Support

Browse my Chrome Guide

Let your CellCog AI agent browse and act on the web using your real Chrome browser — your sessions, your bookmarks, your profile.


What is Browse my Chrome?

Browse my Chrome lets CellCog AI agents drive your real Chrome browser during a chat session. Instead of using a generic cloud browser, agents work inside your Chrome — using the sites you're already logged into, the cookies you've already set, and the bookmarks you've already saved.

Once enabled, an agent can:

  • Open and navigate websites in your authenticated Chrome session
  • Read page content, forms, and accessibility structure
  • Click buttons, fill forms, take screenshots, and capture page state
  • Adopt tabs you already have open and act on them
  • Work across multiple sites in the same chat

Key Principle: You stay in control. The agent works inside a dedicated cyan "CellCog" tab group in your Chrome window — that's the shared workspace. Tabs outside that group stay private. Every action is classified by risk and gated by your approval threshold. Chrome itself shows a yellow safety banner on every controlled tab that you can dismiss to revoke access at any time.


Browse my Chrome vs Cowork on my PC

Browse my Chrome is the peer of Cowork on my PC. Both require the CellCog Desktop app — they extend the agent's reach in two different directions:

table
UseCowork on my PCBrowse my Chrome
Edit local files (code, docs, etc.)
Run terminal commands on your machine
Visit and interact with websites
Use logged-in sessions (Gmail, GitHub, banking)
Fill forms, click buttons, take page screenshots
Trust boundaryWorking directory on diskThe "CellCog" tab group

You can have both enabled in the same chat. Many real workflows use them together (e.g., "research these 8 SaaS products in Chrome, then write a comparison report into my project directory").


Getting Started

Step 1: Install the CellCog Desktop App

Browse my Chrome runs on top of the CellCog Desktop app — the same app used by Cowork on my PC.

  1. Go to Browse my Chrome in the sidebar (or visit /cowork/browse)
  2. If you haven't installed the Desktop app yet, follow the prompts to download it for your platform (macOS, Windows, Linux)
  3. Install, open, and sign in

If you already have CellCog Desktop installed and connected (from using Cowork), skip to Step 2.

For full Desktop setup details see the Cowork on my PC Guide.

Step 2: Install the CellCog Browse Chrome Extension

  1. On the Browse my Chrome page, locate the Chrome profile you want to use
  2. Click "Install →" next to that profile (this opens the Chrome Web Store listing)
  3. Add the extension to Chrome

Step 3: Open Chrome with the Chosen Profile

The Chrome extension lives inside a specific Chrome profile. After installation, make sure that profile is open in Chrome so the extension can connect to the Desktop app.

You'll know it's working when:

  • The Browse my Chrome page shows a green "Ready" strip
  • The profile name appears beneath it (e.g., "Personal" or "Work · acme.com")

Step 4: Set Your Approval Threshold

Before enabling Browse in a chat, decide how much you trust the agent's self-classification.

On the Browse my Chrome page, find the Auto-approve threshold section and pick one of four levels — see Threat Levels and Auto-Approval below for details. The default is Safe, the strictest meaningful setting.

Step 5: Enable Browse for a Chat

  1. Start a new chat or open an existing one
  2. In the message editor toolbar, click the globe icon (the Browse selector)
  3. Pick the Chrome profile you want this chat to use
  4. Send your message

When you enable Browse, Cowork is also enabled automatically — Browse requires the Desktop app to be connected, and Cowork is how the chat talks to the Desktop app.


Using Browse my Chrome in Chats

The CellCog Tab Group

When the agent first opens a tab in Chrome, it creates a cyan tab group labeled "CellCog" in your Chrome tab strip. This group is the shared agent workspace:

  • Every browse-enabled chat in the same Chrome profile sees the tabs inside this group
  • Tabs outside the group are private to you — the agent cannot read or interact with them on its own
  • You can collapse/expand the group, drag it around, close individual tabs, and it survives Chrome restarts (it's a standard Chrome tab group)

Adopting Tabs You Already Have Open

If you already have a tab open that the agent should work with (e.g., your half-completed visa form), the agent can adopt it into the CellCog tab group. Because adopting necessarily moves your personal tab into the shared workspace, this action is classified as moderate and respects your threshold — at threshold Safe it requires your approval each time.

Snapshots, Screenshots, and Form Interaction

The agent perceives pages in three main ways:

  • Snapshot — an accessibility-tree view of the page (headings, links, form fields, button states). This is the primary way the agent understands what's on the page.
  • Screenshot — a visual capture for review or to share with you in chat.
  • Form fills, clicks, navigation — the agent can fill inputs and click buttons just like a person, including across logged-in flows.

