Model Context Protocol

# Plug the case file into your AI agent.

This site runs a free, no-auth remote MCP server. Any MCP-capable agent — Claude Code, Claude Desktop, Cursor, or your own — can search the case file, look up numbers, and pull live statistics.

Endpoint: `https://scammercasefile.com/mcp` · Streamable HTTP · read-only

6 read-only tools, 0 API keys, 3 discovery endpoints (this page, the Server Card, and the well-known catalog) — serving live report data since 2026-07-22.

## Tools

### search\_scammer\_reports

Search the WhatsApp Scammer Tracker case file by name or phone number. Returns self-reported, unverified scam reports -- not confirmed fraud.

### lookup\_scammer\_by\_number

Look up every report filed against one exact WhatsApp phone number.

### list\_reports\_by\_country

List scam reports whose phone number resolves to one country (ISO 3166-1 alpha-2 code, e.g. "NG" or "US"). Country is derived from the number's calling code at submission time. Reports are self-reported and unverified.

### list\_reports\_by\_scam\_type

List scam reports filed under one scam type. Valid types: Wrong Number / Pig Butchering, Romance Scam, Investment / Crypto Scam, Fake Job Offer, Task / Part-Time Job Scam, Advance Fee (Visa/Loan/Customs), Money / Currency Exchange Scam, Phishing Link, Impersonation (Bank/Government), OTP / Verification Code Scam, Lottery / Prize Scam, Package Delivery Scam, Tech Support Scam, Sextortion / Blackmail, Extortion / Threats, Other. Slugs are also accepted (e.g. "romance-scam").

### get\_scam\_trends

Aggregate statistics computed live from the whole case file: total report count, reports by scam type, reports by country, and the most recent report date. The same numbers as the /trends page.

### get\_recent\_reports

The newest scam reports across the whole case file, newest first. Useful for "what was reported recently" questions.

## Connect your client

### Claude Code

```
claude mcp add --transport http scammercasefile https://scammercasefile.com/mcp
```

### Claude Desktop / claude.ai

Settings → Connectors → Add custom connector, then paste the endpoint URL:

```
https://scammercasefile.com/mcp
```

### Cursor / any mcp.json client

```
{
  "mcpServers": {
    "scammercasefile": {
      "url": "https://scammercasefile.com/mcp"
    }
  }
}
```

### Raw JSON-RPC (no SDK)

```
curl -X POST https://scammercasefile.com/mcp \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{
    "name":"lookup_scammer_by_number",
    "arguments":{"phoneNumber":"+17605969234"}}}'
```

Machine-readable discovery: the [Server Card](https://scammercasefile.com/mcp/server-card) and [/.well-known/mcp/catalog.json](https://scammercasefile.com/.well-known/mcp/catalog.json).

## Prompt for your agent

Paste this into your agent's system prompt (or project instructions) so it uses the tools correctly and presents results responsibly:

```
You have access to the WhatsApp Scammer Tracker MCP server
(https://scammercasefile.com/mcp). It exposes a community-maintained log of
self-reported WhatsApp scam reports.

When a user asks about a suspicious WhatsApp number or contact:
1. Use lookup_scammer_by_number for an exact phone number, or
   search_scammer_reports for a name / partial number.
2. Use list_reports_by_country or list_reports_by_scam_type to explore
   patterns, and get_scam_trends for overall statistics.
3. Always tell the user that reports are self-submitted and unverified:
   a match is a warning sign, not proof; no match is not proof of safety.
4. Never advise sending money, codes, or personal data to an unverified
   contact. Suggest filing a report at
   https://scammercasefile.com/report when the user confirms a scam attempt.
```

## What MCP is, per the standard itself

> "MCP (Model Context Protocol) is an open-source standard for connecting AI applications to external systems. … Think of MCP like a USB-C port for AI applications."

— [modelcontextprotocol.io](https://modelcontextprotocol.io/). This server implements the specification's [Streamable HTTP transport](https://modelcontextprotocol.io/specification/2025-06-18/basic/transports) (2025-06-18 revision): a single endpoint accepting JSON-RPC 2.0 over HTTP POST.

## Frequently asked questions

### Do I need an API key or account?

No. The MCP endpoint is public, free, and unauthenticated — connect and call tools directly. All tools are read-only.

### Can an agent file a report through MCP?

No, by design. Filing a report creates a public record naming a phone number, so it stays a human action through the Report a Scammer form. Every MCP tool is read-only.

### How fresh is the data the tools return?

Live. Every tool call queries the same Cloudflare D1 database that serves the website, so results reflect reports the moment they are filed.

### Is the data verified?

No. Reports are self-submitted by the people who filed them and are not independently confirmed. Agents should always present results as community-reported, not as confirmed fraud.

## Been targeted by a scammer?

File a report with the number, name, and what happened — it's published immediately and warns the next person who searches.

[Report a scammer](https://scammercasefile.com/report)