BuilderProofv0.1

Tag

#ai app builders

58 posts tagged.

Does User-Typed Text Render as Text? A Proposed Output-Encoding Posture Axis for AI App Builder Output (September 2026)

A proposed benchmark axis for what a generated app does with a string a person typed when it puts that string back on a page. MDN documents that injected script elements do not execute, so the folk test returns a clean pass on a build that is not safe. Both failure directions are scored: markup that runs, and a surname that renders as punctuation it never contained.

27 min read139
benchmarks

Does It Run in the Browser They Actually Have? A Proposed Browser-Support Axis for AI App Builder Output (September 2026)

A proposed benchmark axis for which browser engines a generated app can actually run in. Baseline defines widely available as 30 months of interoperability and names what it cannot see. Next.js floors at Firefox 111, Vite resolves its default to Firefox 114, and the preview pane reports a pass either way. Seven weighted signals, four postures, a ten-step protocol, and no scores.

19 min read195
benchmarks

What Happens When the Connection Drops? A Proposed Axis for Client Network Failure Handling (September 2026)

A proposed benchmark axis for what a generated app does when the user’s own connection fails mid-action. The HTML Standard defines navigator.onLine so that true means "might be online", and the Fetch Standard defines a failed request as a response with status zero, no headers and no body. Seven weighted signals, four postures, a ten-step protocol, and no scores.

18 min read161
#ai app builders · BuilderProof