ego (lite) 只是一個瀏覽器，ego 則是你跨裝置的個人 Agent。
加入候補名單
下載 Mac 版
解決方案
文件企業版更新日誌Blog
ego lite 對比 Browserbase

最好用的 Browserbase 替代品

Browserbase is a cloud browser platform: your code spins up remote sessions over REST, drives them over CDP, and pays by the browser minute.

ego lite is a Browserbase alternative for your own agent work: real Chromium on your Mac, already signed into your accounts, controlled by your coding agent via /ego-browser. No session to create, no meter, finishing whole tasks 3 to 4x faster than Browserbase.

免費試用 ego (lite)GitHub
獲得來自以下單位開發者的信賴
GoogleAmazonShopifyTikTokHarvardStanfordUSCUCLA

為什麼 ego lite 比 Browserbase 更好

A Browserbase session is a fresh remote browser on someone else's servers. ego lite starts from your imported Chrome profile on your machine. Run /ego-browser in Claude Code, Codex, or Cursor.

在實際任務中快 3 到 4 倍

Before your agent touches a page, you wait for a cloud session to spin up, and the meter starts immediately. Browser time bills by the minute with a one-minute minimum, so a twenty-second check still costs a full minute.

Run /ego-browser and your coding agent controls ego lite on your machine. Every page arrives as a Snapshot: a compressed accessibility tree with stable @N refs it can act on instantly, finishing whole tasks 3 to 4x faster than Browserbase.

Time to finish a task, shorter is better
81.8 sego lite
282.9 sagent browser
Data source - Task: scrape the main posts from the X account from the last 7 days

平行多工，執行更快

How many Browserbase sessions you can run at once depends on your plan: 3 on the free tier, 25 on Developer, 100 on Startup. Go past that and work waits in a queue. To watch a run, you open Live View in a dashboard tab.

In ego lite there's no cap on how many Spaces you open inside the same browser: each one runs its own task at the same time, controlled by a coding agent via /ego-browser on its own imported Chrome profile. None of them ever steals the tab you're using, and you can watch or take over any Space anytime.

不再來回往返 消耗更少 tokens

Drive Browserbase with Stagehand and a simple checkout flow turns into a chain of model calls, each one waiting on a page read over the network before the next click.

Run /ego-browser and your coding agent batches several actions in a few lines of JavaScript per turn to control ego lite. Stacked with Snapshot input, the same task completes on far fewer tokens overall.

Claude Code · ego lite
>/ego-browser Add the first product to the cart, then check checkout loads
●Bash(
ego-browser nodejs <<'EOF' await openOrReuseTab('localhost:3000') await click('.product-card') await click('.add-to-cart') await click('a[href="/checkout"]') cliLog('checkout loaded') EOF
)
⎿Output:
[t=4480ms] checkout loaded
Done. Both steps work.
─── 1 tool call · ↑ 1.1k tokens ───
Stagehand · env: "BROWSERBASE"
>$ npx tsx run_task.ts const stagehand = new Stagehand({ env: "BROWSERBASE" }) await stagehand.init()
await page.goto("http://localhost:3000")
// page = stagehand.context.pages()[0]
navigated: http://localhost:3000
live view: https://www.browserbase.com/sessions/4f1c…
↑ 1.6k tokens sent so far
await stagehand.act("click the first product card")
// LLM call #1 · model: gpt-5.6-sol · cacheStatus: "MISS"
→ { success: true, message: "Action completed successfully",
actions: [{ selector: "/html/body/div[2]/main/div[1]/a",
method: "click", arguments: [] }] }
↑ 3.8k tokens sent so far
await stagehand.act("click add to cart")
// LLM call #2 · model: gpt-5.6-sol · cacheStatus: "MISS"
→ { success: true, message: "Action completed successfully",
actions: [{ selector: "/html/body/div[2]/main/button[1]",
method: "click", arguments: [] }] }
↑ 5.9k tokens sent so far
await stagehand.observe("find the checkout link")
// LLM call #3 · model: gpt-5.6-sol
→ [{ selector: "/html/body/nav/a[href='/checkout']",
description: "Checkout link", method: "click", arguments: [] }]
↑ 8.0k tokens sent so far
await stagehand.act(action)
// no LLM: Action from observe()
→ { success: true, message: "Action completed successfully",
actions: [{ method: "click", arguments: [] }] }
navigated: /checkout
↑ 9.1k tokens sent so far
await stagehand.extract("is the payment form present?", z.object({ present: z.boolean() }))
// LLM call #4 · model: gpt-5.6-sol
→ { present: true }
↑ 11.4k tokens sent so far
await stagehand.extract("list the payment fields", z.object({ fields: z.array(z.string()) }))
// LLM call #5 · model: gpt-5.6-sol
→ { fields: ["card", "expiry", "cvc", "zip"] }
↑ 13.6k tokens sent so far
─── 7 model calls · ↑ 13.6k tokens ───
同樣的任務，同樣的模型。 左邊：ego lite 裡一次打包的 JavaScript 呼叫，總共 1.1k tokens。 右邊：Stagehand 驅動遠程 Browserbase 會話，7 次呼叫，token 漲到 13.6k。

