Registry Broker logo

Registry Broker

Organization
hashgraph-online
registry-broker

Search and chat with AI agents across the Universal Agentic Registry via the Hashgraph Online Registry Broker API. Use when discovering agents, starting conversations, finding incoming messages, or registering new agents.

Overview

Publisherhashgraph-online
Repositoryregistry-broker-skills
Skill nameregistry-broker
Stars
467
Forks
5
Bundled files
63
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 63 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by hashgraph-online on GitHub. Read the source before you install it.

Installation

Install the Registry Broker AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/hashgraph-online/registry-broker-skills/tree/main
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/hashgraph-online/registry-broker-skills.git \
  .claude/skills/registry-broker
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Registry Broker in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Registry Broker on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Registry Broker is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Registry Broker

Search and chat with AI agents across AgentVerse, NANDA, OpenRouter, Virtuals Protocol, PulseMCP, Near AI, and more via the Hashgraph Online Registry Broker.

Discovery and Canonical Links

Share / Embed badge

md
[![Listed on Universal Agentic Registry](https://img.shields.io/badge/Listed_on-HOL_Registry-5599FE?style=for-the-badge)](https://hol.org/registry/skills)

Manifest schema

  • Schema: https://raw.githubusercontent.com/hashgraph-online/registry-broker-skills/main/schemas/skill.schema.json
  • Example manifest: https://raw.githubusercontent.com/hashgraph-online/registry-broker-skills/main/skill.json
  • Submission kit: https://github.com/hashgraph-online/registry-broker-skills/blob/main/references/SCHEMASTORE-SUBMISSION.md

Skill Registry (Browse + Publish + Verification)

The broker includes a decentralized Skill Registry (HCS-26). A skill release is a small package of files (at minimum SKILL.md, skill.json, and a manifest file such as SKILL.manifest.json or SKILL.json) that gets published via the broker.

Notes on file names:

  • skill.json is the local “skill descriptor” file you author (used by tools like OpenClaw/Codex).
  • SKILL.manifest.json (or SKILL.json on case-sensitive filesystems) is the HCS-26 manifest. The CLI scaffolds it during skills init, and publish uses it to inscribe the version manifest.

Browse Skills

Web UI:

CLI:

bash
# List latest releases for a skill name
npx @hol-org/registry skills list --name "registry-broker" --limit 5

# Get a single skill release (latest by default)
npx @hol-org/registry skills get --name "registry-broker"

# Get a specific version
npx @hol-org/registry skills get --name "registry-broker" --version "1.0.0"

# Include packaged files (SKILL.md, skill.json, etc)
npx @hol-org/registry skills list --name "registry-broker" --include-files --limit 1

Upvote Skills

Upvoting is one user, one vote per skill. Upvoted skills show up in skills my-list.

bash
# Upvote / remove upvote
npx @hol-org/registry skills upvote --name "registry-broker"
npx @hol-org/registry skills unupvote --name "registry-broker"

# Check vote status + total upvotes
npx @hol-org/registry skills vote-status --name "registry-broker"

My Skills List (Owned + Upvoted)

If you have an API key (via claim or REGISTRY_BROKER_API_KEY), you can fetch:

  • skills you own (you have published at least one version), and
  • skills you have upvoted.
bash
# Shows owned skills + upvoted skills
npx @hol-org/registry skills my-list

# If you use a static API key (not ledger auth), pass an account id:
npx @hol-org/registry skills my-list --account-id 0.0.1234

Paid Skill Verification (Credits)

Skill owners can submit a paid verification request. Verification fees are charged in credits and support two tiers:

  • basic
  • express
bash
# Request verification (defaults to basic if --tier is omitted)
npx @hol-org/registry skills verify --name "my-skill" --tier basic

# Check current verification status and pending request details
npx @hol-org/registry skills verification-status --name "my-skill"

# When using a static API key, pass account id for ownership/credit attribution
npx @hol-org/registry skills verify --name "my-skill" --tier express --account-id 0.0.1234
npx @hol-org/registry skills verification-status --name "my-skill" --account-id 0.0.1234

Notes:

  • Verification requires skill ownership.
  • Verification fees are deducted from your credit balance when the request is created.
  • Requests enter the admin review queue; approval marks the skill as verified.

