Contributing toward the Rosetta-free goal

[postmelee]

There is no player-ready Rosetta-free preview yet. This track is native runtime engineering toward the final Pear goal; improving the current Rosetta Preview does not automatically qualify M1.

Start by reading the architecture, contributor guide and the current Issue before proposing work. Useful areas include native host observations (#10), startup and Steam integration (#11 and #46), game qualification (#12/#16), runtime assembly (#13) and regression evidence (#14). These are existing work areas, not claims that each is unassigned.

Documentation, portable fixture tests and source-inventory improvements can help without a provisioned runtime or a game. Full Wine/FEX/DXMT reconstruction still has exact source/build prerequisites and prepared local inputs; a clean clone is not yet a complete player runtime. Record missing inputs instead of importing private signing material or authenticated Steam environments.

Keep source inspection, synthetic tests, native host execution, game qualification and ordinary-user delivery distinct. Rosetta-free game claims require the relevant process and library architecture evidence across startup, play, exit and resume, alongside the gameplay requirements.

Reuse upstream implementations and retain their languages unless measurements justify a change. FEX's contribution policy restricts AI-generated contributions; respect upstream policies and licensing independently of the tools allowed for Pear. No AI subscription or internal task-report bundle is required from external contributors.

Public first-build and separate contribution-path guides are being prepared in Task #47. Ask in Q&A or the relevant Issue before taking overlapping work.