> ## Documentation Index
> Fetch the complete documentation index at: https://browseruse-0aece648-agency-v4-agentmail-2fa-docs.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.
# Browser Infrastructure quickstart
> Launch a cloud browser and connect to it from your code.
Every browser includes stealth, proxies, live preview, and recording. Its
**CDP URL** is a WebSocket endpoint for remotely controlling Chrome.
Get an [API key](https://cloud.browser-use.com/settings?tab=api-keys\&new=1) and
export it:
```bash theme={null}
export BROWSER_USE_API_KEY=your_key
```
## Install the SDK
Skip this step if you use curl.
```bash Python theme={null}
pip install --upgrade browser-use-sdk
```
```bash TypeScript theme={null}
npm install browser-use-sdk@latest
```
## Launch a browser
The browser-management resource currently uses the explicit V3 SDK namespace.
The equivalent REST endpoint is API V4.
```python Python theme={null}
from browser_use_sdk.v3 import BrowserUse
client = BrowserUse()
browser = client.browsers.create(proxy_country_code="us")
print(browser.cdp_url)
# When finished:
client.browsers.stop(browser.id)
```
```typescript TypeScript theme={null}
import { BrowserUse } from "browser-use-sdk/v3";
const client = new BrowserUse();
const browser = await client.browsers.create({ proxyCountryCode: "us" });
console.log(browser.cdpUrl);
// When finished:
await client.browsers.stop(browser.id);
```
```bash curl theme={null}
browser=$(curl -sS https://api.browser-use.com/api/v4/browsers \
-H "X-Browser-Use-API-Key: $BROWSER_USE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"proxyCountryCode":"us"}')
export BROWSER_SESSION_ID=$(echo "$browser" | jq -r .id)
export BROWSER_USE_CDP_URL=$(echo "$browser" | jq -r .cdpUrl)
echo "$BROWSER_USE_CDP_URL"
# When finished:
curl -X PATCH \
"https://api.browser-use.com/api/v4/browsers/$BROWSER_SESSION_ID" \
-H "X-Browser-Use-API-Key: $BROWSER_USE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"action":"stop"}'
```
`browser.close()`, disconnecting CDP, or `client.close()` does not stop the
managed browser. Use `client.browsers.stop(browser.id)` or call
`PATCH /api/v4/browsers/{id}` with `{"action":"stop"}`.
## Connect over CDP
Point Playwright or Puppeteer at the CDP URL. Nothing else changes in your
code. The snippets read `BROWSER_USE_CDP_URL`; if you launched with the SDK
above, pass `browser.cdp_url` / `browser.cdpUrl` instead of the variable.
```python Python theme={null}
import os
from playwright.sync_api import sync_playwright
with sync_playwright() as p:
browser = p.chromium.connect_over_cdp(os.environ["BROWSER_USE_CDP_URL"])
page = browser.contexts[0].pages[0]
page.goto("https://example.com")
print(page.title())
```
```typescript TypeScript theme={null}
import { chromium } from "playwright";
const browser = await chromium.connectOverCDP(
process.env.BROWSER_USE_CDP_URL!,
);
const page = browser.contexts()[0].pages()[0];
await page.goto("https://example.com");
console.log(await page.title());
```
```typescript Puppeteer theme={null}
import puppeteer from "puppeteer-core";
const browser = await puppeteer.connect({
browserWSEndpoint: process.env.BROWSER_USE_CDP_URL!,
});
const [page] = await browser.pages();
await page.goto("https://example.com");
console.log(await page.title());
```
Selenium cannot connect to a remote WebSocket CDP URL; use Playwright or
Puppeteer. Stop the browser when you are done (see the warning above), and
the session ends with a [live preview](/cloud/browser/live-preview) and
recording you can open from the dashboard.
Full connection guide and stop semantics.
Configure proxies, screen size, recording, and timeout.