A comprehensive trust and reputation oracle for verifying the credibility of ERC-8004 registered AI agents before interaction or payment.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agentproof
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 agentproof 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 AgentProof Trust Oracle?
AgentProof is a specialized trust oracle designed to bring transparency to the decentralized agent ecosystem. By analyzing over 25,000 agents across major networks like Ethereum, Avalanche, Base, and Linea, it provides real-time reputation scoring based on an advanced 8-factor trust model. This skill is essential for users of Openclaw Skills who need to validate the safety and reliability of third-party agents before committing funds or sharing data.
The system categorizes agents into trust tiers ranging from Unranked to Platinum, allowing for instant decision-making. By surfacing critical risk flags such as frequent endpoint changes or suspicious deployment patterns, AgentProof ensures that your automated workflows remain secure and that you only transact with high-integrity AI entities.
AgentProof Trust Oracle Use Cases
- Verifying an agent's reputation before sending payments via x402 protocols.
- Auditing the publisher of a new skill before installation to ensure security.
- Comparing multiple available agents to select the highest-rated performer for a specific task.
- Monitoring existing agent integrations for sudden changes in URI stability or trust scores.
- Researching deployer history to identify serial deployers or low-quality agent farms.
How AgentProof Trust Oracle Works
- The user or system provides an agent ID (ERC-8004 token ID) or address to the AgentProof skill.
- The skill queries the AgentProof oracle API to retrieve the composite trust score and 8-factor breakdown.
- The system checks for active risk flags like NEW_IDENTITY or SERIAL_DEPLOYER that might mitigate the raw score.
- Deployer reputation data is cross-referenced to determine if the agent inherits credibility from its parent entity.
- A final trust tier is assigned (Bronze to Platinum), providing a clear recommendation for the interaction.
AgentProof Trust Oracle Setup
To use this skill within your environment, ensure you have curl installed. You can interact with the oracle directly via the CLI:
# Install dependencies (Linux/macOS)
sudo apt-get install curl jq
# Query an agent's trust status
curl -s "https://oracle.agentproof.sh/api/v1/trust/{agent_id}" | jq .
AgentProof Trust Oracle Data Schema & Taxonomy
The AgentProof skill returns structured JSON data containing the following metadata taxonomy:
| Field | Description | Type |
|---|---|---|
composite_score |
The primary reputation value (0-100) | Integer |
trust_tier |
Label based on score (e.g., Gold, Platinum) | String |
risk_flags |
List of warnings (e.g., FREQUENT_URI_CHANGES) | Array |
score_breakdown |
Individual scores for feedback, age, and uptime | Object |
deployer_info |
Reputation and agent count of the creator | Object |
AgentProof Trust Oracle Advanced Features
- Multi-chain Leaderboards: Access top-ranked agents across Ethereum, Avalanche, Base, and Linea.
- Deployer Analytics: Evaluate the entire portfolio and abandonment ratio of an agent publisher.
- Network Statistics: Real-time insights into total indexed agents and scoring distributions across the ecosystem.
- Automated Decision Frameworks: Programmatic logic to block transactions if trust scores fall below a custom threshold.
SKILL.md
Loading
Related Openclaw Skills
A2A Platform Integration
A comprehensive protocol skill that teaches AI agents how to discover and interact with the A2A B2B Agent platform for seamless cross-agent communication.
Enoch Tuning
A production-grade configuration framework that provides AI agents with persistent identity, memory architecture, and rigorous security protocols.
Memory Tiers
A sophisticated tiered memory management system that optimizes AI agent context windows using a hierarchy of hot, warm, and cold storage tiers.
Academic Paper Mentor
An expert research guide designed to mentor scholars through a six-stage workflow for publishing in world-class academic journals.
clawdev
A specialized publishing skill that allows AI agents to draft and submit technical tutorials, guides, and snippets directly to the clawdev.to developer community.
Brand Voice Extractor
A professional tool to extract or build a distinct brand voice profile that ensures AI agents produce human-sounding, consistent content.
Featured*
