1. Conduid
  2. Security
  3. scamverifyai/scamverify-mcp
MCP server · Security

scamverifyai/scamverify-mcp

powered scam and threat verification MCP server. Check phone numbers, URLs, text messages, emails, documents, and QR codes against 8M+ threat intelligence records (FTC/FCC complaints, carrier analysis, URLhaus, ThreatFox). Returns risk scores, verdicts, and detailed signals. 10 tools, OAuth 2.1 + API key auth, Streamable HTTP transport.

Unclaimed last commit 6 months ago security
34Low

Scored 5 months ago · breakdown

About scamverifyai/scamverify-mcp

scamverifyai/scamverify-mcp is an MCP server published by scamverifyai in the Security category: powered scam and threat verification MCP server. Check phone numbers, URLs, text messages, emails, documents, and QR codes against 8M+ threat intelligence records (FTC/FCC complaints, carrier analysis, URLhaus, ThreatFox). Returns risk scores, verdicts, and detailed signals. 10 tools, OAuth 2.1 + API key auth, Streamable HTTP transport.

The repository has 1 stars and 0 forks, with the last commit 6 months ago. Six months or more without a commit doesn't mean the server is broken, but check the open issues before depending on it in production.

Install

From source
git clone https://github.com/scamverifyai/scamverify-mcp

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Tools (10)

batch_phoneLook up multiple phone numbers in a single request (max 100).

Parameters not published.

batch_urlCheck multiple URLs in a single request (max 100).

Parameters not published.

check_documentAnalyze a document image for scam indicators. Uses vision AI to extract entities (addresses, officials, citations, phone numbers) and verifies them against government databases.

Parameters not published.

check_emailAnalyze an email for phishing indicators. Checks sender domain, email headers (SPF/DKIM/DMARC), brand impersonation, embedded URLs and phone numbers.

Parameters not published.

check_phoneLook up a US phone number for scam reports, carrier info, robocall flags, and community reports. Returns risk score (0-100), verdict, and detailed signals from FTC, FCC, carrier, and community data.

Parameters not published.

check_qrScan a QR code image and verify its contents. Decodes the QR code server-side and, if it contains a URL, runs full URL verification.

Parameters not published.

check_textAnalyze a text/SMS message for scam indicators. Extracts and cross-references embedded phone numbers and URLs. AI identifies scam type, red flags, and risk level.

Parameters not published.

check_urlCheck a website URL for safety. Analyzes domain age, SSL certificate, redirect chains, brand impersonation, Google Web Risk, URLhaus, ThreatFox, and community reports.

Parameters not published.

get_statusCheck the operational status of ScamVerify API services. No authentication required.

Parameters not published.

get_usageCheck your current API usage quota and rate limits for the billing period.

Parameters not published.

Ask AI

✦

Ask AI about scamverifyai/scamverify-mcp

Powered by Claude · Grounded in docs

I know everything about scamverifyai/scamverify-mcp. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

  • ✓README presentThe repository ships documentation.

5 checks have not been run yet.

README

ScamVerify MCP Server

AI-powered scam and threat verification MCP server. Verify phone numbers, URLs, text messages, emails, documents, and QR codes against 10M+ threat intelligence records from FTC, FCC, URLhaus, ThreatFox, and community reports.

Server URL: https://scamverify.ai/api/mcp

Transport: Streamable HTTP (stateless) or stdio (local proxy)

Authentication: API key or OAuth 2.1 (PKCE)

Tools (10)

Tool Description
check_phone Look up a US phone number for scam reports, carrier info, robocall flags, and community reports. Returns risk score (0-100), verdict, and detailed signals from FTC, FCC, carrier, and community data.
check_url Check a website URL for safety. Analyzes domain age, SSL certificate, redirect chains, brand impersonation, Google Web Risk, URLhaus, ThreatFox, and community reports.
check_text Analyze a text/SMS message for scam indicators. Extracts and cross-references embedded phone numbers and URLs. AI identifies scam type, red flags, and risk level.
check_email Analyze an email for phishing indicators. Checks sender domain, email headers (SPF/DKIM/DMARC), brand impersonation, embedded URLs and phone numbers.
check_document Analyze a document image for scam indicators. Uses vision AI to extract entities (addresses, officials, citations, phone numbers) and verifies them against government databases.
check_qr Scan a QR code image and verify its contents. Decodes the QR code server-side and, if it contains a URL, runs full URL verification.
batch_phone Look up multiple phone numbers in a single request (max 100).
batch_url Check multiple URLs in a single request (max 100).
get_usage Check your current API usage quota and rate limits for the billing period.
get_status Check the operational status of ScamVerify API services. No authentication required.

