Versioned, reproducible benchmarks of AI app builders - methodology-first and version-stamped.
93 posts
We are proposing a fifth BuilderProof axis to score whether an AI app builder ships code that can leave the platform. Five sub-axes, 0 to 100, provisional cohort scores included.
Effective with the H2 2026 ranking, BuilderProof retires binary 'first-build success' as a scored axis, on the basis of documented vendor changes through H1 2026 rather than any first-party run. It becomes a precondition (must pass to be ranked) and the rank weight moves to time-to-first-functional-build.
Two Lighthouse runs on the same deployed AI-builder output rarely return the same score. This note documents the variance phenomenon, names its seven sources from Google's own documentation, and publishes the median-of-five protocol a hands-on deploy-quality harness would have to meet. Corrected August 21, 2026: the June 2026 result table it accompanied has been withdrawn, and the account of how that table was produced is withdrawn with it.
Proposing first-build stability as the fifth BuilderProof axis: the fraction of OQ-7 prompts that complete without manual intervention. Failure-mode taxonomy, measurement protocol, scoring rubric and open questions, dated June 20, 2026.
BuilderProof methodology v1.1: the published rubric, brief OQ-7, environment standards and weights used to score AI app builders on output quality, speed, deploy quality and agency suitability. The four June 2026 result sets were withdrawn on August 21, 2026 as placeholder data, so the lab currently publishes method, not scores.
Agencies build for clients, which changes what matters: can you remove the builder's branding, drive it programmatically, integrate via a stable API and export the code you ship? This page defines the BuilderProof agency-suitability axis across those four capabilities. The June 2026 scored table was placeholder data and was withdrawn on August 21, 2026, so the page documents method only and ranks no builder.
The BuilderProof deploy-quality axis audits the production build of a generated app on three independent dimensions: Lighthouse performance, axe-core accessibility plus a manual keyboard-and-landmark pass, and a structured SEO checklist. The June 2026 audit table was placeholder data and was withdrawn on August 21, 2026, so this page documents method only.
The BuilderProof speed protocol uses two stopwatches rather than one number: speed-to-first-paint (prompt to first rendered preview frame) and time-to-working-app (prompt to all acceptance checks passing with zero manual edits), across five cold runs on a fixed network profile. The June 2026 timing table was placeholder data and was withdrawn on August 21, 2026.
The BuilderProof output-quality axis: brief OQ-7 and a rubric grading visual fidelity, code structure and functional correctness, weighted so correctness and structure outrank visuals. The June 2026 scored table was placeholder data and was withdrawn on August 21, 2026, so this page documents method only and ranks no builder.
