Scrapling

Scrapling is an adaptive Python web scraping library that handles static pages, JavaScript-rendered content, and anti-bot protected sites through a unified API.

File conventions

Unless the user specifies otherwise, write all scraping files to:

• Python scripts → .scratch/jarrettmeyer/scrapling/.py
• Output files (JSON, HTML, PDF, CSV, etc.) → .scratch/jarrettmeyer/scrapling/.

Ensure .scratch/ is listed in .gitignore to keep scraping artifacts out of version control.

Running scrapling

Always run scripts with uv run --with "scrapling[all]" — no installation, venv, or pyproject.toml required. Works in any project.

uv run --with "scrapling[all]" python .scratch/jarrettmeyer/scrapling/scrape.py

Step 1: Choose the right fetcher

Select based on the target site's characteristics:

| Site type | Fetcher | When to use | | --------------------- | ------------------- | ----------------------------------------- | | Static HTML | Fetcher | Fast, no JS needed, no bot protection | | Anti-bot / Cloudflare | StealthyFetcher | Bot detection, Cloudflare, fingerprinting | | JavaScript-rendered | DynamicFetcher | Content only visible after JS executes | | Multi-page with login | *Session variants | Cookie persistence across requests |

If unsure, start with Fetcher. If you get blocked or see empty content, escalate to StealthyFetcher, then DynamicFetcher.

Step 2: Fetch the page

Static HTML — Fetcher

from scrapling.fetchers import Fetcher

page = Fetcher.get('https://example.com', stealthy_headers=True)
print(page.status)  # 200

Anti-bot protected — StealthyFetcher

from scrapling.fetchers import StealthyFetcher

page = StealthyFetcher.fetch('https://example.com', headless=True)
print(page.status)

JavaScript-rendered — DynamicFetcher

from scrapling.fetchers import DynamicFetcher

page = DynamicFetcher.fetch(
    'https://example.com',
    wait_selector='.target-element',  # wait for element before parsing
    headless=True,
)
print(page.status)

Session (cookies / login flows)

from scrapling.fetchers import FetcherSession  # or StealthySession, DynamicSession

with FetcherSession(impersonate='chrome') as session:
    session.post('https://example.com/login', data={'user': '...', 'pass': '...'})
    page = session.get('https://example.com/dashboard')

Step 3: Extract data

All fetchers return the same Selector object — extraction syntax is identical regardless of which fetcher was used.

CSS selectors

# Single value
title = page.css('h1::text').get()

# All matching values
links = page.css('a::attr(href)').getall()

# Element with nested access
for item in page.css('.product'):
    name  = item.css('.name::text').get()
    price = item.css('.price::text').get()
    print(name, price)

XPath

title = page.xpath('//h1/text()').get()
rows  = page.xpath('//table/tr').getall()

Text search

# Find element containing specific text
el = page.find_by_text('Add to Cart', first_match=True)

Structured output

import json

results = []
for item in page.css('.result'):
    results.append({
        'title': item.css('h2::text').get(),
        'url':   item.css('a::attr(href)').get(),
        'desc':  item.css('p::text').get(),
    })

print(json.dumps(results, indent=2))

Step 4: Multi-page crawl (Spider)

Use Spider when scraping more than a handful of pages.

from scrapling.spiders import Spider, Response

class MySpider(Spider):
    name = "my_spider"
    start_urls = ["https://example.com/listings"]

    async def parse(self, response: Response):
        for item in response.css('.listing'):
            yield {
                'title': item.css('h2::text').get(),
                'price': item.css('.price::text').get(),
                'url':   item.css('a::attr(href)').get(),
            }

        # Follow pagination
        next_page = response.css('a.next::attr(href)').get()
        if next_page:
            yield response.follow(next_page, callback=self.parse)

result = MySpider().start()
result.items.to_json('.scratch/jarrettmeyer/scrapling/output.json')
print(f"Scraped {len(result.items)} items → .scratch/jarrettmeyer/scrapling/output.json")

Run with:

uv run --with "scrapling[all]" python scrape.py

Common options

# HTTP method
page = Fetcher.post('https://api.example.com/data', json={'key': 'value'})

# Custom headers / proxy
page = Fetcher.get(
    'https://example.com',
    headers={'Accept-Language': 'en-US'},
    proxy='http://user:pass@proxy:8080',
)

# DynamicFetcher: scroll before capturing
page = DynamicFetcher.fetch('https://example.com', scroll_down=True)

# DynamicFetcher: execute JS before capturing
page = DynamicFetcher.fetch(
    'https://example.com',
    wait_selector='.loaded',
    execute_js="window.scrollTo(0, document.body.scrollHeight)",
)

Troubleshooting

| Symptom | Fix | | --------------------------------- | ---------------------------------------------------------------------- | | Empty selectors / missing content | Switch to DynamicFetcher — page likely needs JS | | 403 / CAPTCHA / bot block | Switch to StealthyFetcher | | Inconsistent results across runs | Use find_by_text or similarity matching instead of brittle CSS paths | | Need to stay logged in | Use a *Session variant | | Large crawl is slow | Increase Spider concurrency settings |

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: jarrettmeyer
• Source: jarrettmeyer/skills
• License: MIT

Install and usage instructions live in the source repository linked above.