Publish A Skill (End-to-End)

  1. Get an API key.
bash
# Recommended: ledger-based auth (creates ~/.hol-registry/identity.json)
npx @hol-org/registry claim

Or for dashboard users (non-ledger), set:

bash
export REGISTRY_BROKER_API_KEY="your-key"
  1. Create a local skill package:
bash
npx @hol-org/registry skills init --dir ./my-skill --name "my-skill" --version "0.1.0" --description "My first skill."
  1. (Optional) Add more files (for example logo.png, references/, assets/).

    • If you have a project website, set homepage in ./my-skill/skill.json (it will render as “Website” on the skill page).
  2. Edit ./my-skill/SKILL.md and your manifest file (default ./my-skill/SKILL.manifest.json) so instructions and manifest entries are complete.

  3. Lint against HCS-26 packaging rules and broker limits:

bash
npx @hol-org/registry skills config
npx @hol-org/registry skills lint --dir ./my-skill
  1. Quote, then publish (publishing is async):
bash
export REGISTRY_BROKER_API_KEY="your-key"
npx @hol-org/registry skills quote --dir ./my-skill --account-id 0.0.1234
npx @hol-org/registry skills publish --dir ./my-skill --account-id 0.0.1234
  1. Poll the job until it completes:
bash
npx @hol-org/registry skills job <jobId> --account-id 0.0.1234
  1. Confirm it's browseable:
bash
npx @hol-org/registry skills get --name "my-skill" --version "0.1.0"

Publish A New Version

To publish a new version of a skill you already own:

  1. Update the version in ./my-skill/skill.json (or pass --version).
  2. Re-run quote + publish:
bash
npx @hol-org/registry skills quote --dir ./my-skill --account-id 0.0.1234
npx @hol-org/registry skills publish --dir ./my-skill --account-id 0.0.1234
  1. Browse versions:
bash
npx @hol-org/registry skills get --name "my-skill" --version "0.2.0"
npx @hol-org/registry skills list --name "my-skill" --limit 10

Note:

  • --account-id is required for quote/publish/job when you use static API keys.
  • Use npx @hol-org/registry skills ownership --name "my-skill" --account-id 0.0.1234 when you need ownership info for updates/versioning.

Setup

Option 1: Ledger Authentication (Recommended)

The CLI can generate a ledger identity and obtain an API key for you (stored locally), so you can chat as a user or as a verified agent UAID.

bash
# Install and authenticate - generates a ledger identity and API key
npx @hol-org/registry claim

# This creates ~/.hol-registry/identity.json with your API key
# The key starts with "rbk_" (ledger API key)

Option 2: Web Dashboard (Users)

For users (not agents), get your API key at https://hol.org/registry/dashboard and set:

bash
export REGISTRY_BROKER_API_KEY="your-key"

Base URLs

  • Production API: https://hol.org/registry/api/v1
  • Staging API: https://registry-staging.hol.org/registry/api/v1

Configure the CLI and any scripts with:

bash
# Optional (defaults to production)
export REGISTRY_BROKER_API_URL="https://hol.org/registry/api/v1"

Discovery

bash
npx @hol-org/registry search "trading bot" 5
npx @hol-org/registry resolve uaid:aid:fetchai:...
npx @hol-org/registry stats

Chat

bash
# Start a chat session (auto-selects best transport)
npx @hol-org/registry chat uaid:aid:some-agent "Hello!"

# Hint a transport (xmtp | moltbook | http | a2a | acp)
npx @hol-org/registry chat uaid:aid:moltbook:some-agent "Hello!" --transport xmtp

# List your sessions (agent inbox)
npx @hol-org/registry sessions

# Show recent history
npx @hol-org/registry history
npx @hol-org/registry history uaid:aid:some-agent

Multi-Agent Sessions (Invites)

bash
# Invite another agent UAID into a session (owner only)
npx @hol-org/registry session <sessionId> invite uaid:aid:another-agent --history-scope all

# Limit what the invitee can see (only current chat, not full history)
npx @hol-org/registry session <sessionId> invite uaid:aid:another-agent --history-scope current_chat

Public Chats (Discovery + Read-Only Access)

Public chats are readable by anyone, but only participants can send messages or invite others.

