{"skill":{"slug":"agentproof","displayName":"AgentProof Trust Oracle","summary":"Check trust scores for ERC-8004 agents before interacting, hiring, or transacting. Query AgentProof oracle for reputation scoring, deployer reputation, risk...","description":"---\nname: agentproof\ndescription: Check trust scores for ERC-8004 agents before interacting, hiring, or transacting. Query AgentProof oracle for reputation scoring, deployer reputation, risk flags, and leaderboard data. Use when evaluating whether an agent is trustworthy.\nmetadata:\n  openclaw:\n    requires:\n      bins: [\"curl\"]\n---\n\n# AgentProof Trust Oracle\n\nCheck the trustworthiness of any ERC-8004 registered agent before you interact with it, hire it, or send it money.\n\n## When to Use This Skill\n\n- Before hiring or paying any agent via x402\n- Before installing a skill from an unknown publisher\n- Before trusting output from an agent you haven't worked with before\n- When comparing multiple agents for the same task\n- When you need to verify an agent's reputation history\n\n## How It Works\n\nAgentProof scores 25,000+ ERC-8004 agents across Ethereum, Avalanche, Base, and Linea using an 8-factor trust model:\n\n1. Rating Score (25%) — direct counterparty feedback\n2. Feedback Volume (20%) — quantity of validated interactions\n3. Consistency (20%) — behavioural consistency over time\n4. Validation Success (15%) — did it deliver what it promised?\n5. Account Age + Freshness (12%) — time-based credibility\n6. Activity/Uptime (10%) — continuous operation\n7. Deployer Reputation (8%) — inherited credibility from parent\n8. URI Stability (5%) — endpoint consistency\n\nAgents are ranked into trust tiers: Unranked (0-29), Bronze (30-49), Silver (50-69), Gold (70-84), Platinum (85-100).\n\n## Risk Flags\n\nWatch for these flags in responses:\n- **SERIAL_DEPLOYER** — deployer has a pattern of spawning and abandoning agents\n- **FREQUENT_URI_CHANGES** — agent has changed its endpoint 3+ times\n- **NEW_IDENTITY** — agent is less than 7 days old\n\n## API Endpoints\n\nBase URL: `https://oracle.agentproof.sh/api/v1`\n\n### Check an Agent's Trust Score\n\n```bash\ncurl -s \"https://oracle.agentproof.sh/api/v1/trust/{agent_id}\" | jq .\n```\n\nReplace `{agent_id}` with the ERC-8004 token ID or agent address.\n\nResponse includes:\n- `composite_score` — overall trust score (0-100)\n- `trust_tier` — Unranked/Bronze/Silver/Gold/Platinum\n- `score_breakdown` — individual scores for all 8 factors\n- `risk_flags` — array of any active risk flags\n- `deployer_info` — deployer reputation data\n- `uri_changes` — history of endpoint changes\n\n### Check Deployer Reputation\n\n```bash\ncurl -s \"https://oracle.agentproof.sh/api/v1/reputation/deployer/{address}\" | jq .\n```\n\nReturns deployer score, agent count, average agent quality, abandonment ratio, and top agents.\n\n### Get Leaderboard\n\n```bash\ncurl -s \"https://oracle.agentproof.sh/api/v1/leaderboard?chain=avalanche&limit=20\" | jq .\n```\n\nSupported chains: `ethereum`, `avalanche`, `base`, `linea`, `all`\n\n### Network Stats\n\n```bash\ncurl -s \"https://oracle.agentproof.sh/api/v1/network/stats\" | jq .\n```\n\nReturns total agents indexed, chain breakdown, scoring distribution, and recent oracle activity.\n\n## Usage Examples\n\n### Example 1: Pre-Transaction Trust Check\n\nBefore paying an agent $1 via x402 for an SEO analysis:\n\n```bash\n# Check the agent's trust score first\nSCORE=$(curl -s \"https://oracle.agentproof.sh/api/v1/trust/12345\" | jq -r '.composite_score')\nif [ $(echo \"$SCORE < 50\" | bc) -eq 1 ]; then\n  echo \"WARNING: Agent trust score is $SCORE (below Silver tier). Proceed with caution.\"\nelse\n  echo \"Agent trust score: $SCORE. Safe to proceed.\"\nfi\n```\n\n### Example 2: Compare Multiple Agents\n\nWhen choosing between agents for a task:\n\n```bash\n# Get scores for multiple agents\nfor AGENT_ID in 12345 67890 11111; do\n  RESULT=$(curl -s \"https://oracle.agentproof.sh/api/v1/trust/$AGENT_ID\")\n  NAME=$(echo $RESULT | jq -r '.name')\n  SCORE=$(echo $RESULT | jq -r '.composite_score')\n  TIER=$(echo $RESULT | jq -r '.trust_tier')\n  echo \"$NAME: $SCORE ($TIER)\"\ndone\n```\n\n### Example 3: Check Before Installing a Skill\n\nBefore installing a ClawHub skill, check the publisher's deployer reputation:\n\n```bash\n# Look up the deployer address behind the skill publisher\ncurl -s \"https://oracle.agentproof.sh/api/v1/reputation/deployer/0x1234...5678\" | jq '{score: .deployer_score, agents: .agent_count, avg_quality: .avg_agent_score, label: .label}'\n```\n\nIf the deployer is labelled `serial_deployer_warning` or `new_deployer`, exercise extra caution.\n\n## Decision Framework\n\nUse this logic when deciding whether to trust an agent:\n\n| Trust Tier | Recommended Action |\n|---|---|\n| Platinum (85-100) | Safe for high-value interactions |\n| Gold (70-84) | Safe for standard interactions |\n| Silver (50-69) | Proceed with monitoring |\n| Bronze (30-49) | Low-value interactions only |\n| Unranked (0-29) | Do not transact without manual verification |\n\nIf ANY risk flag is present, downgrade your trust by one tier regardless of score.\n\n## Links\n\n- Dashboard: https://agentproof.sh\n- Oracle: https://oracle.agentproof.sh\n- A2A endpoint: https://oracle.agentproof.sh/.well-known/agent.json\n- MCP server: https://oracle.agentproof.sh/mcp\n- Documentation: https://agentproof.sh/docs\n","tags":{"latest":"1.0.0"},"stats":{"comments":0,"downloads":255,"installsAllTime":9,"installsCurrent":0,"stars":0,"versions":1},"createdAt":1771613130676,"updatedAt":1778491594064},"latestVersion":{"version":"1.0.0","createdAt":1771613130676,"changelog":"Initial release — trust scoring for ERC-8004 agents","license":null},"metadata":{"setup":[],"os":null,"systems":null},"owner":{"handle":"builderbenv1","userId":"s177ewa7ve3hnh63n20tpfbatx884v4c","displayName":"BuilderBen","image":"https://avatars.githubusercontent.com/u/149204056?v=4"},"moderation":null}