MaxMind minFraud MCP server. MCP server for MaxMind minFraud — score transactions for fraud risk from your AI assistant.
Add to your MCP config, then reload & authorize:
{
"mcpServers": {
"maxmind-minfraud": {
"url": "https://maxmind-minfraud.usefulapi.io/mcp"
}
}
}| Tool | Type | What it does |
|---|---|---|
minfraud_score | read | Minfraud score Score a single transaction for fraud risk and get back the overall risk score (0.01–99), IP risk, a recommended disposition, and remaining query funds — the lightest, cheapest minFraud assessment. |
minfraud_insights | read | Minfraud insights Score a transaction and get detailed risk insights alongside the score — IP geolocation plus email, address, and device attributes with per-component risk — for richer fraud investigation. |
minfraud_factors | read | Minfraud factors Run the most detailed minFraud assessment, returning everything in Insights plus risk-score reasons and per-factor subscores that explain exactly why a transaction scored the way it did. |
minfraud_report_transaction | write | Minfraud report transaction Report a transaction's real-world outcome (chargeback, confirmed fraud, spam or abuse, or cleared) back to MaxMind so minFraud learns from your traffic and sharpens future scoring. |
maxmind_minfraud_usage_status | Usage status (free-tier meter) Report the caller's current free-tier usage this month: calls used, monthly limit, remaining, and whether the cap is reached. Read-only; does not count against the meter. | |
maxmind_minfraud_upgrade | Upgrade to Pro (unlimited) Subscribe to the Pro plan for UNLIMITED MaxMind minFraud tool calls (the free tier caps monthly usage). Choose monthly ($9/month) or yearly ($90/year — 2 months free) billing. Returns a Stripe Checkout link to open in your browser; after payment your account upgrades automatically. Read-only; does not count against the meter. |
| Plan | Price | Limit |
|---|---|---|
| Free | $0 | 100 tool calls / month |
| Pro | $9/mo · $90/yr | Unlimited |
This is a Model Context Protocol endpoint — meant to be connected from an AI client, not opened in a browser. An invalid_token response at the URL is the auth gate working as designed; clients authenticate automatically.