registry-broker

What This Skill Does

The Registry Broker skill serves as a universal gateway for discovering, interacting with, and registering AI agents across a vast, fragmented ecosystem. By integrating the Hashgraph Online Registry Broker, this skill acts as an abstraction layer for over 72,000 agents spanning 14 distinct registries, including AgentVerse, NANDA, OpenRouter, and more. It utilizes the standardized @hashgraphonline/standards-sdk to normalize communication, allowing users to perform semantic searches, inspect agent capabilities, and maintain persistent chat sessions with remote AI entities through a single unified interface.

Installation

To integrate the Registry Broker into your OpenClaw environment, ensure you are in the project base directory and execute the following command: clawhub install openclaw/skills/skills/kantorcodes/registry-broker-hashnet-openclaw. Once installed, navigate to the local skill directory, run npm install to resolve dependencies, and ensure you have an API key from https://hol.org/registry. Providing this key is highly recommended for authenticated operations, which include registering your own agents and accessing higher tier API rate limits.

Use Cases

This skill is designed for power users and developers who need to bridge the gap between isolated AI agent platforms. You can use it to build sophisticated agent-to-agent workflows, perform market research on available autonomous tools, or facilitate cross-platform interoperability. It is particularly valuable for developers building decentralized applications on Hedera or those looking to leverage specialized agents from diverse providers like Virtuals Protocol or Coinbase x402 without having to manually manage individual API credentials for every source.

Example Prompts

1. "Find me an AI agent that specializes in cryptocurrency market analysis and can provide real-time price alerts."
2. "Search for agents available on the NANDA registry that are capable of generating creative images."
3. "Start a chat session with the agent having UAID uaid:aid:7890-xyz-123 and ask it to explain its core function."

Tips & Limitations

To get the best results, use descriptive queries when performing semantic searches, as the system relies on vector-based lookups to match user intent with agent metadata. Note that this skill requires active network access to query the remote Hashgraph registries; ensure your environment configuration allows for external HTTPS calls. For production workflows, always resolve agent UAIDs before initiating long-running chat sessions to ensure the agent's connection state is valid. Be aware that response latency may vary depending on the target agent's hosting platform and current load.