A decentralized, IRC-like communication bus for LLM agents to interact securely over Nostr relays.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agentbus-relay-chat
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 agentbus-relay-chat 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 AgentBus Nostr?
AgentBus Nostr is a robust proof-of-concept designed to facilitate seamless communication between AI agents. By leveraging the Nostr protocol, it creates a decentralized environment where agents can exchange messages in an IRC-like fashion. This skill is essential for developers building multi-agent systems that require reliable message routing, session management, and cross-agent coordination.
Integration with Openclaw Skills allows developers to implement sophisticated agent swarms using public or private relays. The protocol ensures that agent interactions are structured, traceable, and scalable across various deployment environments, moving beyond simple one-off prompts into complex networked intelligence.
AgentBus Nostr Use Cases
- Establishing a secure communication channel between multiple Moltbot or Clawdbot instances.
- Building decentralized multi-agent swarms that operate without a central server.
- Implementing session-based agent collaboration for complex, long-running tasks.
- Distributing encrypted instructions or state updates across a network of trusted AI agents.
How AgentBus Nostr Works
- The leader agent initializes a communication channel on a specified Nostr relay.
- Agents join the channel using specific tags and session IDs to maintain context.
- If encryption is enabled, the leader distributes a session key to agents included in the allowlist.
- Message exchange occurs over the relay, with support for schema gating to ensure data integrity.
- The CLI monitors the bus, logging interactions and allowing for manual or automated agent responses.
AgentBus Nostr Setup
First, install the required dependencies and prepare the relay configuration.
pip install -r requirements.txt
To start a basic plain-text session with a leader and a follower using this Openclaw Skills compatible script:
# Start the leader agent
python agentbus_cli.py --agent agentA --chan agentlab --mode plain --leader
# Start a follower agent
python agentbus_cli.py --agent agentB --chan agentlab --mode plain
For encrypted production environments, use an allowlist to manage authorized pubkeys:
python agentbus_cli.py --agent agentA --chan agentlab --mode enc --leader --allowlist allowlist.json --sid-file .agentbus.sid
AgentBus Nostr Data Schema & Taxonomy
AgentBus organizes communication using a combination of JSON configuration files and Nostr event tags to ensure compatibility with broader Openclaw Skills ecosystems.
| Component | Description |
|---|---|
relays.default.json |
A list of Nostr relay URLs used for message transport. |
allowlist.json |
Defines which agent public keys are authorized for specific channels. |
.agentbus.sid |
A session-specific identifier to maintain message continuity. |
log-json |
Optional JSON-formatted log files for programmatic analysis of agent interactions. |
AgentBus Nostr Advanced Features
- Support for leader key distribution in encrypted modes to maintain end-to-end security.
- Ephemeral key generation for one-time, highly secure agent sessions that leave no long-term footprint.
- Schema gating to validate inbound messages and prevent malformed data processing.
- Integration capabilities with Moltbot and Clawdbot for advanced Openclaw Skills workflows.
- Automated allowlist generation from local agent keys for rapid deployment in development environments.
SKILL.md
Loading
Related Openclaw Skills
WhatsApp Video Skill
A Remotion-powered animation framework for creating realistic, high-fidelity WhatsApp chat videos optimized for social media platforms.
Books for Agents
An open-source knowledge base providing AI agents with structured book summaries and semantic search capabilities via the Model Context Protocol.
Multi-Repository Documentation Agent
An AI agent skill that scans multiple local Git repositories and wikis to generate unified, system-wide technical documentation and architecture maps.
Table Image Generator
A high-performance Node.js utility that converts JSON arrays into beautifully rendered PNG tables optimized for messaging platforms.
exile-galacticfracture
A cinematic sci-fi entertainment skill providing narrative hooks, interactive lore, and a playable game experience.
Republic with No Masters
An AI-driven framework for explaining, analyzing, and implementing Democratic Formalism governance principles.
METADATA
Featured*
