awslabsOfficial·19 Agent Skills Included

cli-agent-orchestrator

Coordinate multiple AI coding agents in parallel from one supervisor

Coordinates multiple AI coding CLIs so a supervisor agent can delegate work to specialist workers in parallel or sequence. Eliminates manual copy-paste between chat windows, lost context across sessions, and untracked long-running tasks. Adds shared memory, resumable workflows, live dashboards, and self-learning so multi-agent coding runs finish faster and more reliably.

1,188|249

↗

npx skills add awslabs/cli-agent-orchestrator --all -g -y

All Skills in This Repository (19)

Pure Emerald Level Indicators

📦 In Repo

awslabs

cao-plugin

Create and integrate custom plugins for the CLI Agent Orchestrator.

cao-provider

Develop and manage custom CLI agents for the CAO framework.

mcp-apps-builder

Create MCP Apps and add interactive ui:// views to MCP servers.

cao-worker-protocols

Define task handling and message formatting protocols for CLI Agent Orchestrator agents.

cao-memory

Store and recall facts across sessions for AI coding agents.

cao-session-management

Manage CAO sessions for parallel code review and scheduled agent runs.

cao-supervisor-protocols

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

workflow-author

Author and validate CAO workflows in a terminal session, outputting YAML.

cao-mcp-apps

Operate and extend the CAO MCP Apps host-rendered fleet UI.

add-app-to-server

Integrate interactive HTML UIs with existing MCP server tools.

migrate-oai-app

Guide migration from OpenAI Apps SDK to MCP Apps SDK with API mapping.

create-mcp-app

Create interactive MCP Apps by linking MCP tools with HTML UI resources.

Official

Intermediate

Frequently Asked Questions

FAQPage Schema

How to install CLI Agent Orchestrator?▼

Run `npx skills add awslabs/cli-agent-orchestrator --all -g -y` in your terminal to install all skills in this suite globally.

How to run multiple AI coding agents at once?▼

CAO launches a supervisor agent that delegates tasks to worker agents running in isolated terminal sessions, either in parallel with assign or sequentially with handoff.

Which AI coding CLIs does CAO support?▼

It works with Claude Code, Kiro CLI, Codex, GitHub Copilot CLI, Cursor CLI, Kimi CLI, OpenCode, and several other terminal-based AI coding tools.

Can agents remember things across sessions?▼

Yes. CAO provides shared persistent memory so facts, conventions, and lessons learned in one session are automatically available to agents in later sessions.

Do I need to write code to use CAO?▼

Basic use only requires terminal commands to launch and monitor sessions. Advanced features like custom workflows and plugins use simple Python scripts with provided templates.