Web scraping in Python, one line

Async usage

Async-first design for modern Python applications.

import asyncio
from alterlab import AsyncAlterLabClient

async def main():
    client = AsyncAlterLabClient(api_key="your-api-key")

    # Scrape a URL — anti-bot bypass handled automatically
    result = await client.scrape(
        "https://example.com/product-page",
        formats=["markdown", "text"],
    )

    print(result.markdown)
    print(f"Credits used: {result.credits_used}")

asyncio.run(main())

Sync usage

Synchronous wrapper for scripts, notebooks, and frameworks that don't use async.

from alterlab import AlterLabClient

client = AlterLabClient(api_key="your-api-key")

# Scrape with bot-protected tier
result = client.scrape(
    "https://example.com",
    formats=["markdown"],
    tier=3,  # Stealth browser mode — bypasses Cloudflare JS challenges
)

print(result.markdown)

SDK vs raw HTTP requests

The SDK handles the plumbing so you don't have to.

import httpx, time

def scrape(url, api_key):
    for attempt in range(3):
        resp = httpx.post(
            "https://api.alterlab.io/v1/scrape",
            headers={"X-API-Key": api_key},
            json={"url": url},
            timeout=30,
        )
        if resp.status_code == 200:
            return resp.json()["markdown"]
        time.sleep(2 ** attempt)
    raise Exception("Failed")

from alterlab import AlterLabClient

client = AlterLabClient(api_key=api_key)

def scrape(url):
    # Retries, auth, tier selection,
    # error handling — all handled
    result = client.scrape(url)
    return result.markdown

What the SDK handles