MCP.Market

Agentradar MCP server

by Bichev·io.github.Bichev/agentradar·v0.1.3·1 stars
No reviews yet

Trust scoring, scam detection, and EAS attestations for ERC-8004 + x402 agents on Base.

B72/100grade B
Write a reviewInstall
What users say
No reviews yet
Be the first
Safety scan
B72/100

full report

Adoption
New

1 stars56 downloads/wk

Reviews

Write one

Nobody has reviewed Agentradar yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Agentradar tools (18, 1 write)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • add_scam_wallet

    Report a malicious wallet address to the AgentRadar scam database. Requires operator key. Include chain, severity level, source of intel, and optional notes.

  • attest_agent

    Score an AI agent and write an on-chain EAS attestation recording the trust score permanently. Costs $5 via x402 or requires an operator key. Use when you need a verifiable, immutable trust record for an agent.

  • batch_verify

    Verify multiple AI agent addresses in a single call (max 10). Returns trust scores for each address. Efficient for evaluating a list of candidate agents.

  • check_scam

    Check if a wallet address is in the AgentRadar scam database. Returns matching scam records with severity, chain, source, and notes. Use this as a quick safety check before any transaction.

  • compare_agents

    Compare two AI agents side-by-side on trust scores and security signals. Useful for choosing between competing agents that offer similar capabilities.

  • explain_score

    Get a human-readable explanation of an agent's trust score. Returns a narrative breakdown of each signal's contribution, risk factors, and actionable recommendations. Best for presenting trust analysis to end users.

  • get_attestation_by_tx

    Look up a specific attestation by its transaction hash. Returns the full attestation record including the scored address, trust score, and on-chain details.

  • get_attestations

    List all on-chain trust attestations written by AgentRadar. Returns paginated attestation records including addresses, scores, and transaction hashes. Requires operator key.

  • get_badge_url

    Generate a trust badge URL for an AI agent. Returns an embeddable SVG badge showing the agent's trust score. Useful for displaying trust status in UIs or documentation.

  • get_health

    Check AgentRadar API health and connectivity status. Returns service status, uptime, and version info. Use to verify the API is reachable before making other calls.

  • get_pricing

    Get AgentRadar API pricing information. Shows costs for each endpoint including x402 micropayment amounts and free tier details.

  • get_score

    Get a cached trust score for an AI agent address. Faster than full verification — returns the last computed score without re-running signals. Free endpoint, no payment required.

  • get_stats

    Get AgentRadar dashboard statistics — total verifications, attestations written, scam wallets tracked, and API usage metrics. Requires operator key.

  • list_signals

    List all trust scoring signals used by AgentRadar with their weights and descriptions. Helps understand how the composite trust score is calculated.

  • lookup_identity

    Look up an AI agent's on-chain identity (ERC-8004 registration, ENS name, contract metadata). Returns who the agent claims to be and whether those claims are verified.

  • lookup_reputation

    Look up an AI agent's on-chain reputation — ERC-8004 feedback history, attestation count, and community signals. Shows how established and trusted the agent is in the ecosystem.

  • resolve_agent_uri

    Resolve an agent address to its registered URI, metadata, and service endpoints. Useful for discovering how to interact with a verified agent.

  • verify_agentwrite action

    Run a full 6-signal trust verification on an AI agent or wallet address. Returns composite trust score (0-100), individual signal scores, risk flags, and recommendations. Use this before transacting with or delegating to an unknown agent.

Public scan report

scanner v0.1.9 · 2026-09-27 · same rubric, same numbers if you re-run it

no findings
  • Code scan21 source files scanned25/25
  • –Live reliabilityno gateway calls yet and no remote to proben/a
  • –Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitystatic API keys via environment variables6/15
  • Maintenancelast push 103 days ago8/15
  • Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
Overall 72/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How Agentradar compares

Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
72/100
higher than 38% of 48,255 servers
Tools
18
the typical server has 7
Tools that change things
6%
typical: 0%. Fewer write tools means less an agent can break
GitHub stars
1
more than 65% of 28,688 servers
Downloads a week
56
more than 46% of 12,191 servers with a package

What its tools let an agent do: 16 read · 1 create · 1 change

Install Agentradar in Claude Code, Cursor or VS Code

Runs npx -y @agentradar/mcp on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add agentradar -- npx -y @agentradar/mcp
Add to Cursor

What the publisher says

From the Agentradar repository's README, as published. We do not edit it. Read it on GitHub

@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

Installation

Cursor

Add to your project's .cursor/mcp.json:

{
  "mcpServers": {
    "agentradar": {
      "command": "npx",
      "args": ["-y", "@agentradar/mcp"],
      "env": {
        "AGENTRADAR_OPERATOR_KEY": "your-key-here"
      }
    }
  }
}

Claude Desktop

Add to ~/Library/Application Support/Claude/claudedesktopconfig.json (macOS) or %APPDATA%\Claude\claudedesktopconfig.json (Windows):

{
  "mcpServers": {
    "agentradar": {
      "command": "npx",
      "args": ["-y", "@agentradar/mcp"],
      "env": {
        "AGENTRADAR_OPERATOR_KEY": "your-key-here"
      }
    }
  }
}

npx (direct)

npx @agentradar/mcp

Local development

git clone https://github.com/Bichev/agentradar-mcp.git
cd agentradar-mcp
npm install
npm run build
node dist/index.js

Environment Variables

Usage Examples

Once connected, you can ask your AI assistant:

  • "Verify this agent: 0x742d35Cc6634C0532925a3b844Bc9e7595f2bD28"
  • "Is 0xdead...beef a known scam wallet?"
  • "Compare these two agents and tell me which is more trustworthy"
  • "Explain why this agent scored 45/100"
  • "Write an on-chain attestation for this verified agent"

Pricing

  • verifyagent**: $0.005 per call (x402) or free with operator key
  • attestagent**: $5.00 per attestation (x402) or free with operator key
  • getscore**: Free
  • Admin endpoints: Operator key required

License

MIT

Nothing above is checked by us. What we check is on the safety report.

Agentradar: common questions

Is Agentradar MCP server safe?
Mostly: it is graded B (72/100). Read the Agentradar safety report
How do I install Agentradar?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does Agentradar need an API key?
Yes. The registry entry asks for AGENTRADAR_OPERATOR_KEY.
Is Agentradar maintained?
The last commit was 140 days ago (2026-06-17). The latest release is v0.1.3.
How does Agentradar compare with other MCP servers?
safety score 72/100 (higher than 38% of 48,255 servers); tools 18 (the typical server has 7); tools that change things 6% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.