What is Web Scraper?
One-liner: pull what you need from any web page in one click and import it into AI Table. A smart web-data collection tool. The browser extension and the AI Table side-panel plugin work together to give you one-click extraction of structured data from any web page.What can you do with it?
Pages and data types you can scrape
- Self-media platforms: Xiaohongshu note lists, short-video lists — capture title, author, like count, etc.
- E-commerce: capture product cover, price, rating, sales, etc.
- News and articles: capture title, body, summary, publish time, etc.
- Comment sections: capture content, author, publish time, etc.
- Listing platforms: capture bid links, body, publish time, etc.
- Internal back-office pages: turn finance, attendance, project-progress back-office tables into AI Table records
Before you start
See the install guideExamples
Scraping a list page
Xiaohongshu note list: paste a Xiaohongshu keyword search URL or a user homepage URL into the input — the note list on that page will be parsed and imported into AI Table. Taobao product list: paste a keyword product list URL on Taobao / Tmall or any e-commerce platform — the cover, price, sales, ship-from location, and other fields will be detected and imported into AI Table. Taobao product comments:- Paste a Taobao product detail URL. If the comment list isn’t auto-detected, click Reselect list.
- Open the product detail page and click View all reviews.
- In the scraper panel at the top-right, click Manual select and pick the comment area — the comment list is detected.
- Scroll the page to capture more rows.
- Click Confirm to write the data back into AI Table.
Scraping a detail page
Take a column of URLs you already have in AI Table (or one you scraped with this plugin) as input, configure which fields to capture from the detail page, and Web Scraper opens each URL in turn and writes the captured data row by row. Bulk-scrape Douyin video detail pages: Click Scrape detail pages. In the Pick the link field to scrape input, choose the link field that holds the Douyin video URLs. Note: every URL in that column must be from the same site and the same page type, or you’ll hit errors. Bulk-scrape Xiaohongshu note detail pages: Click Scrape detail pages. In the Pick the link field to scrape input, choose the link field that holds the Xiaohongshu note URLs. Same caveat: every URL must be from the same site and the same page type.FAQ
After installing the plugin in Chrome / Edge, why am I still prompted to install a new version when I start scraping?
- Make sure you opened AI Table in a browser — the plugin doesn’t work in AI Table opened from the DingTalk client.
- After installing the plugin, refresh the AI Table page before re-entering it.
Will using this plugin trigger anti-crawling defenses or get my account banned?
Web Scraper is essentially a “copy-paste automation tool” running in your local browser — not a server-side crawler. It doesn’t scrape in bulk and doesn’t try to bypass anti-crawler mechanisms. Why doesn’t it trigger anti-crawling?- It uses your own login: the plugin runs in your local browser with your normally signed-in account. No login bypass, no permission cracking — to the site, this looks like a normal user visit.
- No automated bulk scraping: Web Scraper is single-shot, manually triggered — you click to run it. It doesn’t support high-frequency batch requests.
- Paced like a real user: the plugin matches normal browsing pace, doesn’t refresh aggressively, doesn’t fire concurrent requests at multiple pages — fundamentally different from typical crawler behavior (high frequency, batch, automated).
- Respect the site’s access rules and terms of service
- Don’t bypass logins, permissions, or anti-crawler mechanisms
- Stay focused on personal-productivity use cases