Omniscio

AI Browser (in development — enable it in Settings → Lab)

The AI Browser is an in-development upgrade to the Embedded Browser for your own browsing — real multi-tabs, spaces, a command palette, bookmarks, history, downloads, an autofill vault, Memories and Journeys, plus the two point-at-the-page tools (Design mode and Visual Edit). Agents no longer drive it: each agent browses in its own Agent Web browser (amc-web). It ships off behind the Settings → Lab toggle aiBrowserEnabled, and while it is off nothing changes.

On this page

What it is

In development. This feature was retired on 2026-08-02 and un-retired on 2026-08-31; it is gated behind the ai-browser Lab toggle (aiBrowserEnabled, Settings → Lab, off by default) and works when you turn it on. For driving your own signed-in Chrome, the road is My Real Chrome + Browser Logins — a different feature.

Agents no longer drive this browser. It used to give an Omniscio agent its own isolated tab here, driven over an agent tool route and an MCP adapter. Both are retired: every agent now browses in its own Agent Web browser (amc-web), a real Chrome private to its session. This panel is yours — the agent-facing parts below say what changed.

This ships OFF. While the flag is off, the Browser pane is exactly the plain Embedded Browser — one page, no tabs. Nothing on this page happens until you turn the flag on.

The AI Browser turns Omniscio's embedded Browser pane into a real, multi-tab browser for you, with the AI tools pointed at your own tabs: edit a page element and hand the change to a coding session, rewrite a text field, record a workflow as a skill, and search what you've browsed.

Where to find it

Turning it on

  1. Settings → Lab → enable ai-browser. (Or launch with AMC_SHOW_AI_BROWSER=1 to reveal it.)
  2. Make sure the Browser panel is available: Settings → Features → Enable Browser.
  3. Open Browser in the Omniscio sidebar. You'll now see a tab strip instead of a single page.

How it behaves

Autofill vault (saved logins for your own tabs)

The Passwords tab in History is a small password manager for the browser pane's own tabs. Log into a site in one of your own tabs and Omniscio offers to save it: you type the username/password into a save dialog yourself (it doesn't capture what you typed on the page). Come back to that exact site later and a toast offers to fill it in for you. The list shows the site, username, and last-used date; remove a saved login any time. The password itself is encrypted through your OS's own credential store (Keychain / Credential Manager / libsecret) — the management panel only ever shows metadata, never the password.

This is a different feature from Browser Logins, and the two don't share data:

Autofill vault Browser Logins
Stores An encrypted password, in your OS keyring A cookie/session — never a password
Works in Your own tabs, in the AI Browser panel Your own in-app tab (cookies-only, best-effort); agents' browsers once imported into Agent Web
Reached from Browser → History → Passwords Settings → Browser Logins
Agent access None — no tool, no CLI route reaches it Only through Agent Web: import it in Settings → Agent Web → Saved logins (the old amc-browser road is retired)

A saved credential only fills on the exact site (scheme + host + port) it was saved on — it never carries over to a different subdomain. (Agent Web can sign an agent in with a saved Omniscio login through its own sign-in verb, with your approval the first time — see Agent Web.)

Memories (search pages you've already seen)

Omniscio quietly indexes the pages you visit in your own tabs, chunking and embedding the text with the same local embedding model Omniscio uses elsewhere (nothing leaves your machine; a duplicate visit to the same content isn't re-indexed). "What was that pricing page I looked at last week" becomes a search instead of digging through history.

Manage it from History → Memories: browse what's indexed, delete individual entries, clear everything, or turn indexing off entirely (it's on by default). (A session used to search it with an agent tool, browser_search_memory; that tool is retired with the AI Browser's agent API.)

Journeys (your browsing, auto-grouped by topic)

History → Journeys groups your recent browsing into topic clusters automatically — nothing to save by hand. It's computed from the same page embeddings Memories uses, bucketing pages that are topically close together and labeling each group after its most-recently-visited page. It's a lightweight, on-the-fly grouping (computed when you open the tab, not a background job or a precise research-session tracker) — good for "what was I looking at when I was researching X."

