Overview
Crawling is useful when:- Launching new features
- Needing focused exploration (e.g., admin panels, checkout flows)
- Updating navigation or UI structure
- Improving test coverage for critical user journeys
Starting a Crawl
Ask the assistant in chat. Describe the area you want explored, for example “Crawl the admin panel” or “Explore everything behind login”. You can also type/crawl in the chat input, optionally followed by the area to focus on.
The assistant reports back in the same conversation when the crawl finishes, and continues from there: it can reuse what it found to suggest tests, or crawl again from a different starting point.
Settings → Crawling lists your recent sessions and has a prompt field that hands off to chat.
What the Assistant Decides
You describe the goal; the assistant works out the rest:- Where to start: it opens the page closest to what you asked about, rather than always starting at your project URL.
- How deep to go: a targeted request earns a deeper exploration budget than a broad one, because every extra level multiplies the pages to visit.
- Authentication: to reach pages behind a login, it reuses the browser state from a successful login test. If no login test exists, it offers to create and run one first.
- What to skip: your description filters which links and buttons are worth following.
Allowed Domains
Crawling stays within the domains allowed for your project. Adjust these in Settings → Crawling to widen or narrow what can be explored.Understanding Results
Session Status
- Processing: Currently running
- Complete: Successfully finished
- Failed: Encountered an error
- Cancelled: Manually stopped
Iteration Details
Each crawling iteration shows:- Screenshot: Visual capture of the page
- Source Action: How the crawler reached this page (“n/a” for start page)
- Depth: Links away from the start URL (0 = start page)
- Found Actions: Interactive elements discovered (links, buttons, forms)
- Intent Score: Relevance to your goal (0-40% low, 41-70% moderate, 71-100% high)
Best Practices
- Be specific: name a concrete target. “Focus on the e-commerce checkout flow” works better than “crawl everything”, which cannot narrow anything down.
- Say roughly where it lives: mentioning a path or page helps the assistant start close to the target instead of walking there.
- Review results: check screenshots and intent scores to verify exploration.
Troubleshooting
Crawl Failed- Requires authentication: ask the assistant to run or create a login test, then crawl from that state
- Pages not accessible or SSL issues
- Blocked by your host or network: see the IP Access Control guide