How it works

From image to useful questions.

The analysis is intentionally narrow. It reviews visible language and layout, then gives you prompts for further investigation.

1

Choose an image

Submit only content you are allowed to use. Avoid including faces, payment information, addresses, account numbers, or other unnecessary personal data.

2

The image is processed

The service sends the image and limited request metadata through configured systems and an AI provider to analyze visible sale-presentation patterns.

3

The response is structured

SaveHawk normalizes the response into one or more items, each with a score, verdict, and three reasons.

4

You verify independently

Review the actual merchant terms, taxes, shipping, returns, price history, and product information before acting.

What the score means

The score summarizes signals found in the submitted presentation. It is not a probability of fraud, a merchant trust rating, or an objective measure of product value.

Use it as a second opinion—not the final word

AI can misread text, miss context, or overemphasize a design choice. When stakes are meaningful, use primary sources and professional advice.