2.1 KiB
Claude QA Coverage — Period
Status: Current
Owner: _null
Last reviewed: 2026-08-18
Governs: what each QA pass actually reached
Review trigger: Any QA round run
Pass by pass, what was reached and what was not. The point of this file is the Blocked and Not run rows: a pass left out of a report reads exactly like a pass that succeeded, and that is how untested code ships believing it was tested.
No round has been run
There is nothing to report yet, and this section says so rather than leaving the
file looking like a round that found nothing. The first round can only happen
once there is an app to run — the passes in
ClaudeQAPlan.md all require a build, and Batch 01 is what
produces one.
The table below is the shape each round fills in. It is deliberately left with no rows rather than pre-filled with "Not run", because a round that never happened and a pass that was skipped are different facts.
Round N — YYYY-MM-DD at sha
| Pass | Result | Notes |
|---|---|---|
| A — First run | ||
| B — Core loop | ||
| C — Failure paths | ||
| D — Persistence and migration | ||
| E — Forecast under hard histories | ||
| F — Notification privacy on a lock screen | ||
| G — Accessibility | ||
| H — Data ownership and leakage |
Results are Pass, Partial, Blocked or Not run — and the last three carry
what stopped them and the issue number, never a blank.
Standing gaps
Things no round has ever covered, carried forward until they are. This list existing is not a failure; it not existing while the gaps do is.
- Everything. No build exists yet.
- Physical-device coverage is undecided. Passes F and G need a real device with a lock screen and TalkBack; which device that is has not been chosen, and an emulator is not a substitute for either.
- Long-horizon accuracy — whether predictions measurably improve at 3, 6 and 12 confirmed cycles — cannot be reached by a QA round at all. It needs either a simulated history harness or real elapsed time, and until one exists the product's headline claim is tested only at the unit level.