What the evidence actually supports

Evidence levels

A — Official statement: an observed developer listing, announcement or identified developer reply. This verifies what was published; it is not our runtime test.

B — Attributed third-party account: a named source with a traceable page. A single source remains provisional. Two pages quoting the same developer text are not two independent gameplay observations.

C — Individual report: a player’s stated experience. Useful for investigation, but not a population-wide result or a guaranteed fix.

Editorial analysis: our interpretation, checklist or proposed test. It must not silently become a game rule.

Claim register

What we tried to cross-check

We read the official store entries, developer announcement feed, a developer reply attached to a Steam review, individual player reviews, a specialist catalogue and its beginner guide, and two dated editorial articles. Targeted searches for distinctive card names did not locate an independent second account. The specialist site’s general source page did not provide a HANDS OVER-specific build record in the retrieved text.

The console-trailer link could not be fetched because YouTube returned a rate-limit response. We do not substitute a guessed platform list for that missing check. No account login, private Discord access or game installation was used.

What would upgrade a provisional entry

A dated in-game capture can establish current displayed text. A recorded controlled test can establish an outcome under stated conditions. Neither alone proves every interaction or future version. Record the build identifier, card or menu text, prior state, action and result together.

For release changes, preserve the old announcement alongside the new one. For bugs, separate the original report, a developer acknowledgement, an individual resolution and a patch note. These are different milestones. For translated text, record the language so that a naming mismatch is not mistaken for a mechanics change.