// Real-time URL heuristic analysis — no API required
A real-time, client-side URL threat analyzer that detects phishing, typosquatting, and suspicious links using 11 heuristic checks — with zero APIs, zero server calls, and zero data leaving your browser.
PhishGuard analyzes URLs using a weighted heuristic scoring engine. Each URL is passed through 11 security checks — from HTTPS verification to brand impersonation detection. The total score (0–100) determines the verdict: Likely Safe, Suspicious, or Phishing Risk.
No external calls are ever made. Everything runs locally in the browser, making it instant and completely private.
Single-file architecture. No build step, no dependencies, no package.json. Drop it anywhere and open it.
| # | Check | Severity | Weight |
|---|---|---|---|
| 1 | HTTPS Protocol | HIGH | 20 pts |
| 2 | IP Address in URL | HIGH | 25 pts |
| 3 | URL Length (>75 chars) | MED | 10 pts |
| 4 | @ Symbol in URL | HIGH | 15 pts |
| 5 | Subdomain Depth (>2) | MED | 15 pts |
| 6 | Suspicious Keywords | MED | 15 pts |
| 7 | Hyphens in Domain (≥2) | LOW | 10 pts |
| 8 | Suspicious TLD | MED | 15 pts |
| 9 | Brand Impersonation | HIGH | 25 pts |
| 10 | Redirect Indicators | MED | 10 pts |
| 11 | Known Legit Domain Bonus | LOW | −30 pts |
| Score | Verdict | Meaning |
|---|---|---|
| 0 – 25 | LIKELY SAFE | Low risk — URL appears legitimate |
| 26 – 55 | SUSPICIOUS | Some red flags — proceed with caution |
| 56 – 100 | PHISHING RISK | High confidence threat — do not visit |
PhishGuard is a single index.html file. Deploy it anywhere static hosting is supported:
PhishGuard uses heuristic analysis only — it cannot query DNS, check SSL certificates, WHOIS data, or live threat feeds. A "Likely Safe" verdict is not a guarantee. Always verify suspicious links through multiple sources before visiting.
Built by Anam — BCA student, aspiring Cloud Security Engineer, and self-taught ethical hacker.
// GitHub: github.com/anamv007