> ## Documentation Index
> Fetch the complete documentation index at: https://docs.highailabs.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Testing Plan

> Continuous mobile launch testing plan using local checks, Revyl, Argent, manual QA, and agent-driven audits.

# Testing Plan

This is the repeatable test/build/fix/retest plan for High IQ launch readiness.

## Test Layers

| Layer                     | Purpose                                                    | Tooling                                      |
| ------------------------- | ---------------------------------------------------------- | -------------------------------------------- |
| Static code gates         | Catch type, lint, unit, compiler, and configuration issues | TypeScript, Vitest, ESLint, React Doctor     |
| Local simulator diagnosis | Quickly inspect UI and reproduce flow bugs                 | Argent                                       |
| Cloud device workflows    | Prove launch-critical mobile flows on iOS builds           | Revyl                                        |
| Manual TestFlight QA      | Verify real-world native behavior                          | iPhone, StoreKit, OAuth, camera, permissions |
| Agent audits              | Parallel feature, UI, PR, docs, and pricing review         | Codex/Claude/agent subagents                 |

## Persona Fixtures

| Persona             | Needed for                                      |
| ------------------- | ----------------------------------------------- |
| Guest               | Browse/search/detail, logged-out gates          |
| Clean new user      | Onboarding, empty stash, first upload           |
| Free signed-in user | Stash, favorites, manual upload                 |
| Power user          | Rich stash/history/notebooks/stats/dispensaries |
| Pro user            | Notebooks, stats, Gmail, receipt parsing        |
| Lapsed subscriber   | Paywall, restore, data retention                |

## Continuous Loop

For every launch-critical feature:

1. Inspect current code path and feature flag.
2. Confirm intended tier and launch phase.
3. Run or write the smallest useful automated proof.
4. Capture screenshots or manual device notes for UI-heavy screens.
5. Fix code, UX, data, or test blockers.
6. Rerun the same proof.
7. Update the feature matrix and iteration log.

## Immediate Test Backlog

| Priority | Test                                                                                         | Layer                                  |
| -------- | -------------------------------------------------------------------------------------------- | -------------------------------------- |
| P0       | `useFeatureAccess` decision matrix                                                           | Unit                                   |
| P0       | Guest/free/Pro direct-route gating                                                           | Revyl/Argent                           |
| P0       | Current candidate Revyl smoke rerun                                                          | Revyl                                  |
| P0       | Onboarding first-run journey                                                                 | Revyl/manual                           |
| P0       | Strain detail UI screenshot sweep                                                            | Manual/Argent                          |
| P1       | Manual text upload happy path                                                                | Revyl                                  |
| P1       | Populated stash + history + edit mode                                                        | Revyl/Argent                           |
| P1       | Notebook order -> progress -> readable detail -> cancel/recovery -> immutable update/history | Studio + unit/Convex/UI + Revyl/Argent |
| P1       | RevenueCat purchase/restore/lapsed                                                           | Manual/TestFlight                      |
| P1       | Gmail recent sync controlled account                                                         | Manual/Revyl where possible            |
| P2       | Image receipt fixture                                                                        | Revyl/manual                           |
| P2       | Stats rich-user route                                                                        | Revyl                                  |
| P2       | Dispensary CRUD                                                                              | Revyl                                  |
| P2       | Music / now-playing route                                                                    | Revyl/Argent                           |

## Agent Usage

Use background agents when a feature can be reviewed independently:

* Feature surface agent: routes, flags, screens, navigation.
* Test agent: Vitest, Revyl, Argent, coverage gaps.
* UI agent: screenshots, layout density, spacing, empty states.
* Docs/pricing agent: public promise consistency.
* PR/risk agent: recent merged work, open PRs, CI state.

The parent owner consolidates results into this launch readiness system and decides the next certification pass.

## Notebook test lane

Use [Notebook Studio](/planning/notebooks/studio) for provider-free product and edge-state review, then follow the automated, database, staging, and native layers in the [Notebook launch runbook](/planning/notebooks/launch-runbook). Studio approval alone is not release evidence.