Free website
table extractor.
From a table on your page
to a CSV you can use.
Keep the headers, preserve the values, and take the source link with you. Let WebAct handle the repetitive part.
Chrome extension + account required.
25 free automation tasks/month. Browser-provider limits apply.
| Item | Price | In stock |
|---|---|---|
| Desk, compact | $89.00 | 12 |
| Chair | $45.00 | 8 |
| Lamp | $29.00 | 15 |
Item,Price,In stock,Source URL
"Desk, compact",$89.00,12,https://example.com/catalog
Chair,$45.00,8,https://example.com/catalog
Lamp,$29.00,15,https://example.com/catalog
Three small steps.
- 01
Get WebAct ready.
Install the Chrome extension and verify your WebAct account. Choose Browser model → ChatGPT and sign in when prompted.
- 02
Open the table you need.
Activate WebAct from the toolbar on the source page. Select the table as context, then switch to Task mode.
- 03
Paste. Run. Check.
Use the task below, review the proposed actions, then compare the CSV with your table before importing it.
Give the copying to WebAct.
Copy this task and paste it into WebAct on the page with your table.
Website table → CSVTASKExtract the visible, loaded rows of the table I select on this page into a CSV named webact-table.csv. If no table is selected and more than one is available, ask me which one. Keep the original headers and cell values, including IDs, dates, units and currencies. Add a Source URL column with this page’s URL. Leave missing cells empty; never invent values. Use one header row and consistent columns. Escape commas, quotes and line breaks correctly. Neutralize cells that could run as spreadsheet formulas and tell me which values you changed. Start with up to 25 data rows. Keep the file under 48,000 UTF-8 bytes. Do not follow links, load other pages or include hidden rows. Report the exported row count and any omitted or unreadable content. Check the CSV against the source before offering the download. Stop and explain if the table cannot be read.
Paste in Task mode with Browser model selected.
Before you extract.
Start small. Keep the source.
Check what comes back.
Is this website table extractor free?
Yes, this task can use WebAct’s Free plan with a ChatGPT browser connection. You need the Chrome extension, a verified WebAct account and a signed-in ChatGPT account in the same Chrome profile. No WebAct subscription or paid API key is needed. Provider limits still apply.
How many tables can I extract for free?
The Free plan includes 25 user-started automation tasks per calendar month (UTC), shared across all WebAct tools. A task counts when its first action starts; the agent’s steps and necessary retries within that task do not count separately. After 25 tasks, upgrade or wait for the next monthly reset. Ordinary browser Chat stays free, subject to provider limits.
Can I extract tables from any website?
Use a readable HTML table loaded on your current tab. This recipe does not capture tables in images, scanned PDFs or canvas-based grids. Hidden rows, other result pages and content that has not loaded are excluded. Start with up to 25 rows and check the reported coverage; site restrictions may prevent extraction.
Can I open the CSV in Excel or Google Sheets?
Yes. Download the CSV, then use your spreadsheet’s import option. Import identifiers such as ZIP codes as text to preserve leading zeros. Check the headers, row count and a few values against the original table. Each generated file must stay under 48,000 UTF-8 bytes; use a smaller table if needed.
Do I paste a website URL here?
No. This page shows an illustrative example and a task you can copy. The extraction runs through WebAct on the source page you activate. Page context shared with the task is processed by your selected browser AI provider. Review the context before running it.
Learn how shared page context is handled in our privacy policy.
Your table. Ready for what’s next.
Bring WebAct to the page you’re already working on.