Questions about RenderCache

Do all AI crawlers fail to run JavaScript?

Crawler capabilities differ and can change. RenderCache addresses requests that need useful HTML without executing your application. Its editable user-agent patterns are routing hints, not a universal claim about every crawler or a way to verify identity. Test the responses that matter for your own public pages.

Will this guarantee search ranking or AI citations?

No. It can make rendered content available in a response and record what the server wrote. A receipt does not prove a crawler read the body, indexed it or used it in an answer. Reachability, robots rules, canonical metadata, content quality and each service’s policies still matter.

Does it replace server-side rendering?

It provides a pre-rendered cache path for crawler requests while humans continue to receive the app. It does not convert the application to server-side rendering, and it does not make the human browser’s loading path faster. If you already produce useful HTML for all clients, check whether you need a separate crawler cache at all.

Can I cache logged-in or personalized pages?

The shared snapshot path is for public, unauthenticated content. Visitor cookies and authorization never enter render jobs. Private, no-store, cookie-dependent and authentication-required outcomes are rejected or withdraw an older copy. An origin must still classify its content correctly; do not point shared warming at private pages.

What happens before a page has a snapshot?

The cold request proxies the origin and queues background work. It does not wait for Chrome. The response may still be an empty shell, so warm sitemap coverage is a deployment check. A stale saved response remains available during ordinary refresh failures. Positive privacy outcomes have stricter withdrawal behavior.

Is the whole browser inside the Go binary?

No. There is one Go service executable, plus an externally installed Chrome/Chromium. The Dockerfile installs Chromium. The initial supported release is Linux amd64, with Linux-specific secure file operations and locking.

Is the website demo rendering my site?

No. It bundles one real public fixture capture and illustrates cache and receipt behavior locally. Live configured-site comparisons are available in the self-hosted protected admin app. This static site does not accept arbitrary rendering URLs.

Can I use Lightpanda or a BareProxy plugin today?

Those integrations are deferred. The current engine uses Chrome, and the alpha can run as its own reverse proxy or behind nginx/Cloudflare routing. There is no hosted billing service, framework plugin collection or container-registry image in this release.