Audit and Edit Crawled Website Data for Conversation AI
In Bot Training, click the document icon next to a crawled URL to open the scraped content, edit it inline, and save. The bot retrains on the cleaned data. Refreshing the URL reverts to the original crawl.
What's new
You can now review and edit the content scraped from each website during Bot Training. That means catching wrong pricing, outdated hours, or off-brand language before the bot uses it as training data.
How it works
- Open the Bot Training tab.
- In the Uploaded Links table, find the document icon next to each URL.
- Click the icon to open a modal with the scraped content.
- Edit the text directly. Fix typos, remove fluff, correct facts.
- Save. The bot retrains on the updated content.
Edited entries are tagged Edited so it's clear which ones have been touched post-crawl.
Note: Refreshing an edited website pulls fresh content from the live site and overwrites your edits.
Frequently Asked Questions
Where do I edit crawled content?
In the Bot Training tab. Each crawled URL has a document icon next to it — clicking it opens the scraped content for inline editing.
Will my edits stick if I refresh the URL?
No. Refreshing re-crawls the live site and replaces the edited content. Only refresh when you want to pull the latest version from the source.
How can I tell which entries have been edited?
Edited entries display an "Edited" tag in the Uploaded Links table, distinguishing them from untouched crawl results.
Does the bot retrain automatically after I save?
Yes. Saving an edit retrains the bot on the updated content so the next conversation uses the corrected data.
Was this article helpful?
Thanks for the feedback.