Trust Api MCP server

What it does

This MCP server exposes LayerCall's fraud scoring API, allowing you to assess the risk level of IPs, emails, phone numbers, domains, and devices in a single call. It returns a risk score (0-100), a verdict (allow/review/block), and the signals behind the decision. Intended for developers building signup flows, payment systems, and other user-facing features where fraud detection is needed.

• Score an IP address to detect VPNs, proxies, Tor nodes, and datacenters during user signup.
• Verify an email address for syntax errors, disposability, and domain age to filter low-quality accounts.
• Evaluate a phone number for validity across numbering plans and detect line type (mobile, landline, VOIP).

Tools this server exposes

Tool	Does
score_ip	Score an IP address for fraud risk, returning VPN/proxy/Tor/datacenter status, geolocation, and ASN.
verify_email	Verify an email address for syntax, MX records, disposability, role accounts, and domain age.
lookup_phone	Validate a phone number worldwide and identify line type (mobile, landline, VOIP).
score_domain	Score a domain for fraud risk using RDAP age, registrar information, and mail authentication records.
score_device	Score a device fingerprint for reputation and estimate bot probability.
score_user	Score an entire signup by combining IP, email, phone, and device signals into a single verdict.
verify_agent	Verify that an AI agent is who it claims using Web Bot Auth (RFC 9421).

Before you enable it

An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.