同一個 Chrome，AI Agent 原生體驗

Browserbase sessions start with a fresh profile unless you turn on persistence, and Browserbase's own docs say two-step verification and magic links usually need a human in the loop.

Built on Chromium, ego lite imports your entire Chrome setup in one click. When your coding agent runs /ego-browser, it controls ego lite with your real sessions already in place, so it doesn't stall on SSO, 2FA, or captchas on accounts you already have.

ego lite Chrome 個人設定檔匯入：一鍵設定，帶入你所有的登入狀態

ego lite 對比 Browserbase

ego lite 與 Browserbase 的功能比較。
功能ego liteBrowserbase
瀏覽器跑在哪裡在你的 Mac 本機，作為日常瀏覽器Browserbase cloud only; no self-hosted option
已登入的網站（SSO、2FA）One-click Chrome profile import; coding agent controls ego lite with real sessions via /ego-browserFresh user data by default; docs say 2FA and magic links need a human in the loop
AI Agent 如何連接在 Claude Code、Codex 或 Cursor 裡運行 /ego-browser。沒有 SDK，沒有 API key先用 REST 創建會話，再經 CDP 用 Stagehand、Playwright、Puppeteer 或 Selenium 接入
每項任務的 Token 成本Lower: Snapshot input plus batched JavaScript, measured per taskStagehand: a model call per act/observe/extract; plain Playwright/Puppeteer/Selenium scripts: none
價格免費，無訂閱，無帳號Free tier, then $20/mo Developer / $99/mo Startup / custom Scale, plus overage
什麼被計量什麼都不用。瀏覽器已經在你機器上Browser minutes and proxy MB, with one-minute and one-MB minima per session
平行任務Unlimited Spaces in one browser, bounded by your machinePlan caps: 3 free, 25 Developer, 100 Startup, 250+ Scale
流量帶著誰的信譽Your network and your logged-in browserShared cloud IP ranges
你的資料去哪Stays on your machinePages transit Browserbase; recording is on by default unless you opt out
可重複使用的 skills（即將推出）將成功的執行過程萃取成可重複使用的 skills；隨著 AI Agent 重複執行，複雜任務最多可加快 5 倍（限量 beta）沒有內建的對應功能
生產級瀏覽器集群與合規不能。只是一位開發者的機器和帳號Yes: SOC 2 Type II, HIPAA BAA, DPA, SAML SSO, four data regions
最後更新 2026年7月30日

讓轉換無縫接軌

Keep Browserbase for your product's cloud fleet. Move your own logged-in browser work to ego lite in a few minutes.

  1. 下載 ego (lite)

    下載 ego lite，一鍵匯入你的 Chrome 設定檔，登入狀態、Cookie 和擴充功能全部一併帶入。

  2. 用 /ego-browser 執行你的第一個任務

    貼到你的 Agent 裡

    /ego-browser 前往 ego.app 並加入候補名單

    Paste one sentence into Claude Code, Codex, or Cursor and hit enter. No session to create, no API key, no SDK.

  3. 看它實際運作
    ego lite 的 Space 總覽，四項瀏覽器任務並排執行：Claude Code 在 Yahoo Finance 追蹤蘋果股價、Codex 在 cars.com 依年份篩選車款、Hermes 完成一項 SaaS 後台任務、一位使用者在擷取 X 上的資料，還有一隻手正點擊 + 開啟另一個 Space

    The task runs in its own Space, so your tabs stay untouched. Watch live or take over anytime. Results land back in your agent's CLI.

