agentleFS
0xranx0xranx/golembot/skills/multi-bot/SKILL.md

multi-bot

Skill326 starsChanged 9 months agoMITView on GitHub

Coordinates responses between multiple GolemBot instances in a shared fleet. Use when the bot operates in a group chat with other bots, needs to decide whether to respond or pass, or must call a peer bot's API to fetch cross-domain data.

  • Sends data out

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 8 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    When you see [Peers: 小忆 (user research), 小舟 (content creation)] in the prompt:

    line 12
  2. Distinctive:

    Respond: if the message falls within your domain/role

    line 22
  3. Distinctive:

    Respond: if you are directly @mentioned

    line 23
  4. Distinctive:

    [PASS]: if the message clearly belongs to another peer's domain

    line 24
  5. Distinctive:

    Respond: if the topic spans multiple domains — focus on YOUR area of expertise only, don't duplicate what peers would cover

    line 25
  6. Distinctive:

    Respond: if no peer is better suited (don't let messages go unanswered)

    line 26
  7. Very specific:

    Check the conversation history for [bot:PeerName] entries

    line 30
  8. Distinctive:

    If a peer already covered a topic, don't repeat it — add new information or skip

    line 31
  9. Distinctive:

    When a topic spans multiple domains, scope your response to your own expertise

    line 32
  10. Distinctive:

    Use these labels to understand the conversation context and avoid repeating what peers have already said.

    line 92

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Multi-Bot Collaboration

You may be one of several GolemBot instances running in the same fleet. The gateway injects [Peers: ...] into your group chat context so you know who else is present.

Peer Awareness

When you see [Peers: 小忆 (user research), 小舟 (content creation)] in the prompt:

  • These are other GolemBot instances in the fleet
  • Each has its own specialization (shown in parentheses)
  • They may or may not be active in this specific group chat

When to Respond vs [PASS]

In group chats with peers, follow these rules:

  1. Respond if the message falls within your domain/role
  2. Respond if you are directly @mentioned
  3. [PASS] if the message clearly belongs to another peer's domain
  4. Respond if the topic spans multiple domains — focus on YOUR area of expertise only, don't duplicate what peers would cover
  5. Respond if no peer is better suited (don't let messages go unanswered)

Avoiding Redundancy

  • Check the conversation history for [bot:PeerName] entries
  • If a peer already covered a topic, don't repeat it — add new information or skip
  • When a topic spans multiple domains, scope your response to your own expertise

Calling Peer Bots

You can call other GolemBot instances directly via their HTTP API when you need their capabilities:

# Ask a peer bot a question
curl -s -X POST http://<peer-url>/chat \
  -H 'Content-Type: application/json' \
  -d '{"message": "your question here", "sessionKey": "cross-bot-<context>"}'

# Check a peer's status
curl -s http://<peer-url>/health

Error Handling for Peer Calls

Peer bots may be temporarily unavailable. Always check the HTTP status code before using the response:

response=$(curl -s -w "\n%{http_code}" -X POST http://<peer-url>/chat \
  -H 'Content-Type: application/json' \
  -d '{"message": "your question", "sessionKey": "cross-bot-ctx"}')
status=$(echo "$response" | tail -1)
body=$(echo "$response" | sed '$d')

if [ "$status" -ne 200 ]; then
  # Peer unavailable — answer with your own knowledge or inform the user
  echo "Peer bot returned status $status; falling back to local knowledge."
fi

When to Call a Peer

  • You need data or analysis from another domain (e.g., you're the product bot and need user feedback data from the user-research bot)
  • A user asks you something outside your expertise — call the appropriate peer, then synthesize the response
  • You want to verify or cross-reference information

When NOT to Call a Peer

  • The question is within your own domain
  • The peer is likely to respond on their own in the group chat (redundant)
  • Simple factual questions you can answer yourself

Discovering Peers

Peer information is automatically injected by the gateway from fleet discovery. You can also check:

# List all running GolemBot instances
curl -s http://localhost:4000/api/fleet

Conversation History Labels

In group chat history, messages are labeled:

  • [username] — human message
  • [bot:BotName] — message from a peer bot

Use these labels to understand the conversation context and avoid repeating what peers have already said.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.