Release Notes

What’s New

• Smarter XPath generation – Optimized XPath strategy for better compatibility with diverse page structures, including improved key-value structure recognition.
• Faster scroll-loading exploration – Accelerated exploration of long pages with scroll-based loading, reducing wait times.
• Better in-process experience – Refined interaction prompts and feedback for a smoother experience during extraction.

Why It Matters

• Handles a wider variety of page layouts more reliably
• Completes long-page extractions faster
• Improves overall stability with key crash and display fixes

Bug Fixes

• Fixed plugin crash issues
• Resolved error message display problems
• Other known bug fixes

What’s New

• Page exploration beyond fixed workflows – Chat4Data can now explore pages and complete extraction tasks beyond predefined workflows.
• Multi-region extraction – Extract multiple types of data from different areas of the same page in one run.
• Dynamic interaction support – Perform actions like clicking, filtering, and typing to reach data revealed after interaction.
• Human-in-the-loop collaboration – Add or refine instructions during execution, and pause for manual steps like login or verification when needed.
• Improved XPath stability – Added dual-path XPath generation, semantic validation, and auto-repair for better resilience on changing pages.

Why It Matters

• Expands what agentic web extraction can handle
• Completes more complex tasks in a single run
• Reaches data that only appears after interaction
• Performs more reliably on changing, real-world pages

What’s New

• Domain-Specific Optimization – Improved recognition accuracy and stability.

Bug Fixes

• Initial Page Scrolling Optimization – Resolved repeated scrolling during initial page analysis.

What’s New

• Automated Site Navigation – Automatically navigates to a relevant page when no specific URL is provided, proceeding after user confirmation.
• Improved Field Detection Accuracy – Includes a reflection mechanism in the core model to enhance result accuracy.

Bug Fixes

• Fixed Known Issues

What’s New

• Added new subscription plans that replace balance top-ups and offer clearer benefits, more flexible usage, and better cost control.
• Added conversation management that allows users to view, search, edit, and bookmark past chats with one-click resume for any history.
• Added file access in chat history where files uploaded in previous sessions can be downloaded directly. Free plan retains files for 14 days and Pro/Max plans retain up to 90 days.
• Added a unified dashboard that centralizes all account details with transparent credit usage per conversation and detailed credit grant records.

What’s New

• Added field recognition using page content descriptions (e.g., email: [email protected]).
• Improved next-level detail page guidance: if a link fails, users will be prompted to select a different field or start a list collection.
• Added collection field confirmation when switching target URLs.
• Added explanation for token and balance conversion rules on the recharge page.

Bug Fixes

• Fixed missing data count on the detail pages.

Bug Fixes

• Fixed Known Issues

What’s New

• Added auto-confirmation of pages: scrape all pages if pagination exists, or the current page if not.
• Separated progress display for starting pages and subpages.
• Added case-specific prompts for unsupported pages.
• Optimized logic for analyzing next-level entries.
• Enhanced chat guidance to proceed extraction.
• Reduced request time with streamlined processing.

Bug Fixes

• Fixed issues with retrieving page structures.
• Fixed issues with JSON parsing.

What’s New

• Added prompts to re-enter a URL or wait if a URL is invalid or a redirect fails.
• Skipped field confirmation step for natural language inputs.
• Added auto-retry mechanism (up to 3 times) for network request failures.
• Optimized extraction logic to address failures in specific scenarios.
• Improved accuracy in detecting and navigating pagination.
• Added Invite Friends & Earn Tokens Program

Bug Fixes

• Adjusted guidance prompts and interface details.

What’s New

• Added support for the data recognition and extraction of details pages, allowing scraping one-level deeper subpages from the start page.
• Added support for local concurrent extraction with up to 3 tabs for next-level data extraction.
• Enabled tab switching during page detection, allowing auto-returns to target pages.
• Optimized pagination recognition and guidance.
• Added natural language support to configure multiple extraction intents for a specific URL.
• Added Simplified Chinese, Traditional Chinese, and Portuguese language support.
• Added FAQ responses within the plugin.

Bug Fixes

• Improved recognition of the main content area.
• Improved detection of missing link fields.
• Fixed known issues.