Omniscio

Semantic Search (AI-powered meaning-based search)

Semantic Search lets Global Search find sessions by meaning rather than keywords, so a query like how do I speed up my deploy can surface a session titled Fixing CI pipeline latency. It runs on a local embedding model entirely on your machine, ships on by default, and blends invisibly into Ctrl+K results.

On this page

What it is

Semantic Search lets Global Search find sessions by meaning, not just keywords — so a query like "why is the deploy slow" can surface a session titled "Fixing CI pipeline latency" even though none of the query words appear in that title. It's powered by a local 33 MB embedding model (all-MiniLM-L6-v2 via Hugging Face Transformers.js) that Omniscio downloads the first time you enable the setting, then runs entirely on your machine — no data leaves your computer. Once built, each session's title plus first message become a 384-dimensional vector stored in SQLite, and every search query gets embedded the same way and compared against the index. Semantic hits are blended into the top of your Ctrl+K results invisibly — there's no visible "AI" badge, no semantic-vs-keyword toggle, no "Top N matches" wording. Omniscio just returns the most-likely results in order. It ships on by default.

Where to find it

How to use it

  1. Confirm it's on. Open Settings → Sessions → scroll to AI-Powered Search and check the toggle. If it's off and you flip it on, Omniscio downloads the model (~33 MB, one-time) and backfills embeddings for all your existing sessions in batches of 50. The backfill is silent — you can keep working during it.
  2. Search normally. Press Ctrl+K and type a meaning-based query: "bug in auth flow", "help me outline a blog post", "that Stripe thing last week". You don't need special syntax — semantic matching runs automatically when the eligibility criteria are met (default Ctrl+K state matches them).
  3. No UI to flip. As of late April 2026, the previous Semantic / Local toggle in the Ctrl+K palette is gone — semantic ranking blends silently with keyword results. The only user-visible control for semantic search is the master setting in Settings → Sessions.
  4. Want literal-only matching for one query? Use FTS5 syntax to short-circuit semantic ranking: wrap the term in quotes for an exact phrase ("refund flow"), prefix with - to exclude (refund -test), or use OR to combine literals (staging OR production). Semantic still runs, but FTS5 keyword hits dominate when your query is precise. There's no in-UI escape hatch beyond the global setting.
  5. When it won't fire. Semantic search only runs on clean queries: sorted by Most Relevant, no source/date filters, searchIn = all, tool actions excluded, channel type conversation, and first page. A query with quotes in it never gets semantic matches — quotes ask for exact text. The Ctrl+K palette's defaults match the eligibility criteria, so most searches work without thought; if you add filters or flip to Newest/Oldest, you get pure keyword results.
  6. Turn it off if you want. Back in Settings → Sessions, toggle AI-Powered Search off. Existing embeddings stay on disk (harmless) but no new ones will be created and searches will be keyword-only. No indexing happens, no model loads, no CPU spent.

How it behaves

It gives its memory back when you stop searching

The embedding engine only runs while something needs it. After a few minutes with no search, no save and no indexing work, Omniscio shuts the helper down and hands its memory (around 265 MB) back to the computer — three minutes on a weaker PC, ten minutes on a machine with 17 GB of RAM or more. Nothing is lost: the next search starts it again and waits for it, which on this model takes about one second (up to about three on a slow machine), and a search that runs out of that five-second budget still answers from keywords while the engine finishes loading for the next one. On a low-memory computer this is the difference between a helper sitting idle at 265 MB and the same memory being available to your agents.

If it repeatedly crashes on your machine (automatic safety)

The embedding model runs through a native engine (onnxruntime) that, on a small number of machines, can crash hard — historically an intermittent crash right as a new session starts indexing its first message (see crash-recovery.md and slow-computer.md). The engine now runs in its own separate process, isolated from the app: if it crashes, only that helper process dies — Omniscio notices, falls back to keyword-only search, and the app itself stays up. Your sessions and data are never affected — only whether meaning-based ranking is active. As a further backstop, Omniscio also self-heals: if it recovers from 3 of these embedding crashes within two weeks, it automatically turns Semantic Search off on the next launch and raises a dismissible inbox notice with a one-click Open Semantic Search settings button. Turning it back on in Settings → Sessions clears the counter and gives it a fresh start.

Under the hood there are two layers. The primary one is process isolation: the embedding engine runs in an Electron utilityProcess child (the "ML service"), so a hard native abort — the Cannot create a handle without a HandleScope crash, CID 9bf615c2 — kills only that child; the host degrades to keyword search and can respawn it. The backstop is the semantic-search circuit breaker: for an onnx abort that still reaches the main process AND is attributed to embedding, the next-launch crash-sweep records it in a rolling 14-day window in gpu-stability.json, and early bootstrap's applySemanticSearchCircuitBreaker disables semanticSearchEnabled past the threshold — before the engine can spawn again. Attribution matters now that more engines are isolated: embedding, local speech-to-text, and screen-recorder captions each run in their own child, so a main-process onnx abort can only come from the one engine still running onnx inline — wake-word. The crash-sweep therefore reads a persisted last-live-main-onnx-host breadcrumb (onnx-host-breadcrumb.ts, written by the wake-word engine) and trips this breaker ONLY when the abort was embedding's — a wake-word crash never disables semantic search (the pre-breadcrumb misfire). It is a confirmed-cause trigger (safe, unlike the GPU auto-disable, because the keyword-search fallback has no crash vector of its own); with isolation shipped it is now a near-vestigial last resort. Full envelope: ml-engine-reliability-contract.md; ORT teardown mechanics: embedding-worker-offload-contract.md.

  • global-search.md — keyword-side of the same Ctrl+K palette; semantic results augment keyword results, they don't replace them, and they're no longer visually distinguished
  • mempalace-memory.md — MemPalace drawer search uses its own semantic pipeline (different embeddings, different surface)
  • semantic-ai-session-search.md — the whole picture of finding a past session by meaning, including Session Search and escalations to an AI agent

Last verified 2026-09-23