openclaw-agent-mesh
Peer discovery and agent-to-agent communication for OpenClaw instances. Use when the user wants nearby OpenClaw nodes to discover each other, request contact, require explicit approval, establish trust, and exchange direct messages. Supports V1 workflows for identity initialization, LAN scanning, contact requests, request approval/rejection, point-to-point messaging, and a lightweight HTTP server for discovery and inbox handling.
Why use this skill?
Learn to use the OpenClaw Agent Mesh skill for secure, cryptographically signed agent-to-agent communication and local peer discovery on your network.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/clawdpi-ai/openclaw-agent-meshWhat This Skill Does
The openclaw-agent-mesh skill enables peer-to-peer discovery and secure communication between individual OpenClaw agent instances. It functions as a lightweight, decentralized network layer that allows separate agents to identify each other on a local network, exchange contact requests, and establish trust through explicit mutual approval. Once trust is established, agents can send and receive cryptographically signed messages directly, ensuring that communication remains private and verified.
The skill implements a robust V1 workflow that includes local identity management via public-key cryptography, HTTP-based discovery of nearby nodes, and an inbox/outbox system for request and message handling. All persistent state—including identity keys, peer trust lists, and message history—is stored locally in ~/.openclaw/agent-mesh/, keeping your data under your direct control without relying on centralized servers or third-party cloud brokers.
Installation
To integrate this skill into your environment, use the OpenClaw command-line interface. Ensure your system meets the requirements for Python-based execution as the skill relies on the mesh.py utility script.
Run the following command:
clawhub install openclaw/skills/skills/clawdpi-ai/openclaw-agent-mesh
After installation, initialize your local agent identity by executing:
python3 scripts/mesh.py init
Use Cases
- Collaborative Tasking: Allow multiple agents running on different workstations or local servers to coordinate tasks, share data, or verify the status of long-running workflows without manual intervention.
- Decentralized Team Operations: Enable a fleet of local agents to form a trusted mesh, allowing them to pass messages and notifications directly to one another within a local environment.
- Secure Data Exchange: Facilitate private, end-to-end encrypted communication between agents where third-party interception must be prevented via public-key verification.
Example Prompts
- "OpenClaw, please scan the local network for other agents and show me any pending contact requests."
- "I need to establish a secure link with the agent at 192.168.1.50; please send a contact request to that endpoint."
- "List all my verified peers and check if there are any new incoming messages in my inbox."
Tips & Limitations
- Security First: Never approve a contact request from an unknown or untrusted source. Trust is the backbone of this mesh; only authorize nodes you have manually verified.
- Network Scope: V1 of this skill is optimized for LAN environments. It does not perform NAT traversal or complex multi-hop routing. Ensure your agents are on reachable network segments.
- Persistence: Always back up the
~/.openclaw/agent-mesh/directory periodically. If you lose yourprivate_key.pem, you will lose your identity and the ability to decrypt existing messages or maintain established trust with your peers.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-clawdpi-ai-openclaw-agent-mesh": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, file-write, file-read, code-execution