Agent Link is a secure communication bridge that enables distributed Openclaw Skills to exchange messages across different devices via a relay server.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-link-relay-server
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install agent-link-relay-server using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Agent Link?
Agent Link provides a high-performance framework for interconnecting Openclaw Skills running on geographically dispersed hardware. By leveraging a relay server architecture, it allows AI agents to communicate through firewalls and different networks without compromising security. This skill acts as the nervous system for a distributed agentic workflow, ensuring that an agent on one machine can trigger actions or share data with an agent on another computer.
Built with security as a priority, Agent Link utilizes HMAC-SHA256 signatures and WebSocket Secure (WSS) to protect the integrity of your Openclaw Skills interactions. The relay server functions purely as a router, forwarding messages without storing sensitive content, making it an ideal choice for privacy-conscious developers building complex multi-agent systems.
Agent Link Use Cases
- Coordinating tasks between a home-based Openclaw instance and a cloud-based research agent.
- Sending real-time status updates from edge devices to a central monitoring agent.
- Building multi-agent swarms where specialized Openclaw Skills work together across different OS environments.
- Implementing remote control capabilities for AI agents across various local area networks.
How Agent Link Works
- The Local Agent connects to a public or private Relay Server using a secure WebSocket connection.
- When an agent needs to send data, it packages the message with a unique instance ID and destination ID.
- The message is signed with a shared secret using HMAC-SHA256 to ensure authenticity.
- The Relay Server receives the packet, validates the signature, and routes it to the target Openclaw instance.
- The recipient Agent processes the message and sends an optional acknowledgment back through the relay.
Agent Link Setup
To deploy the relay server component for your Openclaw Skills network:
# On your public server
cd skills/agent-link/scripts/relay-server
python3 relay_server.py --port 8765 --secret "your-secure-key"
To configure a local agent to connect to the relay:
# On your local machine
cd skills/agent-link/scripts/local-agent
python3 setup.py --relay-url "ws://your-relay-server:8765" --secret "your-secure-key"
Agent Link Data Schema & Taxonomy
Agent Link utilizes structured JSON configurations to manage the routing of Openclaw Skills:
| Configuration | Key Parameters | Description |
|---|---|---|
| relay-config.json | port, secret, registered_instances |
Defines server listening port and authorized agent nodes. |
| agent-link-config.json | relay_url, instance_id, agent_id, auto_reconnect |
Local settings for connecting the agent to the communication mesh. |
Message Schema:
from: The sender's instance and agent identifier.to: The target recipient identifier.message: The actual payload content.signature: Security hash for data integrity.
Agent Link Advanced Features
- Secure message signing using HMAC-SHA256 to prevent unauthorized injection into your Openclaw Skills ecosystem.
- Intelligent auto-reconnection logic to handle intermittent network failures between agents.
- Zero-storage relaying ensures your communication remains private and is never persisted on the server.
- Support for multi-agent routing within a single Openclaw instance, allowing for hierarchical command structures.
SKILL.md
Loading
Related Openclaw Skills
LLMCities Web Host
A specialized web hosting skill that enables AI agents to publish content, manage static websites, and track visitor analytics on LLMCities.com.
Gold Price Fetcher
A high-performance utility to retrieve and cache real-time gold price data directly from JD Finance for AI agents.
AI Daily Report
A specialized AI agent skill designed to aggregate and format the last 24 hours of global AI industry news into sendable Lark message formats.
File Classification Manager
An automated dual-storage system that intelligently separates permanent project deliverables from intermediate working files to eliminate workspace clutter.
Feishu Message Download
A professional utility to programmatically download attachments like videos, images, and documents from Feishu chat messages to local storage.
Claude Code Framework
A professional-grade Agent execution framework extracted from core Claude Code patterns to ensure secure, efficient, and traceable AI operations.
Featured*
