Agentradar Verify

Name: Agentradar Verify
Author: Bichev

@Bichev

リポジトリ

Agentradar Verify について

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

#x402 #erc-8004 #security #base

基本情報

カテゴリ

その他

トランスポート

stdio

公開者

Bichev

投稿者

Vladimir Bichev

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "agentradar": {
      "command": "npx",
      "args": [
        "-y",
        "@agentradar/mcp"
      ],
      "env": {
        "AGENTRADAR_API_URL": "https://api.vvpro.ai"
      }
    }
  }
}

ツール

Full 6-signal trust verification

Side-by-side comparison of two agents

Verify up to 10 agents at once

Quick cached trust score lookup

Generate embeddable trust badge URL

Score + write on-chain EAS attestation

List all attestations

Look up attestation by tx hash

On-chain identity (ERC-8004)

Reputation signals

Resolve agent URI/metadata

Check scam database

Report malicious wallet

Human-readable score explanation

API health check

Dashboard statistics

List scoring signals + weights

API pricing info

概要

What is Agentradar Verify?

Agentradar Verify is an MCP server that provides 18 tools for verifying, scoring, and attesting AI agent trustworthiness directly from Claude, Cursor, or any MCP-compatible client. It is a thin, MIT-licensed client wrapper around the public AgentRadar API, described as the on-chain trust oracle for the agent economy.

How to use Agentradar Verify?

Install and configure via npx or local build. In Cursor, add a stanza to .cursor/mcp.json using command "npx -y @agentradar/mcp" with the optional AGENTRADAR_OPERATOR_KEY environment variable. For Claude Desktop, add the same entry to claude_desktop_config.json. Run directly with npx @agentradar/mcp or clone, install, build, and run from source.

Key features of Agentradar Verify

Full 6‑signal trust verification
Side‑by‑side comparison of two agents
Batch verify up to 10 agents at once
Generate embeddable trust badge URLs
On‑chain EAS attestation of agent scores
Scam database lookup and reporting

Use cases of Agentradar Verify

Verify an AI agent by its blockchain address
Check if a wallet address is a known scam
Compare two agents to decide which is more trustworthy
Get a human‑readable explanation of a low trust score
Write an on‑chain attestation for a verified agent

FAQ from Agentradar Verify

What is the AGENTRADAR_OPERATOR_KEY used for?

It is an optional environment variable that grants access to admin endpoints and bypasses the x402 per‑call payment for paid tools.

Are there any free tools?

Yes. get_score is always free. Other pricing: verify_agent costs $0.005 per call (or free with operator key), attest_agent costs $5.00 per attestation (or free with operator key), and admin endpoints require the operator key.

Where does the data come from?

All data is served by the live API at https://api.vvpro.ai. The server is a thin client wrapper; the full auditable source code is in the repository under the MIT license.

How do I set up the server in Cursor or Claude Desktop?

Add a JSON entry to the corresponding MCP configuration file. The command is npx -y @agentradar/mcp, and you can optionally set the AGENTRADAR_OPERATOR_KEY environment variable.