Improvements

• Fully restored the operation of the SE::Google scraper on Linux for desktop/mobile modes and HTTP/Browser engines, as well as on Windows in all modes

• Added a new scraper FreeAI::DuckAI for retrieving AI model responses via Duck.ai by DuckDuckGo without signing up. Supported models include GPT-5.4 Nano, GPT-5.4 Mini, Claude Haiku 4.5, Mistral Small 4, GPT OSS 120B, and Gemma 4 31B

• Added new Wikipedia scrapers: SE::Wikipedia and SE::Wikipedia::Article

• Added a new scraper for App Store app data: AppStore::Apps

• In Net::Whois, added support for .рф, .ga, and .gt TLDs, as well as .com.gt, .ind.gt, .net.gt, .org.gt, .edu.gt, .gob.gt, and .mil.gt. For .gt, which has no WHOIS/RDAP server, lookup is performed through its web interface

• In Net::Whois, updated support for .li and .ch TLDs by adding RDAP servers

• In FreeAI::Perplexity, added support for authentication via cookies

• SE::Yahoo has been rewritten using Playwright

• NextGen UI:

  • Added a command palette and keyboard shortcuts for quick access to actions
    
  • Added the ability to collapse all JSON blocks in logs, statistics per query, and filtering by statistics values
    
  • Preset overrides can now be reordered and temporarily disabled without deleting them
    
  • Unsaved form changes are now saved as drafts with an option to discard changes
    
  • Fixed handling of saved preset overrides when switching scrapers
    
  • Fixed processing of query file paths without an explicitly specified folder — they are now properly normalized to queries/
    
  • Task presets and thread configs are now reloaded instead of served from cache, so they always display up-to-date data
    
  • During short-term network glitches, requests are automatically retried instead of displaying an error screen
    
  • If the A-Parser core hangs, the UI notifies immediately without waiting five minutes

Fixes following changes in search results

• In SE::Google, fixed pagination in desktop mode following changes to Google’s SERP
• In SE::Google, restored scraping of the PAA (People Also Ask) block
• In SE::Google, improved empty SERP detection — added a check for the absence of links when the SERP is fetched successfully
• In SE::DuckDuckGo, fixed scraping when queries include search operators and added the “Get first link with browser pages count” option. If the first-page HTTP request returns 202, the next attempt is made via the Browser engine
• In Rank::Ahrefs::KeywordGenerator, fixed incorrect retrieval of $difficulty and $volume
• In SE::Aol, restored the scraper’s functionality
• In SE::Rambler, resolved a recurring IP-ban issue. The scraper now operates via Browser engine
• In SE::You, restored the scraper’s functionality

Fixes

• In SE::Google::Images, fixed an issue where the scraper could stop trying to process subsequent queries
• Fixed display of overrides in saved tasks
• Fixed API request body generation error in the “Get API request” feature