Skip to content

MaxMind minFraud MCP by usefulapi

by usefulapi.io · remote · no sign-in to list tools · v1.16.4 · published by io.usefulapi

MCP server for MaxMind minFraud — score transactions for fraud risk from your AI assistant.

Deletes · 1Final · 3
CompareRegistry entry

4 tools: 1 can delete or cancel things and 3 do something that can't be undone.

Scanned Oct 22, 2026

3 checks matched something in the tool definitions. No finding is not a safety guarantee. How the checks work.

claude mcp add --transport http maxmind-minfraud https://maxmind-minfraud.usefulapi.io/mcp

Findings

Matches of fixed checks, with the text they matched.

3 warn
  • Warnminfraud_score

    The server marks "minfraud_score" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

    Matched text

    annotations: {"readOnlyHint":true}; The input carries an amount (input.order.amount) and a currency, so it moves money.

    What this check looks for
  • Warnminfraud_insights

    The server marks "minfraud_insights" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

    Matched text

    annotations: {"readOnlyHint":true}; The input carries an amount (input.order.amount) and a currency, so it moves money.

    What this check looks for
  • Warnminfraud_factors

    The server marks "minfraud_factors" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

    Matched text

    annotations: {"readOnlyHint":true}; The input carries an amount (input.order.amount) and a currency, so it moves money.

    What this check looks for

Tools

Open a tool for its reasons and full description. Descriptions are the publisher's own words, shown as they are.

4 tools

Deletes or cancels 1Claude Code: deny
  • minfraud_report_transactionWRITE — submits fraud feedback to MaxMind.No findings

    Why it is in this group

    • The description's first verb is "write": a word for a change.
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    WRITE — submits fraud feedback to MaxMind. Report the real-world outcome of a transaction (e.g. it was a chargeback or confirmed fraud) so minFraud can improve future scoring. This mutates MaxMind's model of your traffic; it is not a read. Requires a `tag` and at least one identifier (ip_address, maxmind_id, minfraud_id, or transaction_id). API: POST /minfraud/v2.0/transactions/report (returns HTTP 204).

    Annotations the server declares

    • title: Report a transaction outcome
    • destructiveHint: true
Can't be undone 3Claude Code: ask
  • minfraud_factorsThe most detailed assessment: everything in Insights PLUS risk_score_reasons and per-factor subscores explaining WHY the score is what it is.Can send data out1 finding

    Why it is in this group

    • The input carries an amount (input.order.amount) and a currency, so it moves money.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • outboundThe input takes a destination (input.billing.phone_number): what the tool writes goes there.

    Findings

    • Warn

      The server marks "minfraud_factors" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

      Matched text

      annotations: {"readOnlyHint":true}; The input carries an amount (input.order.amount) and a currency, so it moves money.

      What this check looks for
    Description, as the server wrote it
    The most detailed assessment: everything in Insights PLUS risk_score_reasons and per-factor subscores explaining WHY the score is what it is. Same request body as minfraud_score. API: POST /minfraud/v2.0/factors.

    Annotations the server declares

    • title: minFraud Factors (risk reasons + subscores)
    • readOnlyHint: true
  • minfraud_insightsScore a transaction AND get detailed risk insights: IP geolocation, email/address/device attributes, and per-component risk on top of the risk_score.Can send data out1 finding

    Why it is in this group

    • The input carries an amount (input.order.amount) and a currency, so it moves money.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • outboundThe input takes a destination (input.billing.phone_number): what the tool writes goes there.

    Findings

    • Warn

      The server marks "minfraud_insights" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

      Matched text

      annotations: {"readOnlyHint":true}; The input carries an amount (input.order.amount) and a currency, so it moves money.

      What this check looks for
    Description, as the server wrote it
    Score a transaction AND get detailed risk insights: IP geolocation, email/address/device attributes, and per-component risk on top of the risk_score. Same request body as minfraud_score. API: POST /minfraud/v2.0/insights.

    Annotations the server declares

    • title: minFraud Insights (detailed risk)
    • readOnlyHint: true
  • minfraud_scoreScore a transaction for fraud risk.Can send data out1 finding

    Why it is in this group

    • The input carries an amount (input.order.amount) and a currency, so it moves money.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • outboundThe input takes a destination (input.billing.phone_number): what the tool writes goes there.

    Findings

    • Warn

      The server marks "minfraud_score" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

      Matched text

      annotations: {"readOnlyHint":true}; The input carries an amount (input.order.amount) and a currency, so it moves money.

      What this check looks for
    Description, as the server wrote it
    Score a transaction for fraud risk. Returns the overall risk_score (0.01–99), IP risk, disposition, and remaining query funds — the lightest, cheapest assessment. Send at least device.ip_address (plus email/billing/order for a better score). API: POST /minfraud/v2.0/score.

    Annotations the server declares

    • title: minFraud risk score
    • readOnlyHint: true

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or untrusted-input.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write0
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out3
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Install

Pinned to version 1.16.4. Values in angle brackets are yours to fill in.

claude mcp add --transport http maxmind-minfraud https://maxmind-minfraud.usefulapi.io/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [],
    "ask": [
      "mcp__maxmind-minfraud__minfraud_score",
      "mcp__maxmind-minfraud__minfraud_insights",
      "mcp__maxmind-minfraud__minfraud_factors"
    ],
    "deny": [
      "mcp__maxmind-minfraud__minfraud_report_transaction"
    ]
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as maxmind-minfraud. Review it before you use it.

Capability tags behind 3 rules
  • mcp__maxmind-minfraud__minfraud_score outbound
  • mcp__maxmind-minfraud__minfraud_insights outbound
  • mcp__maxmind-minfraud__minfraud_factors outbound