Track tested and unverified host and game compatibility #21

Candidate status

Deferred by Cycle 4 gate #23 after selecting #22. Keep this issue outside the active milestone; reconsider if manual host/game validation capacity or direct consumer evidence becomes available.

Opportunity

Publish an evidence matrix that distinguishes OpenController's tested SDK report/helper behavior from unverified host, browser, Steam, OBS, and game consumption. Integrators currently have to infer where a generated controller path was actually validated.

Evidence

• Quasar #348, opened 2026-09-23, describes a Steam virtual gamepad whose mapping differed from browser recognition; an attempted host fix then blocked sessions until a malformed controller label was corrected. Concrete cross-layer failure in another project, not an OpenController defect.
• Steam for Linux #13616, opened 2026-09-14, reports a controller that worked in Steam, desktop, and other games but was not passed through Steam Input to Spyro. One report; upstream consumer behavior is outside this SDK.
• OpenController already has profile-to-report conformance docs/tests and native doctor/test commands. Those checks cannot establish that a physical device appears correctly in every OS, browser, Steam configuration, OBS runtime, or game.

Candidate acceptance if selected

• Separate CI-validated SDK encoding/helper contracts from manual host or consumer validation.
• Track backend, profile, OS/runtime, test type, exact environment, date, and outcome; label untested combinations unknown/not tested.
• Link reproducible commands and state explicit non-claims for physical hardware and game behavior.
• Do not duplicate the existing profile-to-report matrix or call dry-run output a host compatibility test.

Initial estimate: medium, 2–4 days, with validation gaps due to hardware and OS availability. This is a troubleshooting aid, not a promise to fix SDL, Steam, OBS, or game-specific mapping.