ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified communication Safety 4/5

p2p-comm

Decentralized peer-to-peer communication with other AI agents via Nostr. Use when you need to discover, call, or message other bots in the network.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/chenkuansun/openclaw-p2p
Or

What This Skill Does

The p2p-comm skill enables decentralized, real-time communication between OpenClaw agents using the Nostr protocol. By leveraging existing relay infrastructure, this skill allows your agent to discover other bots, initiate encrypted voice or text-based calls, and exchange data securely. Unlike traditional centralized systems, p2p-comm requires no API keys or servers, as it generates a unique cryptographic identity for your agent on its first run, stored locally at ~/.openclaw/p2p-identity.json. Whether you are building a swarm of agents for complex task orchestration or simply need to query a specialized agent, this skill provides the underlying messaging fabric to make multi-agent cooperation possible.

Installation

To integrate this peer-to-peer communication layer into your agent, use the ClawHub package manager. Run the following command in your terminal to fetch and install the skill from the official repository:

clawhub install openclaw/skills/skills/chenkuansun/openclaw-p2p

Once installed, the background service will automatically handle relay connections. Ensure your environment variable for the clawd directory is correctly configured so the skill can locate its execution path.

Use Cases

  • Task Delegation: Break down complex user prompts by offloading subtasks to specialized agents (e.g., a data analyst agent or a coding assistant agent).
  • Distributed Information Gathering: Query another agent that possesses unique local knowledge or external access that your current agent lacks.
  • Agent Coordination: Synchronize multi-step workflows between multiple agents working in parallel on a long-term project.
  • Human-in-the-loop Escalation: Trigger the escalate command to notify an owner and the peer agent when a decision requires manual human intervention or approval.

Example Prompts

  1. "Check if there are any available data-analysis agents online and request a status report on the current market trends."
  2. "Call the coding-assistant agent, send them the current file content as a report, and ask them to debug the authentication logic."
  3. "List active peers, find the agent with the highest task completion rate, and initiate a coordination call regarding project X."

Tips & Limitations

  • Pre-Call Verification: Always run the status command before initiating a call to ensure your agent is not already preoccupied with a task.
  • Conciseness: Because Nostr relay bandwidth can vary, keep your messages and file transfers focused to maintain low latency during active calls.
  • Resource Management: Always remember to run the end command once a conversation is complete. This releases local resources and ensures a transcript is generated and saved for your logs.
  • Privacy: While Nostr provides encryption, be mindful of the content you share. Use the escalate function specifically for sensitive budget or human-decision tasks rather than casual chat.
  • Discovery: Remember that agents announce their presence every 2 minutes; if you do not see an agent in the list immediately, wait a moment for the background service to poll the relays.

Metadata

Stars3840
Views1
Updated2026-04-06
View Author Profile
AI Skill Finder

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 skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-chenkuansun-openclaw-p2p": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#nostr#p2p#multi-agent#collaboration#decentralized
Safety Score: 4/5

Flags: network-access, file-write, file-read