idle-Based Message Delivery: Ensures efficient delivery of worker results when the supervisor is idle.
  • Callback Pattern: Automates callback routing for worker results without manual intervention.
  • Direct Supervisor Communication: Enables direct communication with worker terminals for status updates and follow-ups.
  • Interactive Worker Prompts: Handles user prompts from Hermes workers effectively.
  • Practical Workflow: Provides a step-by-step guide for using the skill in real-world scenarios.

Quick Start

Launch the skill with the command: 'cao run cao-supervisor-protocols --assign "Analyze the dataset A."'

Frequently Asked Questions about cao-supervisor-protocols

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

FAQPage Schema
How do I distribute tasks to worker agents in a CAO multi-agent workflow?

Distribute tasks in a CAO multi-agent workflow using assign and handoff primitives, choosing assign for asynchronous execution and handoff for synchronous worker responses.

When should I use assign versus handoff for worker agent orchestration in CAO?

Use assign for asynchronous task distribution when the supervisor can continue other work, and use handoff for synchronous execution when the supervisor must wait for the worker's result.

How does the callback pattern handle worker results in CAO supervisor protocols?

The callback pattern automates routing of worker results back to the supervisor, ensuring efficient idle-based message delivery without manual intervention when orchestrating CAO agents.

Can I send direct messages to worker terminals for status updates in CAO?

Yes, direct supervisor communication uses the send_message primitive to enable direct contact with worker terminals for status updates and follow-ups within the CAO framework.

Do I need the MCP framework to orchestrate worker agents with CAO supervisor protocols?

Yes, orchestrating worker agents with CAO supervisor protocols requires both the CAO framework and MCP primitives to enable task distribution and multi-agent coordination.

How do I start a worker agent task from the command line in CAO?

Start a worker agent task from the command line in CAO by running 'cao run cao-supervisor-protocols --assign' followed by your task description to initiate asynchronous execution.