nanocoai/nanoclaw / .claude/skills/slack-agent-flow/container/skills/slack-construct-agents
slack-construct-agents
VerifiedStanding context for Slack sibling agents — one room per team, creator-posted introductions, bot-to-bot hop discipline. Ships as instructions.md composed into the group's CLAUDE.md at spawn; there is no workflow to invoke.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 1.2 KB |
| instructions.md | markdown | 1.8 KB |
instructions.md
View on GitHub## Your Slack sibling agents
You may be part of a construct of agents on Slack: sibling agents (each with its own bot,
DM, and workspace) plus shared rooms where humans, you, and siblings talk. Standing rules
for the sibling half:
- **Teams get ONE room.** When the user asks for several agents for one project ("build me
a team"), create each agent with `create_agent({ ..., room: 'none' })` — they still get
their operator DM — then open a single shared room with
`create_room({ name, purpose, agents: [all of them] })`. Never let each create open its
own room: that yields N separate three-way rooms nobody wants. `add_to_room` works for
later growth, but Slack group DMs never grow in place — the room MOVES to a new
conversation (everyone is re-wired automatically; the old one keeps working), so prefer
creating rooms complete.
- **You introduce agents you create.** When a shared room comes with an agent you created,
YOU post the introduction in the room — the host posts nothing there. You'll get a
system nudge telling you which destination to use. Keep it to 1-2 lines in your own
voice: say what the new agent is for and tag them with their `<@bot-user-id>` mention
(send it literally; it renders as a mention). No mechanics, no member lists — the room's
canvas tab already holds that.
- **Bot-to-bot hop budget.** The platform may cap consecutive bot-to-bot messages (~6)
until a human speaks again, but do not rely on it — self-limit. Don't ping-pong with
siblings: do the work, converge, hand back to the human.
- **Persist durable facts.** Conversations are per-session; rooms and DMs don't share
history. Anything worth keeping (decisions, preferences, ongoing state) goes in your
memory directory, not just the chat transcript.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |