Browser automation API with persistent sessions, headless browser cloud, and CDP WebSocket access. The remote browser for AI agents and automation workflows.
import browserfabric
async with browserfabric.browser(
persist=True
) as session:
await session.navigate("https://example.com")
await session.click("#login")
page = await session.snapshot()
# Session state persists across restartsBrowser automation as a service
From API key to running automation in under a minute.
Sign up and create an API key from the dashboard. One key is all you need.
Create a cloud browser session via the REST API or Python SDK. Full Chromium, headless or headed.
Navigate, click, type, screenshot, or connect directly via CDP WebSocket with Playwright.
A complete browser automation platform designed from the ground up for AI agents and developer workflows.
Every session exposes a Chrome DevTools Protocol WebSocket. Connect with Playwright, Puppeteer, or any CDP client.
Save cookies, localStorage, and auth state. Restore sessions days or weeks later with full login context preserved.
Share a live view URL. Humans solve CAPTCHAs or 2FA, then hand control back to the agent seamlessly.
AI-optimized page snapshots. Get interactive elements, content, and structure in a token-efficient format.
Simple REST API for any language. Native MCP transport for Claude, Cursor, and other AI tool hosts.
Per-call usage logging with duration, status, and service breakdowns. Built-in API for monitoring your consumption.
Each session returns a CDP WebSocket URL. Use it with the Playwright library you already know. No proprietary SDK required.
from playwright.async_api import async_playwright
pw = await async_playwright().start()
browser = await pw.chromium.connect_over_cdp(
"ws://api.browserfabric.com/cdp"
)
page = await browser.contexts[0].pages[0]
await page.goto("https://example.com")Start free. Scale when you need to.
For experimenting.
For side projects and MVPs.
For production workloads.
For teams and high-volume.
All plans include unlimited persistent contexts, live viewer, MCP, and REST API. 14-day full refund — no questions asked.
Install the skill, set your API key, and ask your agent to browse the web.
Claude Code
Install
npx skills add https://www.browserfabric.comRun
BROWSERFABRIC_API_KEY=bf_your_key claude --dangerously-skip-permissions \
-p "use browserfabric to list the top 5 news on hackernews"Codex (OpenAI)
Install
npx skills add https://www.browserfabric.comRun
BROWSERFABRIC_API_KEY=bf_your_key codex \
"use browserfabric to check what's on the front page of hackernews"Cursor
Install
npx skills add https://www.browserfabric.comRun
# Set BROWSERFABRIC_API_KEY in Cursor Settings → Environment Variables
# Then in Cursor chat:
use browserfabric to open example.com and tell me the page titleCline
Install
npx skills add https://www.browserfabric.comRun
# Set BROWSERFABRIC_API_KEY in your shell, then ask Cline:
use browserfabric to take a screenshot of hackernewsHermes Agent / OpenClaw
Install
npx skills add https://www.browserfabric.comRun
BROWSERFABRIC_API_KEY=bf_your_key hermes \
"use browserfabric to fill out the contact form on example.com"Any agent — raw curl
Example
export BROWSERFABRIC_API_KEY=bf_your_key
# Create a browser session
curl -s -X POST https://api.browserfabric.com/api/v1/services/browseruse/call \
-H "Authorization: Bearer $BROWSERFABRIC_API_KEY" \
-H "Content-Type: application/json" \
-d '{"tool_name":"create_session","arguments":{"headless":true}}'All agents need BROWSERFABRIC_API_KEY set in the environment. Get a free key →
Get your API key in 30 seconds. No credit card required.