Write in field (retired)

The ✨ AI tools menu used to offer Write in field (also called Cursor chat / Compose): it captured the text field focused in your own tab and asked the session in the side chat to rewrite or draft its text. Writing the result back into the page relied on an agent tool that is gone now that agents browse only in Agent Web, so the item could read a field but never write into it. It was retired on 2026-10-17 and the menu no longer shows it. For help with a field's text, ask the session in the side chat and paste its answer yourself.

Workflow recording (turn what just happened into a skill)

The Record workflow item in the ✨ AI tools menu captures your clicks, typed values, form submissions, and page navigations in your own tab while it's on. While recording, a Stop button surfaces in the toolbar itself (never buried in a menu) so you can always end it. Hit Stop and a review panel lists every typed value it saw; you decide which ones become {{variables}} in the resulting skill — nothing is tagged automatically. A value that looks like a password shows up redacted and prompts you to tag it if you want it replaced with a variable rather than baked in literally.

It saves a draft skill — the same shape the Skills gallery (below) already understands, so it shows up there ready to run or edit. When it runs, the session replays the steps in its own Agent Web browser (amc-web), finding each element afresh. The honest ceiling: it's a literal, numbered replay of what happened, generalized only by the variables tagged at save time — it doesn't infer a broader workflow from a single recording. (An agent used to be able to record its own tab too; that went with the agent tools.)

Side chat pane

A collapsible panel on the right side of the Browser view holds a live chat with whichever session you pick — the same chat surface as the main Dashboard, just docked next to the browser so you're not switching views. This is the session Visual Edit and Design mode hand off to. Toggle it from the command palette or its own collapse control.

A small activity ticker under the pane header once narrated an agent's steps while it drove the panel's agent tab; that tab is retired, so the ticker is removed. Below the chat, quiet suggestion chips appear when the composer is empty — rule-based shortcuts for the current page ("Get the transcript" on a YouTube video, "Review this PR" on GitHub, "Summarize this page" anywhere). Clicking a chip only pre-fills the composer; nothing is ever sent until you press send, so a stray click never costs a paid call.

Skills (save a workflow as a one-click command)

A skill is a browser workflow you save in plain English and re-run in one click — "Summarize this page", "Find the cheapest flight and add it to my cart", "Pull the action items off this doc". It's the Omniscio take on Dia's Skills / Comet's Shortcuts.

Open it from the History toolbar button → Skills tab in the browser. There you can create a skill (a name, the plain-English steps, and an optional emoji for its card), edit or delete existing ones, and run any of them. Typing one out isn't the only way in — see Workflow recording above for recording one instead.

Under the hood a skill is a recipe — there's no separate "skill engine". Saving a skill writes a browser-scoped recipe (a normal Omniscio recipe flagged as a browser skill), so it shows up in the Skills gallery. Running one goes through the exact same gated run path as any recipe: it opens the standard Run dialog (pick a project, fill any inputs, confirm) and starts a recipe run — it never spawns anything directly. The session browses in its own Agent Web browser, under Agent Web's site rules: sites you block stay blocked, and an action on a finance, payment or social site you have not set a rule for waits for your yes on a card. A skill is a shortcut to a workflow, not a way around the guardrails.

v1 is your own skills only — there's no community gallery or marketplace yet.

Video pages and PDF pages

A "summarize this YouTube video" or "what does this PDF say" request now goes to a session that browses in its own Agent Web browser (amc-web open <url>, amc-web read), or to a skill that reads files (the pdf-extract skill for a saved PDF). The AI Browser's dedicated transcript and save-as-PDF agent tools are retired with its agent API.

Two other paths do not cover these, for context:

  1. The summarize skill does a plain WebFetch of a URL's HTML — on a YouTube watch page that's just the shell; the transcript loads client-side.
  2. POST /download (yt-dlp) saves a media file (video | audio | original), never subtitle text, and has no extractor for a plain PDF URL.

For a Google Doc, POST /gdoc/import (src/main/services/cli/cli-server-gdoc-routes.ts) pulls its real content as Markdown via the Drive API.

Automations & co-work

Three common jobs are shipped as recipe templates in the Pattern Library (Recipes → New from pattern), not as new machinery. Each is a normal recipe whose session browses in its own Agent Web browser (amc-web open, amc-web read, amc-web tabs) plus the session's own tools; there is no new scheduler, no new engine, and no new file route. Pick one, fill in the blanks, and it becomes a runnable recipe.

