Skip to content

New

How detectable are you right now?

Browsers for AI agents

Give each agent its own isolated browser with a persistent identity. Stay signed in, handle the captcha, type like a human, and take over whenever you want.

  • MCP

    Point Claude, Cursor or your own agent at it, over HTTP or stdio.

  • SDKs

    Python and Node, both on direct CDP. Keep your agent logic, swap the browser underneath it.

  • Direct CDP

    Already running browser-use? One line of connection code and the run starts signed in.

251,710

Successful runs in past 7 days

98.93%

Session goal reached

4.2s

Average time to a signed-in page

1.07%

Runs that needed a human

240 teams run their agents on it

#01 - The browser

Give every agent a machine that stays the same one.

The identity comes from the profile seed. Same seed, byte-identical values across restarts and across machines.

seed b0fb2d81-4ec2-41bf-a865-5acf4ae947c8

Persona

Coherent fingerprint axes, derived from the seed. Read-only.

windows
Runs on
Local native Windows
Fixed at creation: moving a profile to another tier means creating a new one.
User agent
Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/148.0.0.0 Safari/537.36
Platform
Win32
Chrome
148.0.7778.215
Screen
1920×1080 @2x
CPU / Memory
8 cores · 8 GB
Timezone
Europe/Vienna
Locale
de-AT
WebGL renderer
ANGLE (NVIDIA, NVIDIA GeForce GTX 1650 (0x00001F0A) Direct3D11 vs_5_0 ps_5_0, D3D11)
GPU
NVIDIA NVIDIA GeForce GTX 1650 (d3d11)
WebGPU
NVIDIA GeForce GTX 1650
Fonts / Voices
58 fonts · 19 voices
Noise seed
17265654078529133351

One browser per profile: its own logins, its own cookies, its own device. Two of them share nothing.

Passwords live encrypted and get typed into the form. The agent never receives the value: it appears in no argument and no log.

Keys survive a crash and the window closing. Permission bubbles and device pickers get answered instead of stopping the run.

See what the agent gets

#02 - The agent layer

Hand your agent a page it can operate.

The agent looks at the page once, then only at what moved. Long runs stop dying on a full context window.

shopify.com/login

First look

31 lines · 935 B

title: Log in — Shopify
[e_122] heading "Log in"
[e_123] heading "Continue to Shopify"
[e_49] button "Sign in with passkey"
[e_25] link "Continue with Google"
[e_28] link "Continue with Apple"
[e_31] link "Continue with Facebook"
[e_34] link "Continue with WhatsApp"
[e_16] button "Continue with email" (disabled)
[e_3] textbox "Email"
[e_39] link "Get started"
[e_208] link "Need Help?"
and 19 more

After one keystroke

4 lines · 182 B

+ [e_277] StaticText "Continue with email"
+ [e_159] image
~ [e_16] button "Continue with email" → no longer disabled
~ [e_3] textbox "Email" → = "ops@northgate.example"

Every step after the first carries the change, not the page.

Pointer paths with overshoot, log normal dwell and flight times, typos, momentum scrolling. Calibrated against a corpus of real people.

Press and hold, image grids, sliders, rotation, the spoken alternative. Solved end to end on the live site, not in a lab.

Every tool call becomes one step with its result and a screenshot. A click that did nothing and a click that worked are two different lines.

See what it is used for

#03 - Use cases

Run the jobs that need a signed in browser.

The run gets to the login and stops there. Here it signs in, keeps the session, and goes on to the task you actually asked for.

login.live.com
Microsoft

Sign in

Use your Microsoft account.

Email or phone number
Next

Midjourney, your ad manager, half the tools your team pays for. If a person can click it, your agent can too.

Twenty accounts, twenty browsers, nothing shared between two of them. One ban stays one ban.

The same eleven portals every morning, by hand. Describe it once and hand the whole morning to the agent.

See what others say

Runs on your machine

Nobody stands between you and your accounts.

Profiles, cookies, saved passwords, passkeys and every screenshot the agent takes are written to your disk and stay there. Nothing to upload, nothing to trust. A breach somewhere else cannot reach a file that never left your disk.

Hosted browser service

You log in
Their server runs the browser
Their disk keeps the cookie

their infrastructure

Two parties own it

Scalebrowser

You log in
Your machine runs the browser
Your disk keeps the cookie

C:\ProgramData\Scalebrowser

One party owns it

Agent integrations

Keep the agent you
already built.

I use it to run about thirty agents that each need their own logged in account. Since every agent needs a separate identity I wanted something where the profile stays the same between runs.

It keeps the session, the cookies and the passkeys per profile. I haven't had to restart the nightly run since June.

Ines Halvorsen

Ines Halvorsen

Automation lead

Kettle

Not taking our word for it?

Ask a model that has never met us, then read the free check on your own browser.