CLI Agent Orchestrator is a tool that coordinates multiple AI coding command-line programs by running them as separate workers in isolated terminal sessions. A supervisor uses it to delegate software tasks to specialist agents in parallel or in sequence while the workers retain their normal command-line capabilities. The catalogue skills operate this orchestration workflow.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add awslabs/cli-agent-orchestrator --skill cao-worker-protocolsgit clone --depth 1 https://github.com/awslabs/cli-agent-orchestratorWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/awslabs/cli-agent-orchestrator/cao-worker-protocols)<a href="https://agentmods.dev/skills/awslabs/cli-agent-orchestrator/cao-worker-protocols"><img src="https://agentmods.dev/badge/skills/awslabs/cli-agent-orchestrator/cao-worker-protocols.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00021 | $0.00641 |
| Opus 5 | $0.00010 | $0.00320 |
| Sonnet 5 | $0.00004 | $0.00128 |
| Haiku 4.5 | $0.00002 | $0.00064 |
Grade A, and why
cao-worker-protocols scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CAO Worker Protocols
Use this skill when acting as a worker agent inside CLI Agent Orchestrator.
This skill explains how workers should interpret assigned versus handed-off work, when to call send_message, and how to report results back cleanly.
Understand the Dispatch Mode
Workers receive tasks through one of two orchestration modes:
handoff: blocking work where the orchestrator captures your final output automaticallyassign: non-blocking work where you must actively return results to the requesting terminal
Depending on provider and CAO behavior, a handoff may be made explicit in the task text. For example, Codex workers currently receive a [CAO Handoff] prefix for blocking handoffs. Other providers may rely on the task wording and orchestration context instead.
Rules for Handoff Tasks
When the task is a blocking handoff, complete the work and present the result in your normal response. The orchestrator captures that response automatically.
Do not call send_message for ordinary handoff completion unless the task explicitly asks for additional side-channel communication.
Rules for Assigned Tasks
When the task came through assign, send your results back after you finish the work:
- Format the result clearly and concisely.
- Call
send_message(message=...)— omittingreceiver_idroutes the result to the terminal that assigned the task (the recorded caller). This is the reliable default. - If the task message names a different callback terminal (directly or in an appended suffix such as
[Assigned by terminal ...]), pass that ID asreceiver_idinstead.
Do not stop after writing a normal response if the assignment explicitly requires a callback. The requesting terminal depends on send_message to receive the result.
Your own CAO_TERMINAL_ID identifies your terminal, not the callback target. Never pass it as receiver_id.
Message Formatting
Return results that are easy for the supervisor to merge into a larger workflow:
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 59 lines · 21 tokens per session scan A 00df779ac8aa
cao-worker-protocols is a skill published in the GitHub repository awslabs/cli-agent-orchestrator (1,213 stars, last pushed yesterday), licensed Apache-2.0. It adds 21 tokens to every session and 641 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
tool-selection
Internal guidance for choosing between agent-rack's sync and background execution tools, and when to use per-agent shortcuts. Use whenever delegating a task to a claude/codex/opencode/Antigravity/custom sub-agent through agent-rack.
review
Run a structured, read-only code review through agent-rack's agentreview tool.
session-send
Send follow-up input to a running background agent-rack session.
review-handling
Internal guidance for presenting agent-rack's agentreview output back to the user. Use whenever an agentreview call (foreground or via agentsessionstatus) returns a result.
run
Run a one-shot task synchronously with a named CLI sub-agent via agent-rack.
session-logs
Read the raw stdout/stderr event stream from a background agent-rack session.