Keep Browserbase for high concurrency, unattended runs, and anonymous targets that need proxies and captcha solving.

何時該用哪個工具

適合選擇 ego (lite) 的情況

  • Your tasks touch logged-in sites: dashboards, admin panels, anything behind SSO or 2FA.
  • You want a free local loop with no session create step and no browser minutes to bill.
  • Pages and credentials should stay on your machine, not transit a cloud provider.
  • You want parallel Spaces you can watch or take over from Claude Code, Codex, or Cursor.

選擇 Browserbase 的情境

  • You need real cloud concurrency (100+ browsers). ego lite is one machine.
  • Work must keep running when your laptop is closed.
  • You're hitting anonymous targets at volume and want managed proxies, stealth, and captcha solving.
  • You need SOC 2, HIPAA, SAML, and session recording because your product runs browsers for customers.

給你的 AI Agent 一個真實瀏覽器

免費，在你的 Mac 上執行，一鍵匯入你的 Chrome 個人設定檔。適用於 Claude Code、Codex、Cursor，以及任何會寫程式碼的 CLI AI Agent。

立即試用 ego (lite)

繼續閱讀

還在權衡？看看 Browserbase 和同賽道其他工具怎麼比。

ego (lite) vs Puppeteerego (lite) vs Playwright MCPego (lite) vs Browser Useego (lite) vs Stagehandego (lite) vs agent-browserego (lite) vs Playwrightego (lite) vs Seleniumego (lite) vs Codex browser所有比較瀏覽器自動化解決方案

FAQ

Browserbase is a cloud browser platform with docs and SDKs on GitHub. You create sessions over REST, drive them over CDP with Playwright, Puppeteer, Selenium, or Stagehand, and pay for browser time plus proxy bandwidth. It handles the ops layer most teams would rather not build: proxies, stealth, captcha solving, session recording, and compliance. For products that need browsers at volume for their own users, it is the reference choice among browserbase competitors. It is cloud-only; there is no self-hosted option.

For your own logged-in browser work, yes: free, local, already signed in, driven via /ego-browser. For production cloud fleets that need 100+ concurrent browsers, managed proxies, captcha solving, and compliance paperwork, Browserbase is still the right category.

Browserbase bills browser minutes and proxy bandwidth: a free tier, then Developer at $20/mo, Startup at $99/mo, and custom Scale plans, plus overage. ego lite is free. The browser is already on your machine, so there is nothing to meter.

Not by default. Sessions start with fresh user data unless you persist a context and sign in by hand through Live View. Browserbase's docs say two-step verification and magic links usually need a human in the loop. ego lite imports your Chrome profile once locally, and every task starts signed in.

Browserbase publishes product docs and open SDKs on GitHub, including Stagehand. That repo is the usual landing point for browserbase github searches. ego lite is separate: a Chromium browser on your Mac, controlled through the MIT ego-browser shell, not a cloud session API.

They sit at different layers. Browser Use is a Python agent framework that decides what to do. Browserbase is the cloud browser infrastructure those agents often run on. ego lite is neither: run /ego-browser and your coding agent controls ego lite locally, with no session API and no per-minute billing.

operator.browserbase is Browserbase's operator-style surface for driving cloud browser sessions from an agent. ego lite is not a hosted operator. It is local Chromium your coding agent controls with /ego-browser, already signed into your accounts.

Among browserbase competitors, the usual names are cloud browser platforms like Browserless, Steel, and hosted Playwright or Puppeteer fleets, plus agent stacks that bundle their own remote browsers. ego lite is a different category: a Chromium browser on your Mac for your own agent work, not a fleet you operate for customers.

Yes, but not directly from /ego-browser. You create a Browserbase session over REST, attach over CDP, and drive it with Stagehand, Playwright, Puppeteer, or Selenium from a script your agent runs. That path fits cloud automation. For logged-in work inside your own accounts, ego lite is simpler: run /ego-browser in Claude Code, Codex, or Cursor with no session create step.

Shared cloud IP ranges carry shared reputation, so some sites treat traffic as automated even when you mean well. Proxies and stealth help on anonymous targets. With ego lite, traffic comes from your own network and logged-in browser.

No. Browserbase's FAQ says there is no self-hosted option. ego lite runs entirely on your Mac. Free on macOS 12+, no account required.