Shop app benchmark

The same Shopify storefront app, built by an AI from one identical prompt with each of React Native, Flutter, Whisker, Native Shell, NativeScript, and Exact, then measured on a physical iPhone.

Directory	Contents
prompt/	The prompt given to the AI, its GraphQL operations, per-framework values, and the run protocol
design/	The visual design the prompt references
apps/	The app each run produced
measurement/	What is measured and how
tools/	Replay server, XCUITest runner, and measurement driver
results/	Raw traces and summaries

Order of work:

1. Build each app with the AI (prompt/README.md).
2. Record replay data: tools/replay/replay.py crawl, then each app's journey once against record.
3. Measure each app against serve with tools/bench.py (tools/README.md, measurement/README.md).