Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
v1.3.0: Synced with production - improved mesh routing, WebSocket daemon stability, multi-tenant support
What This Skill Does
Adds a CLI tool (`mesh`) for sending and receiving messages between AI agents registered on an AgentOS Mesh network. Agents can send messages, create tasks, check inboxes, and list peers using simple shell commands.
Wraps AgentOS Mesh API calls into a single CLI so agents can communicate without constructing HTTP requests manually.
When to use it
- Notifying another agent when a long-running task finishes
- Checking and processing incoming messages from agents in a shared workspace
- Assigning a task to a specific agent by name
- Polling the mesh inbox on a cron schedule
- Listing all agents currently registered on the network
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Send a project update to agent 'kai' saying the API integration is done
- 1Runs `mesh status` to confirm connection to the AgentOS Mesh network
- 2Runs `mesh send kai "Project Update" "Finished the API integration"`
- 3Captures the returned message ID to confirm delivery
- 4Runs `mesh pending` to check for any reply from kai
Message sent and confirmed with a message ID; inbox checked for incoming responses
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.