Decompose launcher/config.ts into focused domain parser and CLI normalization modules to reduce refactor risk while preserving command behavior. Align Jellyfin launcher config with session-based auth by dropping config token/userId dependency.
- add subtitleStyle.hoverTokenColor config default + validation
- normalize hover color payloads and propagate configured color to mpv runtime
- refresh invisible overlay tokenization with current subtitle text and tighten hover overlay cleanup hooks
- record TASK-98 and subagent coordination updates
Disable Anki-dependent shortcuts when AnkiConnect is off and require jellyfin.enabled for remote startup warmups to avoid initializing disabled integrations.
Split config defaults, option metadata, runtime-option registry, and template sections into domain modules while preserving the composed public API in definitions.ts. This keeps ConfigService behavior stable and makes future config extensions easier to add with focused regression coverage.
Move OSD log appends off sync fs calls to reduce main-process blocking under frequent OSD activity. Add buffered flush wiring into quit cleanup so pending log lines are drained best-effort during shutdown.
Collapse src/config/resolve.ts into an orchestrated pipeline over domain modules, wire launcher regression coverage into test scripts, and sync backlog/subagent tracking artifacts for completed TASK-74/TASK-96/TASK-98 follow-up planning.
Split note-update and field-grouping orchestration out of AnkiIntegration so the facade remains focused on composition and shared policy wiring. This keeps mining behavior stable while creating focused workflow seams with dedicated regression coverage and clearer ownership docs.
Introduce runtime domain barrel exports and registry composition support, and document/check main runtime fan-in workflow with related backlog planning artifacts.
Move IPC, shortcuts, startup lifecycle, and app-ready assembly behind dedicated runtime composers so main.ts stays focused on boot wiring while preserving behavior and test coverage.
