# Robots, private URLs and JavaScript pages

A page can open in your browser and still be unsuitable for hosted extraction. Preserve the failed operation before changing the URL or starting more work.

## Robots refusal

For `robots_disallowed`, respect the source site's extraction restriction. Confirm that you used the intended public source, then choose a permitted source if appropriate. Do not change identities, endpoints or tools to evade the restriction. A paid Zilch plan does not override the source owner's controls.

## Private or credential-bearing destination

For `not_public`, inspect the address without sharing its secrets. Local hosts, private network destinations and URLs carrying credentials are not valid public extraction targets. Redirects can also lead away from the original public destination.

Use a genuinely public page that is safe to submit. Do not paste cookies, signed links, private IPs or session tokens into another request to force access. A page accessible through your VPN or signed-in browser may remain inaccessible to the hosted service.

## JavaScript-dependent content

For `javascript_unavailable`, keep JavaScript off; the hosted service does not provide script execution. A completed document that contains only a site shell can also indicate that the desired text was not present in the initial HTML.

Inspect the normal source page for a public static version if one exists. Full-page scope is still static HTML, not a rendering engine. Do not repeatedly submit the same script-only page expecting a different browser session.

## Verify before continuing

Check the corrected source and the reason for the original refusal before deliberately creating new work. If the condition has not changed, stop. Retain the operation ID and sanitized error for [support](/docs/shared/support), and review [static extraction boundaries](/docs/zilch/static-html-and-public-url-boundaries) before another paid request.