layout lab: hero 1 — hero 2 — hero 3 — hero 4 · you are here — hero 5 — hero 6

Om-E · intelligence dashboard

Everything you've ever read, ready to answer.

Om-E is an agentic AI app living inside your browser. It takes the unstructured mess you work in — pages, PDFs, transcripts — and transforms it into searchable knowledge on your machine, engineered into exactly the context your AI needs. Sharp, cited answers, from your world.

Get the beta What is Om-E? ↓

macOS · Chrome extension · local-first

· explore ·

What is Om-E

An app that lives in your browser.

Not another website you paste things into. Om-E installs on your Mac, connects to Chrome, and turns the browsing you already do into knowledge you keep. This is it:

The real Om-E dashboard glowing as the single bright panel in a vast near-black glacial cathedral hall — chats and knowledge bases in the sidebar, a live agent answer in the main pane — the small ice robot watching from the shadows

01 An app, not a tab

A macOS app with a Chrome extension — it wraps around the browser and the tabs you already work in.

02 Your machine, your knowledge

Everything you capture is vectored into local knowledge bases. Nothing uploads, nothing expires.

03 An agent, not a chatbot

It browses, captures, drafts and saves work back — answering from your context, not guesswork.

A great ice hourglass encased in a shell of stacked database rings: chaotic glowing pages pour through the neck and settle as a neat sealed stack

How this works

Open tab to cited answer, in six moves.

scroll to walk the loop ↓

Om-E · your agent

Sitting on everything you keep.

Every page you capture becomes another layer under it. Ask a question and Om-E deals out exactly the right slices of what you know — which is the part nobody else does:

The Om-E robot carved from crystalline ice, sitting on a glowing tower of stacked knowledge-base discs, one arm raised toward the headline beside it with violet document tablets trailing from its open hand

World first

The world's first AI context engineering tool.

Every answer is only as good as the context it was given. Om-E makes that window yours to run — attach a KB for retrieval, load a topic in full, or send the question bare. What the model sees, what it ignores, and what you pay: dialled per turn.

Attach vs load

Retrieval-only when the KB should answer, full text when the model must read every word.

Scoped per chat

Each chat sees the knowledge you point it at — nothing else bleeds in.

Cheaper by design

Tokens are the bill. Sending less, better-chosen context is the discount.

Sharper answers

Less noise in, less noise out — the model reasons over what matters, not everything.

the full story →

The real Om-E attach panel floating in the glacial knowledge library — threads of violet light flowing from the glowing shelf books into the chat's context, the ice robot on a ladder pulling one more book — Attaching knowledge, literally — the library flows into the chat's context, one topic at a time.

Om-E's full-context load telemetry — characters injected, estimated prompt tokens, character cap, and the exact loaded scope — No black box: every turn shows exactly what was injected — 24,720 chars, ~6,180 tokens, 93 source messages — and what it cost.

The machine

Four moving parts. One workspace.

Chats

Where the work happens — every conversation saved, searchable, resumable.

Topics

The unit of memory. Anything worth keeping, saved once and reused forever.

Knowledge Bases

Libraries of topics that search by meaning — your private index, on your machine.

Multi-session

Everything runs at once. Fire a research run in one chat, keep working in another.

Two real Om-E browser windows streaming side by side in the ice hall, documents pouring into both, the ice robot conducting between them with threads of light to each panel — Not a metaphor — two runs streaming at once, and the agent conducts.

gate

Built for trust

Answers you can hand to someone else.

Everything Om-E hands you carries its receipts — where each claim came from, how it scored against your rubrics, and how the sources stack up. Ship the work; the evidence travels with it.

Citations

Sources travel with the answer — whoever you hand it to can check it themselves.

Verification

Score sources against your own rubrics — evidence, not vibes.

Comparison

Source vs source on your criteria, laid out where each one wins.

Templates

Turn captured research into finished documents in your format.

Receipts

Every claim cited. Every answer verifiable.

Answers are built from your captured sources, and every claim carries its citation — the document, the section, the evidence reference. Then run a verification pass: the agent re-checks every claim against the sources and stamps a verdict under each one — including the ones that fail.

The real Om-E verification report with green CONFIRMED banners floating beneath a giant handshake carved from ice, the ice robot pressing an emerald seal onto the document's corner — A verified answer, claim by claim — sealed under the handshake.

Om-E's Claims Needing Review table — claims marked Partially Confirmed, Insufficient Evidence or Contradicted, each with the reason and its evidence references — It shows its doubts — weak or contradicted claims are flagged for review, not hidden.

An Om-E context verification report — assessment summary showing 33 of 38 claims supported, the method, duration, sources assessed and verdict breakdown — The audit trail: 33/38 supported, method, duration, verdicts — saved as a topic you can hand to anyone.

Citations

Sources travel with the answer — whoever you hand it to can check it themselves.

Verification

Score answers against rubrics you define — the judgement is inspectable, not vibes.

Comparison

Two proposals, two policies, two vendors — where they differ, who wins on which criterion.

Templates

Turn captured research into reports, briefs and memos in your format — citations intact.

gate

The agent's reach

From one question to a finished report.

Deep research

Multi-source research runs that come back as sourced, finished reports.

Web search

Built-in search — the agent searches, opens the tabs and reads them for you.

Agentic

An agent with tools, not a textbox — it browses, captures, drafts, and saves the work back into your KB.

Personas

Same knowledge, different eyes — review it as a researcher, a governance officer, a lawyer.

Deep research

Multi-source research runs that come back as sourced, finished reports — not a pile of links.

Web search

The agent searches, opens the results, reads what's worth reading — and what comes back lands in your KB as knowledge.

Local-first

Nothing uploads. Nothing expires.

Your KBs, files and keys live on your machine — the index is local, retrieval is local, and the only thing that ever leaves is the turn you chose to send.

how we handle your data →

Tokens, not seats

Buy a pack. Spend it on turns.

No subscription, no per-seat maths. Credits ride a hard-capped key, sharper context costs less per answer, and the meter is always yours to read — that's it on the right, straight from the app.

see the packs →

The real Om-E account panel — purchases, unit balance, usage chart — floating beside a sealed ice vault, glowing credit discs arcing from the vault's keyhole into the meter while the ice robot inspects one — The meter, the ledger, the top-up — credits flow from your vault, and the meter is always yours to read.

Stop losing knowledge in tabs.

Ask the page. Save the answer. Build your knowledge base.

Get the beta See it in use