Think Before You Click.
Understand common warning signs in suspicious messages and links, learn why they matter, and choose a safer next step.
Your message and URL text are analysed in this page. LIGHTHOUSE does not open links or send your inputs to a server.
● Local-only prototype · No network lookup🔍 Message Checker
Find common pressure tactics, suspicious offers and requests for sensitive information.
Maximum 5,000 characters. Avoid entering real passwords, OTPs, or private information.
🔗 Link Checker
Inspect URL text for common warning patterns without visiting the destination.
Paste the text only. Do not open a suspicious link just to test it.
💬 Ask LIGHTHOUSE
Explore common digital-safety situations through a conversational, assistant-style experience. Choose a scenario or type a question to receive guidance.
Welcome to LIGHTHOUSE
Choose a scenario above or ask a question below. I will explain common warning signs and suggest practical next steps.
Responses are predefined and selected by simple keyword rules. Your question stays in this webpage; it is not sent to a live AI service.
🛡️ Stay Safer Online
Before you respond
- Never share passwords, PINs or OTPs in response to unexpected requests.
- Verify urgent requests through an official app or website you open independently.
- Check the actual registered domain, not just familiar words elsewhere in the URL.
- Do not pay a fee to claim a surprise prize or job offer without independent verification.
If something feels wrong
- Pause when someone pressures or threatens you to act immediately.
- Do not send money because a caller claims you are under “digital arrest.” Talk to a trusted person.
- If you entered a password, change it using the official service and secure affected accounts.
- If money was lost or you suspect a cybercrime in India, report it through official government channels as soon as possible.
ℹ️ What LIGHTHOUSE can and cannot do
- It matches text and URL patterns against a small set of rules and explains the matches.
- A warning is not proof of fraud; a clean result is not proof of safety.
- It does not visit links, scan files or apps, query threat-intelligence databases, or verify a sender's identity.
- The conversational experience is a simulation: there is no real Alexa+ integration, live generative AI model, or backend service.
- Do not use this prototype as a substitute for professional security support or official reporting channels.