Skip to main content
​
June 2, 2026

​
Improved

  • Instagram scrapers rebuilt — Instagram Posts, Instagram Profile, and Instagram Reels scrapers have all been rebuilt to work with Instagram’s current API. Single-post URLs are now also handled correctly in the Posts scraper.
  • Credit endpoints: status and is_deleted fields — /v1/squids/{hash}/credits rows now include a status field showing the run’s final state. /v1/credits/squids rows now include an is_deleted flag and retain deleted squids in the list for 30 days.
  • Reddit Scraper: improved reliability — browser upgraded to curl_cffi with Chrome impersonation; proxies are now rotated automatically on 403 blocks.

​
Fixed

  • fetch_since: seconds precision in absolute dates — absolute fetch_since values that include seconds (e.g. 2026-05-01 12:00:00) are now parsed and validated correctly.
  • fetch_since: clamped to maximum allowed range — extremely large fetch_since durations no longer cause overflow errors.
  • Instagram Post Comments: session expiry handling — a dead session now correctly raises cookies_expired instead of triggering a 429 retry loop.
  • TikTok Video: WAF bypass updated — the previous DataDome bypass has been replaced with correct SlardarWAF handling to fix blocked requests.
  • YouTube Channel Emails: page parsing fixed — ytInitialData is now extracted from the full page body rather than an XPath text node, fixing failures on recent page layouts.
  • SeLoger: preserve annonce_id — listings whose detail page JSON omits the ID field no longer lose their annonce_id.
  • ParuVendu: handle 404 on list pages — 404 responses on paginated list pages are now handled gracefully.
  • PAP: retry on connection timeouts — proxy-related connection timeouts now trigger an automatic retry with a fresh proxy.
  • Leboncoin Auto Message Sender: daily quota lock corrected — the message rate-limit cooldown was being applied for the wrong duration. It is now correctly locked for 24 hours to match the documented daily limit.
  • Yelp Reviews & Yelp Search: Cloudflare tunnel errors handled — 521–524 proxy tunnel errors from Cloudflare now trigger an automatic retry instead of failing the task.
  • Idealista: browser impersonation updated — the scraper now uses an updated browser fingerprint to handle Idealista’s current bot-detection layer.