
Agentradar
Wraps the AgentRadar API to bring trust scoring and scam detection for blockchain agents into Claude or Cursor. You get 18 tools covering verification (six signal checks per agent), batch scoring, side-by-side comparisons, and EAS attestation writes on Base. The verify_agent tool runs full trust checks, attest_agent writes on-chain attestations for ERC-8004 agents, and check_scam queries their scam database. Most reads are free, verification costs $0.005 via x402 micropayments, attestations run $5 unless you supply an operator key. Useful if you're building or auditing AI agents on Base and want trust data without leaving your editor.
- tools
- 18
- auth
- Secret needed
- transport
- STDIO
- env vars
- 2
- last updated
- May 15, 2026
Tools
18 toolsPublic tool metadata for what this MCP can expose to an agent.
verify_agentFull 6-signal trust verification
Full 6-signal trust verification
No parameter schema in public metadata yet.
compare_agentsSide-by-side comparison of two agents
Side-by-side comparison of two agents
No parameter schema in public metadata yet.
batch_verifyVerify up to 10 agents at once
Verify up to 10 agents at once
No parameter schema in public metadata yet.
get_scoreQuick cached trust score lookup
Quick cached trust score lookup
No parameter schema in public metadata yet.
get_badge_urlGenerate embeddable trust badge URL
Generate embeddable trust badge URL
No parameter schema in public metadata yet.
attest_agentScore + write on-chain EAS attestation
Score + write on-chain EAS attestation
No parameter schema in public metadata yet.
get_attestationsList all attestations
List all attestations
No parameter schema in public metadata yet.
get_attestation_by_txLook up attestation by tx hash
Look up attestation by tx hash
No parameter schema in public metadata yet.
lookup_identityOn-chain identity (ERC-8004)
On-chain identity (ERC-8004)
No parameter schema in public metadata yet.
lookup_reputationReputation signals
Reputation signals
No parameter schema in public metadata yet.
resolve_agent_uriResolve agent URI/metadata
Resolve agent URI/metadata
No parameter schema in public metadata yet.
check_scamCheck scam database
Check scam database
No parameter schema in public metadata yet.
add_scam_walletReport malicious wallet
Report malicious wallet
No parameter schema in public metadata yet.
explain_scoreHuman-readable score explanation
Human-readable score explanation
No parameter schema in public metadata yet.
get_healthAPI health check
API health check
No parameter schema in public metadata yet.
get_statsDashboard statistics
Dashboard statistics
No parameter schema in public metadata yet.
list_signalsList scoring signals + weights
List scoring signals + weights
No parameter schema in public metadata yet.
get_pricingAPI pricing info
API pricing info
No parameter schema in public metadata yet.
Sponsors
Configuration
AGENTRADAR_API_URLAgentRadar API base URL. Defaults to https://api.vvpro.ai (production, Base mainnet).
AGENTRADAR_OPERATOR_KEYsecretOperator key for admin endpoints and bypassing x402 payment. Optional — leave empty for public, x402-paid usage.
Readme
@agentradar/mcp
MCP Server for AgentRadar — the on-chain trust oracle for the agent economy. Provides 18 tools for verifying, scoring, and attesting AI agent trustworthiness directly from Claude, Cursor, or any MCP-compatible client.
- Live API: https://api.vvpro.ai
- Web: https://vvpro.ai
- MCP Registry:
io.github.Bichev/agentradar
This package is a thin client wrapper around the public AgentRadar API. The full source is auditable here — anything npx @agentradar/mcp runs is in this repo, MIT-licensed.
Tools
Details
- Repository
- github.com/bichev/agentradar-mcp
- Package
@agentradar/mcp- Transport
- STDIO
- Auth
- Required
- Registry
- MCP registry, active