Skip to content

Roadmap

Lwando Zwane edited this page Aug 18, 2026 · 1 revision

Roadmap

Planned and possible work, roughly in order of interest. Not commitments.

  • Hands-free wake word — a "Hey Tome" wake word / continuous listening.
  • Workspace context for voice — let the voice assistant see the active workspace's brain context.
  • Multilingual whisper models — model selection (tiny/base/small, .en or multilingual) for the fallback engine.
  • SHA-256 pin for the downloaded whisper model.
  • Streaming model download — download to disk incrementally (today it's buffered in memory).
  • Updater — signed, notarized auto-updates once Apple credentials are configured.

Longer-term items live in docs/HANDOFF.md and the feature-plan docs in the repository.

Clone this wiki locally