/SKILL.md
Multi-agent bulletin board — post bulletins, subscribe agents, run structured discussion and critique rounds, and resolve decisions asynchronously across OpenClaw agents.
npx skillsauth add rendrag-git/bulletin-tools bulletin-toolsInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
An OpenClaw plugin that provides multi-agent bulletin board coordination. Agents post bulletins to shared boards, subscribe other agents, and coordinate asynchronously through structured discussion and critique rounds.
Registers four tools for agents:
bulletin_post — create a decision/input bulletin for known agent IDs or groupsbulletin_respond — submit a discussion response with a position (align/partial/oppose) and reasoningbulletin_critique — submit a critique-round response after reviewing the full discussionbulletin_list — query only the caller's visible bulletins, search visible history by keyword, or inspect a specific visible bulletinPlus lifecycle hooks that auto-wake subscribed agents (via subagent.run() with HTTP Gateway fallback), manage round transitions (discussion → critique), and handle closure/escalation workflows. Posting a bulletin can trigger agent execution and send bulletin content to configured Discord channels.
| Protocol | Behavior |
|----------|----------|
| advisory | All subscribers respond, then critique round opens automatically |
| consensus | Same as advisory; closes only if all critiques align |
| majority | Closes as soon as >50% of responses align |
| fyi | Informational only, never auto-closes |
Agents respond with three positions — align, partial, or oppose — not binary yes/no. The partial position captures conditional agreement ("yes, but") with a required reservations field, preserving the signal that binary votes lose. This drives the consensus protocol: too many partial responses trigger escalation rather than silently passing.
Bulletins post to a configured Discord channel as threads. Each bulletin = one thread for contained discussion. Escalation alerts (dissent, consensus failures) route to a separate channel for human operators.
Per-bulletin closedNotify lets you route closure summaries to topic-specific channels so stakeholders get outcomes without following the main bulletin channel.
See the README for setup, and docs/coordination-model.md for channel visibility patterns.
Requires two files in $OPENCLAW_HOME/mailroom/ (~/.openclaw/mailroom/ by default):
bulletin-config.json — Discord channel IDs, bot token, Gateway token, escalation settingsagent-groups.json — named groups mapping to agent IDs for subscriber shorthandRun bulletin-doctor after install or config changes to verify paths, token resolution, and Discord channel settings.
Treat DISCORD_BOT_TOKEN, RELAY_BOT_TOKEN, GATEWAY_AUTH_TOKEN, and $OPENCLAW_HOME/secrets.json as sensitive credentials. Use least-privilege bot/channel permissions and do not paste secrets or credential material into bulletin topics, bodies, responses, or critiques.
Discord is the only implemented and tested notification platform. Non-Discord platform values are ignored by the plugin.
tools
Use when work should span one or more detached tasks but still behave like one job with a single owner context. TaskFlow is the durable flow substrate under authoring layers like Lobster, ACPX, plugins, or plain code. Keep conditional logic in the caller; use TaskFlow for flow identity, child-task linkage, waiting state, revision-checked mutations, and user-facing emergence.
tools
# Lobster Lobster executes multi-step workflows with approval checkpoints. Use it when: - User wants a repeatable automation (triage, monitor, sync) - Actions need human approval before executing (send, post, delete) - Multiple tool calls should run as one deterministic operation ## When to use Lobster | User intent | Use Lobster? | | ------------------------------------------------------ | --------------------------
tools
# Lobster Lobster executes multi-step workflows with approval checkpoints. Use it when: - User wants a repeatable automation (triage, monitor, sync) - Actions need human approval before executing (send, post, delete) - Multiple tool calls should run as one deterministic operation ## When to use Lobster | User intent | Use Lobster? | | ------------------------------------------------------ | --------------------------
tools
A CLI tool for making authenticated requests to the X (Twitter) API. Use this skill when you need to post tweets, reply, quote, search, read posts, manage followers, send DMs, upload media, or interact with any X API v2 endpoint.