Stijnman Web Scraper
When to Use
- User says scrape page or task matches this capability
- User says extract from website or task matches this capability
- User says get page data or task matches this capability
- User says web extract or task matches this capability
Workflow
- Fetch via sandbox-internet-handler or WebFetch.
- Parse HTML to text, tables, or JSON per user spec.
- Respect robots.txt and rate limits.
- Return data with source URL and timestamp.
Integrations
sandbox-internet-handlerhumanization-stealth-browsinginternet-enabler
Error Handling
| Failure | Response |
|---|---|
| Blocked by site | Try stealth mode or ask user for export. |
Gotchas
- Never scrape authenticated pages without user session.
Example
Input: User request matching triggers above. Output: Structured result per workflow with integrations invoked as needed.