Stats
Actions
Tags
How this skill is triggered — by the user, by Claude, or both
Slash command
/hivemind:hiveThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
All Hivemind commands are available as MCP tools. When the user runs `/hive <command>`, call the corresponding MCP tool.
All Hivemind commands are available as MCP tools. When the user runs /hive <command>, call the corresponding MCP tool.
Run /hive setup to configure your environment:
| User Command | MCP Tool to Call |
|---|---|
/hive or /hive help | hive_help |
/hive setup | hive_setup |
/hive whoami | hive_whoami |
/hive agents | hive_agents |
/hive status | hive_status |
/hive message <target> <text> | hive_message with target and body |
/hive changes [n] | hive_changes with optional count |
/hive task [description] | hive_task with optional description |
/hive inbox [n] | hive_inbox with optional limit |
/hive read_message <id> | hive_read_message with id |
/hive clean_inbox | hive_clean_inbox |
Parse $ARGUMENTS and call the corresponding MCP tool:
help: Call hive_helpsetup: Call hive_setupwhoami: Call hive_whoamiagents: Call hive_agentsstatus: Call hive_statusmessage <target> <text>: Call hive_message with {"target": "<target>", "body": "<text>"}changes [n]: Call hive_changes with {"count": n} (default 20)task [description]: Call hive_task with {"description": "<description>"} (empty to clear)inbox [n]: Call hive_inbox with {"limit": n} (default 10)read_message <id>: Call hive_read_message with {"id": "<id>"}clean_inbox: Call hive_clean_inboxThe MCP tools handle all coordination logic.
/hive setup
-> Call hive_setup (configures status line)
/hive
-> Call hive_help
/hive whoami
-> Call hive_whoami
/hive message bravo Please hold off on auth.ts
-> Call hive_message with {"target": "bravo", "body": "Please hold off on auth.ts"}
/hive task Implementing user authentication
-> Call hive_task with {"description": "Implementing user authentication"}
/hive task
-> Call hive_task with {"description": ""} to clear
/hive changes 10
-> Call hive_changes with {"count": 10}
/hive inbox
-> Call hive_inbox
/hive inbox 5
-> Call hive_inbox with {"limit": 5}
/hive read_message msg-1708963200-12345-6789
-> Call hive_read_message with {"id": "msg-1708963200-12345-6789"}
/hive clean_inbox
-> Call hive_clean_inbox
npx claudepluginhub ai-origo/claude-hivemindGuides multi-agent coordination using agent-swarm MCP tools: join swarms as leader/worker, delegate/claim tasks, check status, send/read messages, manage workflows.
Coordinates multi-agent swarms with queen-led architecture, Byzantine consensus, and shared persistent memory. Use for orchestrating agent teams under a lead coordinator or building resilient cross-agent collaboration.
Publishes messages to topics for inter-agent handoffs, notifications, broadcasts across hive sessions using `hive msg pub` CLI with flags for message, file, stdin, wildcards.