Standing context for Slack sibling agents — one room per team, creator-posted introductions, bot-to-bot hop discipline. Ships as instructions.md composed into t
复制下面这句话,粘贴给 Claude Code、Codex、Cursor 等 AI 编程工具,它会读取安装说明并在你确认后完成安装。
请阅读 https://ai.atlankj.com/install/asset/gh-slack-construct-agents-886dcbd335ee ,按照其中的说明把「slack-construct-agents」安装到你(当前 AI 工具)中。执行前先告诉我将运行的命令和写入的位置,等我确认。
查看 AI 将读取的安装说明正在读取 GitHub 原文…
内容来自 GitHub 原始文件,由原作者维护。在 GitHub 查看
This skill is a carrier for standing context, not an on-demand workflow. Its
payload is instructions.md in this directory: the rules an agent needs when
it shares Slack with sibling agents it can create (create_agent) and room
with (create_room / add_to_room) — team-room shape, who posts the
introduction, and the bot-to-bot self-limit.
The host composes every container skill's instructions.md into each group's
CLAUDE.md at spawn (src/project-doc-compose.ts), so if you are reading this
from inside a session, those rules are already part of your standing
instructions. There is nothing further to load or run here.
The room-and-canvas half of the Slack standing context (mention engagement,
canvas discipline, access rules) ships separately with the Slack channel
payload as the slack-construct skill; this fragment layers the
sibling-agent rules on top and arrives with the slack-agent-flow skill.