ego(lite) Features, Pricing, and Alternatives

What is ego(lite)?

ego(lite) is a Chromium browser that you and your AI agents use together. You browse in it the same way you browse in Chrome, and agents such as Claude Code, Codex, Cursor, Gemini CLI, and OpenCode use the same browser to do web tasks for you: research, filling in forms, collecting data from websites, and testing web apps. During setup you can import your Chrome profile, including tabs, bookmarks, passwords, extensions, cookies, and login sessions, so your agent starts already signed in. The app is made by Citro Labs, runs on your own computer, and is free to download and use. It is built for developers, analysts, and anyone who wants an agent to work on sites behind a login.

What sets ego(lite) apart?

Each agent task runs in its own Space, a separate set of tabs inside the same browser. The agent does not switch your tabs or take keyboard focus, so you keep working while several tasks run at once, and you can open any Space to watch, stop, or take over. Most browser automation tools start a blank browser with no logins, or run in the cloud where you have to hand over cookies. ego(lite) keeps your logins on your device and lets your own agent use them. In the company's Real-World Bench of 31 live-site tasks, it was 47% to 63% faster than four other browser CLIs.

How does ego(lite) work?

Installing ego(lite) also installs the ego-browser skill and adds it to the agent CLIs it finds on your computer. In your agent, you type /ego-browser and describe the task. The agent writes a short Node.js script that ego-browser runs in one call: open a page, read it, click, fill in a form, and report back. To read a page, the agent gets a Snapshot, a text list of page elements with short references such as @3 that usually fits in 200 to 400 tokens. Snapshots also reach iframes, shadow DOM, and embedded widgets such as Stripe checkout.

Who should use ego(lite)?

Software engineers and QA engineers can have an agent test a checkout or review a local dev server in a signed-in browser instead of maintaining login scripts. Sales reps and lead generation specialists can pull leads from tools like Sales Navigator. Marketers and analysts can collect posts and competitor data from sites behind a login. Finance teams can gather invoices from Stripe, Shopify, and bank portals at month-end, and legal researchers can search case-law sites like PACER and get cited filings back.

Is ego(lite) free?

Yes. ego(lite) is free to download and use, with no subscription, no per-task fee, and no charge for browser time. You do not need an account or an email address. You still pay for your own agent and model as you do today, and you can point it at a local model if you want page content to stay on your machine.

Which platforms does ego(lite) support?

ego(lite) runs on macOS, natively on both Intel and Apple Silicon Macs. A Windows version is available in beta, and Linux is on the roadmap with no release date yet. Any agent that can run shell commands can drive it, including Claude Code, Codex, Cursor, Continue, Gemini CLI, Hermes Agent, OpenClaw, OpenCode, and your own custom agent harness.

ego(lite) Use Cases

Scraping data from sites behind a login
Agent-driven end-to-end testing of web apps
Lead collection from sales platforms
Month-end invoice gathering from billing portals
Automated job search and applications

Features and Benefits

Agent Spaces
Each agent task runs in its own Space with separate tabs, so agents work in parallel without switching your tabs or taking keyboard focus.
Chrome Profile Import
Copy tabs, bookmarks, passwords, extensions, cookies, and login sessions from Chrome during setup so your agent starts already signed in.
Compact Page Snapshots
Agents read pages as a text list of elements with short references, usually 200 to 400 tokens, instead of raw HTML or CSS selectors.
Multi-Step Script Calls
The ego-browser CLI runs a whole Node.js script in one call, so opening, reading, clicking, and reporting take fewer model turns.
Iframe and Shadow DOM Access
Snapshots are built inside the modified Chromium engine, so agents can reach nested iframes, shadow DOM, and widgets like Stripe checkout.
Works With Any Agent
Setup wires ego-browser into Claude Code, Codex, Cursor, Gemini CLI, and other agent CLIs it finds, with no SDK to integrate.

ego(lite) Pros and Cons

Pros

Free to use with no subscription, per-task fee, or account
Agents start signed in using your imported Chrome logins
Agent tasks run in separate Spaces while you keep browsing
Works with Claude Code, Codex, Cursor, and other shell-capable agents
Logins and browser data stay on your own computer
Supports local models so page content can stay on your machine

Cons

Windows version is still in beta and Linux is not yet available
Tasks run on your computer, so they stop when it is off
You still pay for your own agent and model usage
Does not bypass CAPTCHAs, 2FA prompts, or other security checks
Saved reusable workflows (experience accumulation) are not yet released

ego(lite) — AI Browser