> ## Knowledge Base Index
> Fetch the complete knowledge base index at: https://product-upload.crisp.help/sitemap.xml
> Use this file to discover available pages before exploring further.
> Pure-Markdown content can be obtained by appending a '.md' suffix to the content URLs listed in the sitemap (without the trailing slash).

# I'm seeing images or content from "similar products" or recommendation sliders in my imports

Some supplier sites include recommendation sliders, "customers also bought" sections, or related-item galleries. The scraper can occasionally pull these into the main product by mistake.

If you notice this happening:

* Set **AI quality** to ✨ **Quality** under **Advanced** on the **Content** tab. The more powerful model is better at filtering out non-primary content.
* Turn on **Skip search-engine images** on the **Data Sourcing** tab, so Product Upload relies only on the product page itself.
* Add a rule to **Extra instructions** (Content tab, under Advanced) such as `Only include images and content that belong to the main product on this page. Ignore related, similar, or recommended products.`

| If it keeps happening on a specific site, send the URL to support and we'll tune our extraction for it.