Multiple Profiles

If you have multiple Chrome profiles (e.g., Personal + Work), each one needs the extension installed separately. Once installed, each chat is bound to one specific profile via the Browse selector in the editor. The Browse my Chrome page shows the install state of every profile so you know exactly which one is missing the extension.


How the Agent Interacts with Pages

The agent uses Chrome's built-in chrome.debugger API (the same API used by Chrome DevTools and other AI-browser tools like OpenAI Codex and Browserbase) to navigate and act on pages. There's no separate browser launched and no special launch flag — it's your real Chrome doing real things.

When the agent reads a page, it gets:

  • Page URL and title
  • Visible text and accessibility tree
  • Form field values (passwords and credit-card fields are always redacted)
  • Form state (checkboxes, radio buttons, dropdown selections)

When the agent takes an action, it can:

  • Navigate to a URL
  • Click buttons and links
  • Fill form fields
  • Press keys
  • Take screenshots
  • Read or set cookies / localStorage (declared as moderate or dangerous depending on direction)

Threat Levels and Auto-Approval

CellCog uses a transparent agent-classifies, user-threshold-gates model. The agent classifies every action by what data it touches; you control which classifications auto-approve.

How the Agent Classifies Actions

Every browser action falls into exactly one of three risk levels, based on a simple question: what data does this command touch?

table
LevelWhat it meansExamples
SafeNo private read, no data change.Reading a Wikipedia article, navigating to a public URL, expanding a menu, searching on google.com
ModerateReads your private data but changes nothing.Snapshotting Gmail / a private repo / banking, reading document.cookie, listing or adopting personal tabs
DangerousChanges data or triggers an irreversible action.Submitting forms, sending email, paying, deleting, writing cookies, downloading files

The agent shows you both the classification AND a one-sentence reason on the approval card. You're always picking how much to trust the agent's own self-classification — not picking a category of content the system identifies.

Setting Your Threshold

The Browse my Chrome page has four threshold levels. Anything above your threshold still requires your manual approval on the approval card in chat.

table
ThresholdWhat it auto-approves
NoneNothing — every browser action requires your approval. The strictest setting.
Safe (default)Side-effect-free actions on public content. Reading your private pages and any data changes still require approval.
ModerateAlso auto-approves actions that read your private data — logged-in page snapshots, cookies, personal-tab discovery. Anything that changes data still requires approval.
DangerousAuto-approves everything the agent declares — including actions that change data. The most permissive setting.

Most users settle into Safe or Moderate. Dangerous is intended for advanced workflows where the user trusts a specific agent role completely (e.g., a project-scoped agent reviewing their own work) — not a default.

The Yellow Banner — Chrome's Enforced Safety Indicator

Chrome shows a yellow banner across the top of every tab CellCog is driving:

"CellCog Browse started debugging this browser."Cancel

This is Chrome's built-in safety indicator. CellCog cannot hide or suppress it. Click "Cancel" at any time to immediately revoke debugger access for that tab — the agent will see an error on its next attempt and can prompt you to re-enable in the Browse my Chrome settings.

Revoking Access Entirely

You have multiple escape hatches, at different scopes:

  • Single tab: Click "Cancel" on the yellow banner
  • Single chat: Toggle the globe icon off in the editor toolbar
  • All chats in this Chrome profile: Disable the extension at chrome://extensions/
  • All of CellCog on this machine: Quit the CellCog Desktop app

Privacy and Safety

What the Extension Can See

When Browse is enabled for a chat:

  • Page content (text, URLs, accessibility tree) of tabs inside the CellCog tab group
  • Form values on those tabs (excluding passwords and credit-card fields — those are auto-redacted)
  • Screenshots of those tabs when the agent captures them
  • The list of all open tabs in your Chrome profile (only when the agent runs Browser_Tabs list, which is classified moderate)

What the Extension Cannot See on Its Own

  • Tabs outside the CellCog tab group (your personal browsing)
  • Browsing history
  • Bookmarks
  • Saved passwords (Chrome's password manager)
  • Tabs in a different Chrome profile (each profile is independent)
  • Anything when Browse is not enabled for a chat

Sensitive-Data Redaction

The extension automatically redacts before any data leaves Chrome:

  • Password fields (type=password)
  • Credit-card number fields (autocomplete=cc-number)
  • Credit-card CSC fields (autocomplete=cc-csc)

These values are replaced with <redacted> in everything the agent reads.

Cookies, localStorage, Screenshots

These are not auto-redacted because they sometimes legitimately need to flow through the agent (e.g., the agent reads a session cookie to make an authenticated API call on your behalf):

  • Cookies and localStorage — only read when the agent's script explicitly requests them. Classified moderate for reads, dangerous for writes.
  • Screenshots — capture whatever is visible. If you're on Gmail or a banking page, that's what the screenshot will show. Classified at the agent's discretion based on the page content (typically moderate for logged-in surfaces).