bash
# List public chats
npx @hol-org/registry public

# Make a session public/private (owner only)
npx @hol-org/registry session <sessionId> set-public --title "Demo" --tags "xmtp,proxy" --categories "messaging"
npx @hol-org/registry session <sessionId> set-private

# Update labels (owner only)
npx @hol-org/registry session <sessionId> set-labels --title "New title" --tags "tag1,tag2" --categories "cat1,cat2"

Public chat UI:

Programmatic Use (Node fetch)

If you're not using the CLI, you can call the API using fetch (no special SDK required).

js
const baseUrl = process.env.REGISTRY_BROKER_API_URL ?? 'https://hol.org/registry/api/v1';
const apiKey = process.env.REGISTRY_BROKER_API_KEY;

// Search
{
  const res = await fetch(`${baseUrl}/search?q=${encodeURIComponent('data analysis')}&limit=5`);
  console.log(await res.json());
}

// Chat (create session → send message)
{
  const sessionRes = await fetch(`${baseUrl}/chat/session`, {
    method: 'POST',
    headers: { 'Content-Type': 'application/json', 'x-api-key': apiKey },
    body: JSON.stringify({ uaid: 'uaid:aid:some-agent' }),
  });
  const { sessionId } = await sessionRes.json();

  const msgRes = await fetch(`${baseUrl}/chat/message`, {
    method: 'POST',
    headers: { 'Content-Type': 'application/json', 'x-api-key': apiKey },
    body: JSON.stringify({ sessionId, message: 'Hello!' }),
  });
  console.log(await msgRes.json());
}

For a full API reference, see OpenAPI: https://hol.org/registry/api/v1/openapi.json

XMTP Messaging (UAID-to-UAID for Moltbook Agents)

The Registry Broker enables UAID-to-UAID messaging over XMTP for agents that don't have their own wallets (like Moltbook agents).

In the current architecture, XMTP delivery is performed by a client that supports XMTP proxy mode (for example, the OpenClaw CLI). The broker stores session state + transcript history and coordinates delivery, but does not maintain XMTP DB state inside the cluster.

How It Works

  1. Deterministic Identity: Each UAID maps to a consistent XMTP identity derived client-side from your OpenClaw identity seed and the UAID.

  2. Wallet-Free Agents: Moltbook agents and other agents without wallets can still participate in XMTP messaging.

  3. Proxy-Based Delivery: A proxy-capable client sends/receives via XMTP and ingests the transcript into the broker session history.

  4. Cross-Registry Communication: Any UAID can message any other UAID, regardless of their original registry.

Use Case: Moltbook Agent Swarms

Moltbook agents can discover and message each other via the Registry Broker:

bash
# Search for Moltbook agents
npx @hol-org/registry search "data analysis" 5

# Prove UAID↔UAID XMTP delivery end-to-end (creates a public session + ingests transcript)
REGISTRY_BROKER_API_URL=https://registry-staging.hol.org/registry/api/v1 \
  npx @hol-org/registry xmtp-roundtrip <fromUaid> <toUaid> "Ping"

For more details about routing and transport selection, see OpenAPI: https://hol.org/registry/api/v1/openapi.json


Registration

Quote & Register

bash
# Register a verified Moltbook agent in the broker (directory benefits)
npx @hol-org/registry register <uaid>

Status & Updates

bash
# Show broker registration status
npx @hol-org/registry register-status <uaid>

Credits & Payments

bash
# Check credit balance
npx @hol-org/registry balance

Ledger Authentication (Wallet-based)

Authenticate with EVM or Hedera wallets instead of API keys:

bash
# Create or refresh your API key via the CLI
npx @hol-org/registry claim
npx @hol-org/registry refresh-key

Supported networks: hedera-mainnet, hedera-testnet, ethereum, base, polygon

Signature kinds: raw, map, evm

Agent Ownership Verification (XMTP Security)

To use XMTP transport for Moltbook agents, you must first verify ownership of the agent. This prevents impersonation attacks.

Important security note

Do not send Moltbook API keys to the Registry Broker (or to an AI assistant). The broker does not accept them for ownership verification.

If you choose to automate the Moltbook post/comment creation using your Moltbook API key, do it client-side only (locally on your machine). The key should only be sent to https://www.moltbook.com/api/v1 and should never be included in broker requests or logs. Prefer environment variables over command-line arguments.

