webpeel 0.13.0

Fast web fetcher for AI agents — smart extraction, stealth mode, structured data

WebPeel Python SDK

Fast web fetcher for AI agents — smart extraction, stealth mode, structured data.

Zero dependencies. Pure Python 3.8+ stdlib.

Installation

pip install webpeel

Quick Start

Basic Scraping

from webpeel import WebPeel

client = WebPeel()

# Scrape a URL and get clean markdown
result = client.scrape("https://example.com")
print(result.title)
print(result.content)  # Clean markdown content
print(result.metadata)  # Structured metadata

Search the Web

# Search via DuckDuckGo
results = client.search("python web scraping")

for item in results.data.get("web", []):
    print(f"{item['title']}: {item['url']}")

JavaScript-Heavy Sites

# Use browser rendering for SPAs and JS-heavy sites
result = client.scrape(
    "https://twitter.com/elonmusk",
    render=True,  # Enable browser mode
    wait=2000,    # Wait 2s for JS to load
)

Stealth Mode (Bypass Bot Detection)

# Bypass Cloudflare, reCAPTCHA, and anti-bot systems
result = client.scrape(
    "https://protected-site.com",
    stealth=True,  # Enable stealth mode
)

Structured Data Extraction

# Extract specific data using CSS selectors
result = client.scrape(
    "https://amazon.com/product/...",
    extract={
        "selectors": {
            "title": "h1#title",
            "price": "span.price",
            "rating": ".review-rating",
        }
    }
)

print(result.extracted)
# {"title": "Product Name", "price": "$29.99", "rating": "4.5"}

Crawl a Website

# Start an async crawl job (requires API key)
client = WebPeel(api_key="your-api-key")

job = client.crawl(
    "https://docs.example.com",
    limit=100,
    max_depth=3,
)

print(job.id)  # Job ID for tracking

# Check status later
status = client.get_job(job.id)
print(status["status"])  # pending, running, completed, failed

Map a Domain

# Discover all URLs on a domain
result = client.map("https://example.com")

print(f"Found {result.total} URLs")
for url in result.urls[:10]:
    print(url)

Batch Scraping

# Scrape multiple URLs in batch (requires API key)
client = WebPeel(api_key="your-api-key")

urls = [
    "https://example.com/page1",
    "https://example.com/page2",
    "https://example.com/page3",
]

job = client.batch_scrape(urls, max_tokens=5000)
print(job.id)

API Reference

WebPeel Class

WebPeel(
    api_key: Optional[str] = None,
    base_url: str = "https://api.webpeel.dev",
    timeout: int = 30,
)

• api_key: API key for authentication (optional for free tier)
• base_url: Base URL for the WebPeel API
• timeout: Request timeout in seconds

Methods

scrape(url, **options) -> ScrapeResult

Scrape a URL and extract content.

Options:

• formats: Output formats (default: ["markdown"])
• max_tokens: Maximum token count for output
• render: Use headless browser (default: False)
• stealth: Bypass bot detection (default: False)
• wait: Wait time in ms after page load
• extract: Structured data extraction config
• headers: Custom HTTP headers

search(query, limit=5) -> SearchResult

Search the web via DuckDuckGo.

crawl(url, limit=50, max_depth=3) -> CrawlResult

Start an async crawl job (requires API key).

map(url) -> MapResult

Discover all URLs on a domain.

batch_scrape(urls, **options) -> BatchResult

Batch scrape multiple URLs (requires API key).

get_job(job_id) -> Dict

Check status of an async job.

WebPeel vs Firecrawl

Feature	WebPeel	Firecrawl
Pricing	$0 local / $9-$29 cloud	$16-$333/mo
Free Tier	125 fetches/week	500 credits one-time
License	AGPL-3.0	AGPL-3.0
Python SDK Deps	Zero (pure stdlib)	httpx, pydantic
Smart Escalation	✅ Auto HTTP→Browser→Stealth	Manual mode selection
Token Budget	✅ --max-tokens	❌
Quality Scoring	✅ 0-1 per response	❌
Local CLI	✅ Free, unlimited	Requires API key
LangChain	✅	✅
LlamaIndex	✅	✅

WebPeel is the free, fast, open-source alternative to Firecrawl.

Authentication

Free tier: No API key needed. Anonymous usage with rate limits.

Paid tier: Get an API key at webpeel.dev.

client = WebPeel(api_key="wp_...")

Error Handling

from webpeel import WebPeel, WebPeelError, RateLimitError, TimeoutError

client = WebPeel()

try:
    result = client.scrape("https://example.com")
except RateLimitError:
    print("Rate limit exceeded. Upgrade or wait.")
except TimeoutError:
    print("Request timeout. Try again.")
except WebPeelError as e:
    print(f"Error: {e}")

License

AGPL-3.0 © Jake Liu

Links

• Homepage
• Documentation
• GitHub
• Issues