Newly

Assessed October 25, 2026

A conditional fit for founders and non-specialist builders who specifically want a native iOS/Android app and value source-code access, for MVP and prototype work where iterative AI development is acceptable. In our October 25, 2026 build of a test app (StackCheck), Newly generated a real React Native / Expo app, handled iterative changes, built a backend with authentication and per-user isolation, and produced a substantial source export and an Android APK. The condition: its embedded iOS simulator was not consistently reliable, the export still depends on some Newly-hosted services, and Android and notification delivery were not verified on a device — so teams needing predictable testing infrastructure or production-grade device QA should budget for extra work.

Field-by-field assessment

• Prompt-to-native-app generationHands-on

  Generated a working React Native / Expo app from the first prompt; dashboard, detail view, add, edit, delete with confirmation and dashboard recalculation verified on the iOS simulator.

  Also added unrequested features (yearly cost, haptics) — adherence good, not perfect.

• Iterative modificationsHands-on

  Monthly/annual billing added without a redesign; a $120/year tool normalized to $10/month and existing records were preserved. Roughly 3 credits in our account.

• Native features (reminders)Hands-on

  Date workflow, reminder toggle, permission behavior and displayed reminder dates verified; both native builds succeeded.

  We did not confirm the OS fired the notification. Observed a tap-through date picker and helper text that did not always restore.

• Authentication and cloud backendHands-on

  Sign-up, login/logout, session persistence, local-to-cloud import, cloud CRUD, two-account isolation and rejection of unauthenticated requests verified. The agent fixed a real auth-middleware bug itself.

• Embedded iOS simulatorHands-on

  Worked and was usable, but multiple launches failed by exceeding a 20-minute install deadline, including consecutive failures.

  Platform infrastructure friction; the generated app was not shown to be the cause.

• Portability and exit riskHands-on

  Source-code exportability: High. Operational portability: Medium — leaving appears feasible without a rewrite, but needs an infrastructure migration.

  The inspected ZIP still referenced @newly/sdk-react-native, Newly analytics, Newly-hosted API/auth endpoints and build-provenance tooling.

• AndroidHands-on

  Standalone production APK generated (Ready, 133.8 MB), downloaded and statically inspected.

  Not installed on a physical device or run in an emulator.

• Model selection and creditsObserved

  Multiple models with different credit multipliers were selectable; the default Newly Agent was used for our test.

• Observed pricingObserved

  No free plan. Our October 25, 2026 complimentary test account showed a $25 plan with 100 credits. Separately, Newly's main pricing page and FAQ advertise an entry plan at $25/month with 100 credits/month, verified October 28, 2026; the partner also confirmed this standard allowance in writing.

  The account observation is hands-on account evidence; the later allowance confirmation is vendor-documented and partner-confirmed, not a new test. Some secondary official pages, including /create and /mobile-app-builder, still show 50 credits for $25/month. The FAQ describes per-interaction credit pricing, while the Terms describe variable usage based on work performed (reading the project, writing code, building, checking and automatic fixes). Documentation is not fully consistent; verify the applicable terms before buying. Our billing iteration consumed roughly 3 credits; Newly is investigating this observation, which is not a confirmed bug.

• App Store / Google Play publishingNot assessed

  Not assessed

• True cost riskNot assessed

  Not assessed

• Stack overlap riskNot assessed

  Not assessed