Prompts (4)

Prompt Description
investigate_phone Investigate a suspicious phone number for scam indicators and provide a safety recommendation.
verify_url Check if a website URL is safe to visit and provide a detailed safety assessment.
analyze_text Analyze a suspicious text message for scam indicators and explain the findings.
check_email Analyze a suspicious email for phishing indicators and provide a safety assessment.

Data Sources

  • FTC Do Not Call Complaints - 9.7M+ records with scam type classification
  • FCC Consumer Complaints - 443K+ telecom violation reports
  • URLhaus - 74K+ active malicious URL indicators
  • ThreatFox - 60K+ IOCs (indicators of compromise)
  • Carrier Intelligence - Line type, CNAM, VoIP detection, high-risk carrier flagging
  • Community Reports - User-submitted scam reports with verification

Setup

Option 1: Remote Server (Recommended)

Connect directly to the hosted ScamVerify MCP server. No installation required.

Claude Desktop:

{
  "mcpServers": {
    "scamverify": {
      "type": "streamable-http",
      "url": "https://scamverify.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Claude Desktop (OAuth - no API key needed):

{
  "mcpServers": {
    "scamverify": {
      "type": "streamable-http",
      "url": "https://scamverify.ai/api/mcp"
    }
  }
}

When using without an API key, Claude Desktop will initiate the OAuth 2.1 flow. You'll be redirected to sign in at scamverify.ai and authorize access.

Cursor / Windsurf / Other MCP Clients:

{
  "type": "streamable-http",
  "url": "https://scamverify.ai/api/mcp",
  "headers": {
    "Authorization": "Bearer YOUR_API_KEY"
  }
}

ChatGPT:

ScamVerify is available as a ChatGPT connector. Add it through the ChatGPT plugin/connector settings using:

  • MCP URL: https://scamverify.ai/api/mcp
  • Auth: OAuth

Option 2: Local Proxy (stdio)

Run a local MCP server that proxies all tool calls to the ScamVerify API. Useful for MCP clients that only support stdio transport.

Install and run:

npm install -g @scamverifyai/scamverify-mcp
SCAMVERIFY_API_KEY=sv_live_... scamverify-mcp

Or run directly with npx:

SCAMVERIFY_API_KEY=sv_live_... npx @scamverifyai/scamverify-mcp

Claude Desktop (stdio):

{
  "mcpServers": {
    "scamverify": {
      "command": "npx",
      "args": ["@scamverifyai/scamverify-mcp"],
      "env": {
        "SCAMVERIFY_API_KEY": "sv_live_..."
      }
    }
  }
}

Docker:

docker build -t scamverify-mcp .
docker run -e SCAMVERIFY_API_KEY=sv_live_... scamverify-mcp

Environment Variables

Variable Required Description
SCAMVERIFY_API_KEY Yes Your API key (prefix: sv_live_ for production, sv_test_ for testing)
SCAMVERIFY_API_URL No Override the API base URL (default: https://scamverify.ai)

Getting an API Key

  1. Sign up at scamverify.ai
  2. Go to Settings > API Keys
  3. Generate a key (prefix: sv_live_ for production, sv_test_ for testing)

OAuth 2.1 Authentication

The remote server supports OAuth 2.1 with PKCE (S256). Discovery endpoints:

  • Authorization Server: https://scamverify.ai/.well-known/oauth-authorization-server
  • Protected Resource: https://scamverify.ai/.well-known/oauth-protected-resource

OAuth scopes: phone:lookup, url:lookup, text:analyze, email:analyze, usage:read

Development

git clone https://github.com/scamverifyai/scamverify-mcp.git
cd scamverify-mcp
npm install
npm run build
SCAMVERIFY_API_KEY=sv_test_... npm start

Pricing

A free tier (50 lookups/month) is automatically provisioned on first OAuth login or API key creation.

Plan Monthly Price Lookups/mo Rate Limit
Free $0 50 10 RPM
Starter $19/mo 1,000 30 RPM
Growth $49/mo 5,000 60 RPM
Pro $149/mo 25,000 120 RPM
Scale $499/mo 100,000 300 RPM
Enterprise Custom Custom Custom

Full API documentation: docs.scamverify.ai

Links

License

MIT

README copy captured when this server was indexed; it may lag the repository. The original is authoritative.

Questions

About scamverifyai/scamverify-mcp

How do I install scamverifyai/scamverify-mcp?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is scamverifyai/scamverify-mcp safe to use with an AI agent?

Its trust score is 34 out of 100 (low). It passes 1 of the 1 static security checks run so far. It has no ConduID identity yet, so agent calls to it are not receipted.

Is scamverifyai/scamverify-mcp still maintained?

The last commit was 6 months ago. That's long enough that you should check whether the maintainer is responding to issues before depending on it.