You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Edge returns a 403 JavaScript crawler challenge to search-engine bots — hosted sites can't be crawled, indexed, or previewed #75
Every hostname I serve through Spacefast (apex domain, the app subdomain, and 1,300+ pair subdomains of migrateme.tech) returns HTTP 403 with a JavaScript challenge page to crawler user-agents, while the same URLs return 200 to browsers. The challenge page's title is Checking search engine crawler... and it requires JavaScript, which search crawlers don't run in this form — so the sites are effectively invisible to search engines and link-preview services.
This looks like the crawler-blocking behaviour I raised on 2026-10-14; it is still reproducing on 2026-10-18, so I'm opening it here with fresh evidence.
Evidence (2026-10-18, ~22:28–22:37 CDT)
Same URLs, two user-agents:
URL
Browser UA
Googlebot UA
https://migrateme.tech/
200
403
https://migrateme.tech/sitemap.xml
200
403
https://migrateme.tech/sitemap-pairs.xml
200
403
https://migrateme.tech/robots.txt
200
403
https://universal.migrateme.tech/
200
403
https://lovable.wordpress.migrateme.tech/ (one of 1,332 pair hosts)
200
403
Other crawler user-agents tested against https://migrateme.tech/ — all 403:
Google Search Console (domain property migrateme.tech, verified via DNS on 2026-10-18) accepted the sitemap submission but reports "Couldn't fetch" for https://migrateme.tech/sitemap.xml, with 0 discovered pages — consistent with the 403 challenge.
Because even robots.txt is challenged, crawlers can't reliably fetch crawl rules or sitemaps at all.
Facebook's crawler is also blocked, so Open Graph link previews fail.
Expected
Verified search-engine crawlers (at minimum Googlebot and Bingbot) should receive the real content (200) for public spaces, the way browsers do. If the bot check is an intentional anti-scraping feature, a per-space or per-domain setting to allow search crawlers — or to disable the challenge — would resolve it.
Everything above was verified from multiple networks/clients on 2026-10-18; browser traffic is unaffected and the sites themselves are healthy.
Happy to provide full request/response captures or test again after any change — I can re-run the same UA matrix and the Search Console fetch on request.
Activity