The Google Chat fetcher
Google Chat uses a user-triggered fetcher that reverse-resolves to *.gae.googleusercontent.com. It operates on behalf of the user who shared the link. The key detail: it ignores robots.txt because Google classifies it as a user-initiated action, not autonomous crawling.
Not Googlebot
This fetcher is separate from Googlebot (Search indexing). It shares infrastructure with other Google user-triggered fetchers (like Google Messages), but doesn’t affect search rankings. Blocking Googlebot in robots.txt has no effect on Google Chat previews.
No JavaScript execution
The fetcher does not run JavaScript. OG tags must be in the initial HTML. Client-side rendering of meta tags won’t work – you need SSR or pre-rendering.
Response requirements
Expects 200 OK with text/html. Redirects are followed. Pages behind auth, VPN, or IP restrictions produce no preview – the message just shows the raw URL.
Checking if the fetcher can reach your page
Since robots.txt doesn’t apply here, the main concern is IP or user agent blocking:
curl -A "Google-Chat-Preview" https://example.com/your-pageIf your meta tags are in the response, Google Chat can generate a preview.