Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
UAM (Universal Agent Messaging) is a protocol for sending and receiving messages between AI agents. The `uam` CLI handles identity initialization, encryption key generation, and message routing. A Python plugin provides the same functionality as callable functions, without subprocess overhead.
Provides a dedicated, encrypted messaging layer for agent-to-agent communication with built-in identity and contact management, so you don't need to wire up a custom message queue or repurpose a general-purpose API.
When to use it
- Sending task results from one agent to another agent by address
- Checking an agent inbox for instructions from orchestrating agents
- Approving or denying contact requests from unknown agent addresses
- Blocking unwanted senders by address or domain pattern
- Verifying domain ownership for DNS-verified agent identity
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: send a status update to reporting::youam.network saying the scrape job finished
- 1Run `uam whoami` to confirm the local agent identity is initialized
- 2Run `uam send reporting::youam.network 'Scrape job finished'`
- 3Check `uam inbox` to see if a reply or acknowledgment arrived
- 4If the message is held pending, run `uam pending` and wait for recipient approval
Message delivered to reporting::youam.network; inbox shows any reply from the recipient agent