FAQ

FAQ

Is my data private?

Yes. SenseTree runs offline and computes embeddings on your machine. LLM reasoning/vision calls go only to the endpoints you configure. The one way data leaves your device is if you point a slot at an external cloud API. No telemetry, no background uploads.

Where is my data stored?

Under %APPDATA%\com.virgi.sensetree:

• settings.json — configuration,
• a SQLite database — file metadata, indexing queue, extracted summaries,
• a LanceDB directory — vectors.

Deleting that folder resets SenseTree completely.

Does SenseTree move or modify my files?

Never without your explicit approval. It reads files to index them. Any move/rename/delete is proposed as a Dry-Run plan you Approve or Discard, and apply is transactional with rollback. See AI Chat & Actions.

What's the difference between the .exe and .msi?

Both install the same app as a normal Windows program.

• -setup.exe (NSIS) — lightweight wizard, per-user, no admin needed. Recommended for individuals.
• .msi — enterprise packaging for mass deployment (GPO/Intune/SCCM), silent install, usually per-machine (admin).

See Installation.

Will the next release update my install?

• Manual reinstall = yes, in place. Download the next installer of the same type and run it; because the app identifier is stable, it upgrades the existing install (single entry, data preserved).
• Automatic (app updates itself) = not yet. There is no updater configured, so the app doesn't self-update. You download new versions from Releases. In-app auto-update is planned for a future release.

Do I need a GPU?

No, but it helps. Embedding runs fine on CPU. The LLM reasoning/vision models are much happier on a GPU. See per-hardware picks in Models & Providers.

Do I need Ollama or LM Studio?

Only for reasoning and vision. Embedding (and thus basic semantic search over text) works with the built-in local engine, no extra install. To chat, plan reorganizations, or caption images, point those slots at a local runner.

Why did indexing restart after I changed a setting?

Changing the embedding model or its dimensions invalidates all vectors, so SenseTree does a full re-index. Changing the block/recursive slider or folder_classify prompt makes it forget prior folder classifications. Reasoning/vision changes need no re-index. See Configuration.

Which folders should I not index?

You don't have to micromanage this — the block/recursive classifier automatically treats venv, node_modules, app bundles, DAW packs, caches, and build artifacts as opaque blocks. Tune its aggressiveness with the slider.

Is it cross-platform?

The current builds target Windows. The stack (Tauri + Rust + fastembed/LanceDB) is cross-platform in principle, but Windows is what's built and tested today.

How do I report a bug or request a feature?

Open an issue or a PR. Include your OS, the version (Help/About or the installer name), and relevant tracing logs if it's a runtime problem.