Scrapeless Weekly Changelog: November 2–8, 2026 Updates

TL;DR:

• Grok Scraper no longer accepts mode. Remove it from request payloads. Grok now uses MODEL_MODE_FAST by default.
• Perplexity Scraper adds queries. Capture search queries alongside the answer.
• Amazon Rufus adds sources. Retain source information alongside Rufus results.
• ChatGPT HTML output can render maps. This applies when the returned answer includes map content.
• Scrapeless has a redesigned website navigation bar. Use it to explore products, solutions, and resources.

---

The November 2–8 release changes how applications submit Grok requests and expands what they can capture from Perplexity, Amazon Rufus, and ChatGPT. If your integration sends mode to Grok, update that payload first.

Grok Scraper: Remove the Mode Parameter

Grok Scraper no longer supports the mode request parameter and now uses MODEL_MODE_FAST by default. Remove mode from requests, including requests that already send MODEL_MODE_FAST as the field value. If a settings screen exposes a Grok mode selector, update it to reflect the current request options.

Check shared request builders as well as individual calls. The JSON data interchange standard defines the payload format; the Grok Scraper reference defines the inputs the actor accepts.

Perplexity Returns Search Queries

Perplexity Scraper now supports queries data for search queries associated with the response. Store the original prompt and the returned queries separately. That gives reviewers a way to compare what they asked with the searches associated with the resulting answer.

Start Scraping with Scrapeless

Power up your web scraping and automation workflow with Scrapeless!
Sign up today and get $5 in free credit — no credit card required.

Claim your free credit now in the Scrapeless Dashboard.

Amazon Rufus Adds Source Information

The Rufus action in Amazon Scraper can now return source information in result.sources. Availability depends on the product and prompt. Teams reviewing shopping answers can keep this information with the answer so the source context remains available during review. If you map results into a separate reporting model, the Schema.org citation vocabulary is a useful reference for representing source relationships. It does not define the Rufus response format.

ChatGPT HTML Output Can Render Maps

The ChatGPT Scraper now includes map rendering in its html output when map content is available. Use the documented should_fetch_html option to request HTML. This gives teams inspecting an answer’s layout a way to review the map in the returned markup. The HTML Living Standard describes how browsers interpret that markup.

A Redesigned Scrapeless Navigation Bar

The Scrapeless website has a redesigned navigation bar. It gives visitors a starting point for browsing products, finding workflows under Solutions, and locating resources.

Conclusion

For existing integrations, the action this week is to remove mode from Grok requests. The new Perplexity, Rufus, and ChatGPT outputs help teams keep search queries, source information, and rendered maps with the answers they review. Explore the Web Unlocker and Scrapeless pricing as you plan collection. Last week’s Scrapeless Weekly Changelog covers Google AI Mode places and Perplexity shopping citations.

Ready to Build With Scrapeless?

Join developers building AI-answer and web-data workflows: Discord · Telegram.

Sign up at app.scrapeless.com to get started.

FAQ

Q: Should I send mode: MODEL_MODE_FAST in Grok requests?

No. Remove the mode field entirely. Grok uses MODEL_MODE_FAST by default.

Q: What is the difference between Perplexity queries and Rufus sources?

Perplexity queries provides search query data for the response. Amazon Rufus sources provides source information for the Rufus result.

Q: How do I request ChatGPT HTML output with map rendering?

Enable the documented should_fetch_html option. The returned HTML can render a map when the answer includes map content.