website-fraud-check

# Website Fraud Check Skill A skill for checking if a website could be fraudulent or pose security risks. ## Description This skill analyzes various factors to determine if a website might be fraudulent, including domain age, URL patterns, known blacklists, and other suspicious indicators. Note that having a valid SSL certificate does not guarantee a website is legitimate, as many fraudulent sites also possess valid certificates. ## Implementation When the user requests to check a website for fraud: 1. Execute the fraud detection script: `nodejs {baseDir}/scripts/website_fraud_check.mjs [WEBSITE_URL]` 2. The script will: - Check domain registration information and age - Verify SSL certificate validity (without considering certificate age) - Analyze URL patterns for suspicious characteristics - Check against known threat intelligence feeds - Examine website content for common phishing/fraud indicators - Compare website content with known legitimate sites to detect impersonation attempts - Assess overall risk level and provide recommendations ## Usage When the user says "check if [website] is fraudulent", "is [website] safe?", "check [website] for scams", or similar requests: - Execute: `exec command="nodejs {baseDir}/scripts/website_fraud_check.mjs [WEBSITE_URL]"` - The script will analyze the website and return a risk assessment - Display formatted results including: - Risk level (Low/Medium/High/Critical) - Specific risk factors detected - Impersonation detection (which legitimate site it might be imitating) - Recommendations for user safety - Confidence level in the assessment ## Dependencies - nodejs: to run the fraud detection script - internet connectivity: to connect to the target website and threat intelligence feeds - whois: to check domain registration information - OpenSSL: for SSL certificate validation ## Error Handling - Handle cases where the website is inaccessible - Handle connection timeouts - Handle invalid URLs - Gracefully degrade when threat intelligence feeds are unavailable - Provide best-effort assessment even with incomplete data