What it does
The server scores a counterparty from several inputs: a domain (WHOIS age, MX, SPF and DMARC records, SSL certificate), an IP address (geolocation, VPN, proxy or Tor detection, datacenter versus residential), an EVM wallet address (transaction count, wallet age, counterparty diversity), and a company name screened against OFAC SDN, EU consolidated and UN consolidated sanctions lists. Each request returns a score from 0 to 100 in one of four tiers, from low to critical. A score of 0 marks a hard match, such as an exact sanctions hit, and overrides all other signals. Named risk flags show which signals reduced the score, and the flag names are stable and machine-readable.
It is built for AI agents that transact via x402 on Base. An agent can check a seller or payee before paying, or before creating a job on the Virtuals Protocol Agent Commerce Protocol. A Safe Agent Payments wrapper runs the same check automatically before an x402 payment and can block, warn or only log when the score falls below a set threshold.
Each scored request costs $0.01 USDC on the Base network and is paid through x402, so the calling wallet must hold funded USDC. The documentation states that no separate API key or account is needed. At least one of domain, ip, wallet_address or company_name must be supplied, and the more inputs given, the higher the confidence. The health endpoint is free.
Limitations and requirements
Paid requests require a funded wallet holding USDC on the Base network, at $0.01 per request. At least one of domain, ip, wallet_address or company_name is required. The email field is accepted but not yet scored. The tool is informational only: it is not a compliance certification, legal advice or a guarantee of counterparty legitimacy, and the user remains responsible for transaction decisions. The publisher advises never hardcoding private keys and using environment variables or a secrets manager.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
io.github.AlexanderLawson17/revettr-mcp
Counterparty risk scoring for agentic commerce via x402 micropayments.
v0.2.3As shown on registry.modelcontextprotocol.io when we last read it