example.app-submission-readiness-mobile-proof

@lssm-tech/example.app-submission-readiness-mobile-proof

Deterministic ReviewReady proof that audits the real Expo app at packages/apps/mobile-monolith and emits a stable readiness report.

This proves ReviewReady works on a real brownfield mobile app, not a toy fixture: the proof reads and parses the actual mobile-monolith source files from disk at run time, maps them into the runtime's audit input, and runs @lssm-tech/lib.app-submission-readiness-runtime. If the real app config changes, the parsers pick it up and the inline expected values in the test will catch the drift.

Real files it reads

parse-mobile-monolith.ts reads these files via node:fs (no Expo execution, no mutation):

The Expo config is parsed as source text without executing the application or requiring generated native projects.

Honest gaps surface as findings

Declared-in-repo items are pass; genuinely absent submission materials become real findings:

• legal.account-deletion — no account-deletion URL in the Expo config.
• assets.screenshots.store — assets/ holds only icon/adaptive-icon/splash.
• review.google.dataSafety — no Google Play Data Safety answers in repo.

Deterministic outcome

Fixed checkedAt = 2026-06-07T00:00:00.000Z, fixed ids, no Date.now() / Math.random(). Current real audit outcome:

• decision: blocked
• score: 60
• findings: 3
• stores: apple-app-store, google-play (store: "both")

Usage

import {
	buildMobileMonolithReadinessReport,
	describeMobileMonolithProof,
} from "@lssm-tech/example.app-submission-readiness-mobile-proof";

console.log(describeMobileMonolithProof());
const { report } = buildMobileMonolithReadinessReport();
console.log(report.decision, report.score);

Verify

cd examples/app-submission-readiness-mobile-proof
tsgo --noEmit
bun test

The proof asserts: real parsed values, real findings for absent materials, deterministic deep-equality across two runs, no secret-like strings in the serialized report, and inline expected decision/score/finding ids.