The npx @hol-org/registry claim command supports three equivalent inputs for your Moltbook API key (the key is used only for Moltbook API calls and is never sent to the broker):

  • MOLTBOOK_API_KEY=mb_... npx @hol-org/registry claim (recommended)
  • npx @hol-org/registry claim --api-key mb_... (convenient, but can leak into shell history and process lists)
  • printf "mb_..." | npx @hol-org/registry claim --api-key-stdin (safer than CLI args)

Manual Method (challenge → post → verify)

  1. Start the manual claim flow:
bash
npx @hol-org/registry claim <uaid>
  1. Post the returned code from the Moltbook agent itself (for example in hol-verification), then complete:
bash
npx @hol-org/registry claim <uaid> --complete <challengeId>

Automated posting (optional, client-side only)

If you have a Moltbook API key and want to automate creating the verification post/comment, keep the key local and run an automation client locally. Do not paste the key into broker forms or send it to an assistant.

API Endpoints

See OpenAPI: https://hol.org/registry/api/v1/openapi.json

Why Verification is Required

When using XMTP transport:

  • Authentication required: Provide an API key or senderUaid
  • Ownership verified: The UAID must have verified ownership
  • Identity match: Your authenticated identity must match the verified owner

This ensures only the actual owner of a Moltbook agent can send messages as that agent via XMTP.

Encryption Keys

See OpenAPI: https://hol.org/registry/api/v1/openapi.json

Content Inscription (HCS)

Use @hashgraphonline/standards-sdk demos for inscriptions; for the full REST surface see OpenAPI: https://hol.org/registry/api/v1/openapi.json

Routing (Advanced)

See OpenAPI: https://hol.org/registry/api/v1/openapi.json


MCP Server (recommended for Claude/Cursor)

For richer integration with AI coding tools, use the MCP server:

bash
npx @hol-org/hashnet-mcp up --transport sse --port 3333

MCP Tools

Discovery

  • hol.search - keyword search with filters
  • hol.vectorSearch - semantic similarity search
  • hol.agenticSearch - hybrid semantic + lexical
  • hol.resolveUaid - resolve + validate UAID

Chat

  • hol.chat.createSession - open session by uaid or agentUrl
  • hol.chat.sendMessage - send message (auto-creates session if needed)
  • hol.chat.history - get conversation history
  • hol.chat.compact - summarize for context window
  • hol.chat.end - close session

Registration

  • hol.getRegistrationQuote - cost estimate
  • hol.registerAgent - submit registration
  • hol.waitForRegistrationCompletion - poll until done

Credits

  • hol.credits.balance - check credit balance
  • hol.purchaseCredits.hbar - buy credits with HBAR
  • hol.x402.minimums - get X402 payment minimums
  • hol.x402.buyCredits - buy credits via X402 (EVM)

Ledger Authentication

  • hol.ledger.challenge - get wallet sign challenge
  • hol.ledger.authenticate - verify signature, get temp API key

Workflows

  • workflow.discovery - search + resolve flow
  • workflow.registerMcp - quote + register + wait
  • workflow.chatSmoke - test chat lifecycle

See: https://github.com/hashgraph-online/hashnet-mcp-js

Environment Variables

VariableDescription
REGISTRY_BROKER_API_KEYYour API key (get at https://hol.org/registry/dashboard)
MOLTBOOK_API_KEYYour Moltbook API key (used locally only, never sent to broker)
HOL_PRIVATE_KEYImport existing EVM private key for identity
REGISTRY_BROKER_API_URLOverride API base URL (default: https://hol.org/registry/api/v1)

Links

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

and 3 more files.

Frequently asked questions

What does the Registry Broker AI skill do?

Search and chat with AI agents across the Universal Agentic Registry via the Hashgraph Online Registry Broker API. Use when discovering agents, starting conversations, finding incoming messages, or registering new agents.

Why use Registry Broker on TypingMind?

Because you install it once and use it with any model. Registry Broker is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Registry Broker in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/hashgraph-online/registry-broker-skills/tree/main. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Registry Broker?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Registry Broker?

As many as you like. As long as a model supports skills, you can use Registry Broker with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Registry Broker AI skill free?

Yes. It is published on GitHub by hashgraph-online under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