Check it before you trust it

ScamShield AI helps individuals, families and teams examine suspicious messages before they click, pay, reply, install software or share credentials.

What it checks

• Emails, texts and social-media DMs
• Supplier invoices and changed payment details
• Job offers, rental listings and marketplace messages
• Parcel, toll, bank and account-alert messages
• Suspicious URLs, shortened links and lookalike domains
• Requests for gift cards, cryptocurrency, passwords, one-time codes or remote access

What you receive

• A cautious concern tier: High concern, Needs independent verification, Limited concern based on provided evidence, or Insufficient evidence
• An evidence brief linking each warning to an observable detail
• A list of facts, unknowns and plausible benign explanations
• Ordered verification steps using contact details found independently
• A privacy-safe share card with no raw message or personal identifiers
• Immediate containment guidance if money, credentials or device access were already shared

How to use

1. Paste the suspicious content or upload a screenshot. Redact passwords, one-time codes, full payment numbers and identity documents.
2. Explain where it arrived and whether you already clicked, paid or replied.
3. Review the observable signals and verify the request through an established, independent channel.

Privacy and safety

The packaged local engine makes no network calls and does not include the raw message in its structured output. The selected Capafy hosting model may process uploaded content under its own terms.

ScamShield AI does not click links, scan malware, verify a sender's identity, contact anyone, file reports, reverse payments or guarantee that content is safe or fraudulent. It is a decision-support tool, not legal, financial or cybersecurity advice.