For full disclosure of these data flows, see the Privacy Policy.

Where the Data Flows

  • Between Chrome and CellCog Desktop: locally via Chrome's Native Messaging — nothing leaves your machine for the extension's own operation
  • From CellCog Desktop to CellCog's servers: encrypted in transit (TLS), as part of your authenticated chat session
  • From CellCog to the LLM provider serving your chat: page content the agent reads becomes part of the chat context and is transmitted to the LLM provider (Anthropic, OpenAI, or Google depending on the model chosen for that chat). Each provider has zero-data-retention API terms.

Troubleshooting

Extension installed but the page says "Not connected"

  1. Make sure the matching Chrome profile is actually open in Chrome. The extension only connects when its profile is running.
  2. Make sure CellCog Desktop is connected. Check /cowork for a green "Connected" status.
  3. Reload the extension. Go to chrome://extensions/, find CellCog Browse, click the reload icon. Then revisit the Browse my Chrome page.
  4. Open any cellcog.ai tab in that Chrome profile. Some connection paths use a wake signal from a cellcog.ai page to kick the extension's service worker awake faster.

"Browse my Chrome is ready" but my chat says Browse is not connected

This usually means the chat is bound to a different Chrome profile than the one with the extension installed.

  1. Open the globe icon in the editor toolbar
  2. Confirm which profile is selected for this chat
  3. Either install the extension into that profile OR switch the chat to a profile that has the extension

Yellow banner won't dismiss

The yellow banner is Chrome's built-in indicator and only goes away in two situations:

  • The agent finishes its operation and the debugger detaches automatically
  • You click "Cancel" on the banner (which forcibly revokes access for that tab)

If neither has happened and the banner persists, check that the CellCog Desktop app is still running. If Desktop quits mid-session, the debugger may stay attached until you reload that tab.

Agent says "Tab not in CellCog group"

The agent tried to act on a tab that hasn't been adopted into its workspace. Either:

  • Ask the agent to adopt the tab first (this is a moderate-level action and respects your threshold)
  • Or open the URL fresh inside the CellCog group (the agent can do this with Browser and goto)

Wrong Chrome profile keeps getting picked

When you click "Install →" in CellCog, Chrome opens new tabs in whatever profile is currently focused — not necessarily the profile you intended. If the extension landed in the wrong profile:

  1. Disable / uninstall the extension from the wrong profile via chrome://extensions/
  2. Switch to the intended profile in Chrome
  3. Re-click "Install →" from the CellCog Browse my Chrome page

Frequently Asked Questions

Why do I need both the Desktop app AND the Chrome extension?

The Desktop app is the trust boundary — it authenticates you and brokers every command to and from CellCog's servers. The Chrome extension is the small piece that talks to Chrome on the Desktop app's behalf. Splitting them keeps the powerful pieces (the Desktop app) under your direct control, and keeps the extension intentionally minimal.

Does this work on Firefox or Safari?

Not currently. Browse my Chrome supports Google Chrome only at launch. Microsoft Edge and other Chromium-based browsers are on the roadmap; Firefox and Safari are not currently planned because they don't expose an equivalent to chrome.debugger.

Can the agent see tabs I have open outside the CellCog group?

The agent can list them (only when you grant approval — Browser_Tabs list is classified moderate and requires your threshold to be ≥ moderate to auto-approve). It cannot interact with them. To interact with one, the agent must adopt it, which moves the tab into the CellCog group and is a separate, also-moderate action.

Will my passwords or credit card numbers be sent to CellCog?

No. Password fields and credit-card fields (number + CSC) are always redacted before any data leaves Chrome. The agent sees <redacted> in those fields and cannot retrieve the real values.

What happens to screenshots? Are they sent anywhere?

Screenshots are saved to a temporary directory on your machine and made available to the chat session (so the AI can analyze them). They're transmitted to your chosen LLM provider as part of the chat context. They're not stored permanently by CellCog or shared elsewhere.

What about banking pages or email?

You can use Browse my Chrome on any site you like. Snapshotting a banking dashboard is classified moderate (reads private data, changes nothing); clicking "Transfer $1000" would be dangerous. Your threshold setting decides what runs without your click — most users keep the threshold at Safe or Moderate when working with sensitive surfaces.

Does Browse my Chrome cost extra?

No. Browse my Chrome is included with CellCog. Standard credit usage applies to the agent operations the same way Cowork does — credits are consumed by the agent's reasoning and tool calls, not by the act of having Browse enabled.

Can teammates in my organization use my Browse session?

No. Browse runs against your Chrome profile on your machine, authenticated as you. Organization members can't access your machine through CellCog. If they want to use Browse, they install the Desktop app and the extension on their own machine.


Related Guides