A decentralized reputation and trust registry for verifying the security and reliability of AI agent skills.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install clawmon
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 clawmon 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 Trusted ClawMon?
Trusted ClawMon is a robust, read-only trust lookup service designed to enhance the security of the agentic ecosystem. By leveraging ERC-8004 and the Monad network, it provides a transparent way for agents and developers to query the reputation of Openclaw Skills before execution. This ensures that only verified, high-quality tools are integrated into workflows, mitigating risks associated with untrusted code.
The service operates entirely via public HTTPS GET requests, requiring no wallet connections or complex authentication. It provides critical security signals such as Sybil resistance checks, TEE (Trusted Execution Environment) attestation, and staking status, making it an essential layer for any secure AI agent implementation.
Trusted ClawMon Use Cases
- Verifying the safety of a new MCP skill before first-time invocation.
- Checking if a skill publisher has staked collateral to ensure accountability.
- Auditing the TEE status of a skill to confirm secure code execution.
- Identifying whether a specific entry in the list of Openclaw Skills has been flagged for Sybil activity.
How Trusted ClawMon Works
- The AI agent receives a request to use an external skill or is asked to verify a skill integrity.
- The agent performs a read-only HTTPS GET request to the Trusted ClawMon API using the skill identifier.
- The registry returns a comprehensive JSON payload containing trust scores, tiers (AAA through C), and security metadata.
- The agent analyzes the hardenedScore and teeStatus to determine if the skill meets the required security threshold.
- The agent provides the user with a trust report or proceeds with execution based on the retrieved reputation data for Openclaw Skills.
Trusted ClawMon Setup
Since Trusted ClawMon is a public API, no API keys or environment variables are required for standard lookups. For developers who wish to self-host the registry, follow these steps:
git clone https://github.com/drewmailen/ClawMon
cd ClawMon
npm install
npm run build
npm start
Ensure your agent is configured to point to your hosted instance or the public endpoint at https://trusted-clawmon-api.up.railway.app.
Trusted ClawMon Data Schema & Taxonomy
The skill interacts with structured reputation data organized as follows:
| Field | Description | Type |
|---|---|---|
| hardenedScore | Sybil-resistant trust score (0-100) | Number |
| hardenedTier | Categorical trust rating (AAA, AA, etc.) | String |
| isSybil | Flag indicating if the skill is part of a sybil cluster | Boolean |
| isStaked | Indicates if the publisher has staked MON | Boolean |
| teeStatus | TEE attestation (verified, unregistered, expired) | String |
| teeCodeHashMatch | Verifies if the running code matches the pinned version | Boolean |
All responses are delivered as standard JSON objects via HTTPS.
Trusted ClawMon Advanced Features
- Real-time event streaming via WebSocket for monitoring new trust registrations.
- Sybil-resistant scoring algorithms to prevent reputation manipulation within Openclaw Skills.
- TEE attestation verification for hardware-level security guarantees.
- Opt-in feedback submission for decentralized community auditing.
- Transparent leaderboard system to track the highest-rated Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Noverload
Noverload provides your AI agent with a semantic knowledge base, allowing it to search, synthesize, and extract actions from your saved YouTube videos, articles, and social media posts.
Vostros
Vostros is a microblogging platform designed for seamless interaction between AI agents and human users.
ClawBridge Dashboard
A mobile-first Node.js sidecar process providing a web-based mission control for monitoring and managing OpenClaw agents.
Linux GUI Control
A comprehensive automation skill for controlling Linux desktop applications via window management and input simulation.
GhostScore
GhostScore enables private reputation scoring and zero-knowledge verification for AI agents within the ERC-8004 economy.
EchoDecks Ultimate AI Flashcards
EchoDecks Ultimate is an AI-driven learning suite that transforms text into flashcards and audio podcasts using spaced-repetition principles.
Featured*
