菜单
Use for any agent that communicates with other agents or the user via a chat channel. Defines the reply/check_messages tool mechanics, @-addressing, and the conversational posture — engaged, concise, patient, and grounded.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use for any container agent that posts progress to the coordination server's message board. Defines the curl format, channels, message types, payload conventions, posting responsibilities, and verbosity levels.
Use when a review agent must produce structured output. Defines the BLOCKING/NOTE/Summary/Verdict template, the two-tier confidence rubric, and the JSON shadow block consumed by POST /tasks/:id/reviews. Compose with review-process and a domain skill.
Use when an agent evaluates the output of other agents or decides whether work meets the bar. Enforces rigorous quality standards — reject shortcuts, demand evidence, no rubber-stamping.
Use when an agent is the FSM arbitrator on a durable task. Defines the two triggers, the three ruling types, the POST contract, the addendum-file rules for contradiction rulings, and the 409 handling. Project-agnostic.
Tells container agents where the Unreal Engine source tree is mounted inside the container, so they can grep and read engine headers to verify API signatures and class hierarchies.
Use when an agent must review code changes. Defines the universal review sequence — identify files, read context, check against domain criteria, score, filter. Compose with a domain skill and review-output-schema.
| name | chat-etiquette |
| description | Use for any agent that communicates with other agents or the user via a chat channel. Defines the reply/check_messages tool mechanics, @-addressing, and the conversational posture — engaged, concise, patient, and grounded. |
| axis | protocol |
Rules and posture for communicating in a multi-agent chat channel.
EVERY message you want others to see MUST go through the reply tool. Text you write outside of tool calls is invisible to other agents — it goes to your local log, not to the chat channel. If you want to say something, call reply. There is no other way to communicate.
Call check_messages with your room ID to read the conversation. It returns ALL messages since your last reply as a structured chat log. If there are no new messages, it returns "No unread messages."
You will receive channel notifications when new messages arrive, but these are just alerts — always call check_messages to read the actual conversation in context.
@your-agent-name), reply as soon as you are able — even if only to say you need a moment.@agent-name. To address everyone, use @everyone.@agent-1 @agent-2).Be engaged, concise, patient, and grounded.
This is a working conversation, not a status report. No preamble, no summaries of what you already said. If a message from user asks you to change approach, prioritize it — user messages are directives.