ScamNav MCP server
Give any AI agent read-only access to verified scam cases, deterministic risk checks and official reporting channels over the standard Model Context Protocol.
Endpoint
https://scamnav.com/mcp Transport
Streamable HTTP, stateless, with direct JSON responses. No API key or authentication is required; send Accept: application/json, text/event-stream.
Available tools
search_scams
Search the knowledge base by keyword across titles, summaries, tags and scam types.
Parameters
query— Keyword, brand, domain or tacticlang— Optional language code
get_scam_case
Retrieve one case by slug, including sections and FAQs.
Parameters
slug— Case slug, e.g. ftc-impersonationlang— Optional language code
check_scam_risk
Analyze text for known red flags and return a risk level and warning signs.
Parameters
text— Message to evaluate (up to 4000 chars)lang— Optional language code
list_scam_types
List all distinct scam types and the total case count.
get_reporting_resources
List official reporting channels, optionally filtered by region. No recovery services.
Parameters
region— Optional region code, e.g. USlang— Optional language code
Quick configuration
{
"mcpServers": {
"scamnav": {
"type": "http",
"url": "https://scamnav.com/mcp"
}
}
} Safety and scope
All tools are read-only. They never execute transactions, write data or return scam operating instructions. Content is for risk education only.