• The content is AI-generated. Published websites, their source, and their screenshots were produced by coding tools from a frozen brief. They are demonstrations of what a workflow produced, not products, not maintained software, and not advice.
• As-is, without warranty. The archive is made available for inspection and research with no warranty of any kind, including fitness for a particular purpose. Published sites may contain defects, dead interactions, or incorrect content; that is part of what is being recorded.
• No license is granted. No MIT, Creative Commons, or other license is applied to the published source or media, and no additional rights are claimed over it. Being able to read something here is not permission to reuse it.
• Assets and brands. Challenge assets are self-created or clearly licensed, and every asset carries a recorded provenance and license entry. Brands used in Challenges are fictional unless a documented basis exists for a real one.
• Do not attack the previews. Published sites run on separate origins under a restrictive security policy. Probing for private ingestion archives, admin endpoints, or a way out of a preview origin is not welcome.

• No account, no sign-up. Browsing the archive requires no account and no personal data. The only sign-in that exists belongs to the operator.
• No advertising, no visitor tracking. There are no advertising networks, no third-party trackers, and no profiling of visitors. No visitor cookie is set for reading the archive. If usage measurement is ever added, it will be a cookieless aggregate and this notice will say so before it is switched on.
• Preview domain and cookies. Published sites are served from preview subdomains of the same registrable domain as this platform. Each Deployment still has its own origin, and previews cannot read or frame platform pages, but this arrangement is not a cookie boundary: a generated preview can set a Domain=uibenchmark.dev domain cookie and cookie flooding remains possible. Platform sessions therefore use __Host- cookies, which browsers do not send to previews; that mitigation is not the same as full isolation, so do not put sensitive data in a preview.
• Infrastructure logs. The hosting and storage providers that serve this site keep their own operational request logs. They are used to run and protect the service, not to build a profile of you.

Published archives are retained indefinitely. There is no automated deletion: the point of the archive is that a Revision published today can still be inspected, byte-identical, much later. Withdrawn Revisions keep their stored bytes so a withdrawal can be reversed if a claim turns out to be unfounded, while their public pages show only a tombstone and offer no live action. True deletion happens only as a manual, audited act when there is a legal obligation or an exposed secret.

Rights holders and anyone affected by published content can ask for its removal. A plausible claim leads to withdrawal first, review second, and restoration only if the claim turns out to be unfounded. See Report content for what to send and how quickly you can expect a reply.