The browser for computer use agents.
Give your AI agent a real browser. Act, extract, observe, and automate any page with built-in stealth and smart retry.
import { SpiderBrowser } from "spider-browser" const browser = new SpiderBrowser({ apiKey: process.env.SPIDER_API_KEY!, mode: "cua", }) await browser.init() await browser.page.goto("https://example.com") // AI agent completes tasks autonomously const result = await browser.agent({ maxRounds: 20 }) .execute("Find the cheapest flight to Tokyo") await browser.close()
Key Features
Lightning fast
Commands dispatch in microseconds. The fastest browser engine for automation, delivering results before others even connect.
100x cheaper sessions
Efficient resource usage means fewer compute hours. Run the same workload at a fraction of what others charge.
Near zero flaky runs
Automatic retry engine with stealth rotation and smart proxy management. Sessions recover on their own.
How it works
Three steps from install to autonomous browsing. No infrastructure to manage.
Connect a browser
Install the SDK and call init(). Spider provisions a pre-warmed browser in the cloud and returns a live session in milliseconds.
Let the agent work
Use agent() to give your AI a task in plain English. It browses, clicks, fills forms, and extracts data autonomously. Stealth and retries are automatic.
Get structured results
Receive structured JSON, screenshots, or session recordings. Review every step the agent took, then close and pay only for what you used.
What you can build
From autonomous agents to high-volume scraping, one browser handles it all.
Computer Use Agents
Give your AI a real browser. The agent() API lets models browse, click, fill forms, and complete multi-step tasks autonomously.
Natural language actions
Use act() and observe() to control any page with plain English. No selectors, no fragile scripts.
Structured data extraction
Extract product listings, pricing, reviews, or any structured data using natural language prompts and typed schemas.
Protected site scraping
Scrape sites behind Cloudflare, DataDome, and PerimeterX with automatic stealth escalation and CAPTCHA solving.
Autonomous monitoring
Set an agent to monitor prices, inventory, or page changes. Smart retry ensures you always get the data, even from flaky pages.
RAG pipeline ingestion
Feed JavaScript-heavy pages into your RAG pipeline. Spider renders SPAs and returns clean content ready for embedding.
Your language, same API
Official SDKs for TypeScript, Python, and Rust. Same API surface across all three.
import { SpiderBrowser } from "spider-browser" const b = new SpiderBrowser({ apiKey }) await b.init() await b.page.goto("https://example.com") const data = await b.extract("Get prices")
from spider_browser import SpiderBrowser async with SpiderBrowser(api_key=key) as b: await b.page.goto("https://example.com") data = await b.extract("Get prices")
use spider_browser::SpiderBrowser; let b = SpiderBrowser::new(api_key) .init().await?; b.goto("https://example.com").await?; let data = b.extract("Get prices").await?;
Why Spider Browser
Skip the infrastructure. Ship the agent.
| Feature | Spider | Self-hosted | Other clouds |
|---|---|---|---|
| Stealth & anti-bot | Built-in | DIY | Partial |
| CAPTCHA solving | Automatic | Manual | Add-on |
| Smart retry | Auto | Custom | Limited |
| Multi-browser | Multiple | 1 | 1-2 |
| AI extraction | Native | None | None |
| Command latency | ~4ms | ~50ms | ~20ms |
| Cold start | 0ms | 3-10s | 1-5s |
| Session recording | Built-in | Custom | Extra cost |
Frequently asked questions
What is a Computer Use Agent (CUA)?
A Computer Use Agent is an AI model that controls a real browser. It navigates pages, clicks buttons, fills forms, and extracts data just like a human would. Spider Browser provides the infrastructure for CUAs: pre-warmed cloud browsers with the act(), observe(), extract(), and agent() APIs that let any LLM browse the web autonomously.
How is Spider Browser different from Puppeteer or Playwright?
Puppeteer and Playwright require you to manage your own browser infrastructure: servers, Chrome versions, memory, scaling. Spider Browser gives you pre-warmed browsers in the cloud with built-in stealth, smart retry, CAPTCHA solving, and proxy rotation. It also adds CUA capabilities that traditional tools lack, like natural language actions, AI extraction, and autonomous multi-step agents.
What programming languages does Spider Browser support?
Spider Browser has official SDKs for TypeScript (npm install spider-browser), Python (pip install spider-browser), and Rust (cargo add spider-browser). All three share the same API surface: init, goto, act, observe, extract, and agent.
How do the act, observe, and extract APIs work?
These are natural language interfaces for browser control. act("Click the Sign Up button") performs actions. observe("Find all Add to Cart buttons") discovers interactive elements. extract("Get all product prices") returns structured JSON. The agent() method chains these into autonomous multi-step workflows.
How does stealth and anti-bot bypass work?
Spider Browser automatically handles anti-bot protections including Cloudflare, DataDome, and PerimeterX. Stealth levels escalate on failure and CAPTCHA solving is built-in. Just send your request and Spider takes care of the rest.
What is the pricing for Spider Browser?
Spider Browser pricing is simple: $1 per GB of bandwidth and $0.001 per minute of compute. No monthly minimum. You pay only for what you use, and get 500 free credits on sign-up with no credit card required. Stealth, proxy rotation, and CAPTCHA solving are all included.
Can I use Spider Browser for CUA workflows at scale?
Yes. Spider Browser supports up to 100 concurrent sessions with zero cold start. Each session runs on pre-warmed browser instances, so there is no boot time. Combined with smart retry and automatic recovery, Spider Browser is built for running Computer Use Agents at scale across even the most protected websites.
Start automating in minutes
500 free credits on sign-up. No credit card, no infrastructure, no cold starts.