Recurring browser jobs

"Every morning, open these dashboards, pull the numbers, and write me a summary." That is the Recurring browser report pattern (recurring-browser-report): a one-step recipe whose prompt tells the agent to open and read each URL with amc-web, extract what you asked for, and save a dated markdown summary via its own Write tool. It is read-only by design (no clicking or submitting), which is what makes it safe to run unattended.

To make it recurring, wire the instantiated recipe to a cron job on the existing cron system — nothing new is built for scheduling:

  1. Instantiate the pattern into a recipe (it lands in your recipe list with an id).
  2. Create a recurring cron job of type: "recipe" pointing at it. Over the CLI control server that is a POST /cron/jobs with runMode: "recurring", a cronExpression (e.g. 0 8 * * * for 8am daily), and jobConfig: { "recipeConfigId": "<the recipe id>" }. The cron recipe executor (cron-recipe-executor.ts) already dispatches it on schedule.
  3. A cron job created via the API always registers as pending approval — you approve it once in the inbox before it starts firing.

Agent Web's site rules apply to a scheduled run as to any other: sites you block stay blocked, and an action on a finance, payment or social site waits for your yes on a card, so a recurring job is fine for reading dashboards, prices, or status pages and cannot quietly buy or post anything.

Research to report (co-work)

The Browser research to report pattern (browser-research-to-report) is the co-work shape: one session researches a topic with amc-web open, amc-web read and amc-web tabs, cross-checks across sources, then writes a cited markdown report with its own Write tool to its working directory. No file route or share pipeline is involved — the report is just a file the session wrote where you can open it. Run it as a normal session (you watch and steer) or on a schedule like above.

Guided shopping / booking (confirm-gated)

The Guided shopping pattern (guided-shopping) is a three-step, confirm-gated recipe for "compare these, then buy the best one":

  1. Compare — the agent reads listings/prices across the sites you named with amc-web and produces a shortlist plus a proposed purchase plan. It changes nothing.
  2. Approval gate — the recipe pauses at a human approval step that shows you the plan. Nothing proceeds until you approve.
  3. Checkout — only after approval, the agent adds the item to the cart and drives to the checkout page, then stops at the payment step. It never enters card/address details and never places the order; you review and pay by hand.

This is the load-bearing invariant, enforced at two layers: the recipe's own approval gate sits before the checkout step, and underneath it Agent Web asks you on a card before acting on a payment site it has no rule for. Booking flows follow the same shape — compare, approve, drive up to the confirm/pay step, hand off.

Privacy + safety notes

  • No agent drives this panel. Agents browse only in their own Agent Web browser; the agent tool route that drove this panel's agent tab (/ai-browser/tool) is retired and answers 404, and the agent tab itself is retired too, so every tab in the panel is yours.
  • History and bookmarks are yours — only your own tabs are recorded.
  • Memories only indexes your own tabs.
  • The autofill vault is invisible to agents — no tool or CLI route reaches it; it's exclusively for your own tabs.
  • Everything is local — no browsing data leaves your machine, and the panel's CLI routes (history, spaces, downloads, bookmarks) are bound to 127.0.0.1 behind the CLI bearer token.
  • The embedded page is treated as untrusted and runs sandboxed, exactly like the plain Embedded Browser — the AI Browser doesn't weaken that.

What's still coming

Design mode's element reordering is button-stepped (move up/down among siblings), not drag-and-drop. Workflow recording generalizes only the values explicitly tagged as variables at save time; it doesn't infer a broader workflow from one recording.

Agents browse in Agent Web. The road for driving a browser you are signed into yourself is My Real Chrome, which drives your own signed-in Chrome through a companion extension, and Browser Logins for reusing a captured session. The plain single-page pane it was built on is Embedded Browser, and the local HTTP surface its people-side routes live on is cli-control.md.

Material that outgrew this page — the two point-at-the-page tools (Design mode and Visual Edit), spaces, tabs and quick navigation, the retired agent tab and the per-site permission tiers — continues in AI Browser (part 2).

Last verified 2026-10-06