pangolinfo-amazon-scraper

Official Python SDK for the Pangolinfo Scrape API — structured e-commerce data from Amazon, Walmart, eBay, Shopify and more, delivered as JSON, clean HTML or LLM-friendly Markdown.

Built and maintained by PANGOLIN INFO TECH PTE. LTD.

Why this SDK

The Pangolinfo Scrape API dynamically adapts to page-structure changes so you never maintain fragile selectors. You get:

• Structured JSON for product details, keyword search, rankings, reviews and more
• Raw HTML and Markdown output for AI/LLM pipelines
• 99.9% uptime with built-in anti-bot handling and residential IP rotation
• Sync + async interfaces from a single package
• Coverage of Amazon, Walmart, eBay, Shopify, Shopee

Installation

pip install pangolinfo-amazon-scraper

Quick start

1. Get a free API key

Sign up at tool.pangolinfo.com — open-source contributors get 200 free API calls and full parser templates. See the free-tier details.

2. Use the SDK

from pangolinfo_amazon_scraper import PangolinfoClient

# Authenticate with a token from your dashboard...
client = PangolinfoClient(token="your_bearer_token")

# ...or with email/password (the SDK logs in for you)
# client = PangolinfoClient(email="you@example.com", password="secret")

# Fetch an Amazon product by ASIN
result = client.amazon.get_product("B0DYTF8L2W", zipcode="10041")
print(result.json_data)
# {"asin": "B0DYTF8L2W", "title": "...", "price": ..., "rating": ..., ...}

Async usage

import asyncio
from pangolinfo_amazon_scraper import AsyncPangolinfoClient

async def main():
    async with AsyncPangolinfoClient(token="your_bearer_token") as client:
        result = await client.amazon.get_product("B0DYTF8L2W")
        print(result.json_data)

asyncio.run(main())

Features

Multiple output formats

# Structured JSON (default)
result = client.scrape(
    url="https://www.amazon.com/dp/B0DYTF8L2W",
    parser_name="amzProductDetail",
    formats=["json"],
)

# LLM-friendly Markdown
md = client.scrape(url="...", parser_name="amzProductDetail", formats=["markdown"])
print(md.markdown)

# Raw HTML
html = client.scrape(url="...", parser_name="amzProductDetail", formats=["rawHtml"])
print(html.html)

# All three at once
result = client.scrape(url="...", parser_name="amzProductDetail", formats=["json", "rawHtml", "markdown"])

Platform helpers

# Amazon
client.amazon.get_product("B0DYTF8L2W", zipcode="10041")
client.amazon.search("wireless earbuds", domain="www.amazon.com")

# Walmart
client.walmart.get_product("123456789")
client.walmart.search("blender")

Generic scrape (any platform / parser)

from pangolinfo_amazon_scraper import PARSERS

# Use a known parser key
result = client.scrape(
    url="https://www.amazon.com/dp/B0DYTF8L2W",
    parser_name=PARSERS["amazon_product_detail"],
    formats=["json"],
    zipcode="10041",
)

# Or pass the raw parser name the API accepts
result = client.scrape(
    url="https://www.ebay.com/itm/...",
    parser_name="ebayProductDetail",
    formats=["json"],
)

Async callbacks for high-throughput jobs

result = client.scrape(
    url="https://www.amazon.com/dp/B0DYTF8L2W",
    parser_name="amzProductDetail",
    formats=["json"],
    callback_url="https://your-server.com/webhook",  # async mode
)

API reference

PangolinfoClient

Method	Description
PangolinfoClient(token=...)	Create a client with an existing bearer token
PangolinfoClient(email=..., password=...)	Create a client that logs in lazily
client.scrape(url, parser_name, formats, zipcode, ...)	Core scrape call
client.amazon.get_product(asin, zipcode)	Amazon product detail
client.amazon.search(keyword)	Amazon keyword search
client.walmart.get_product(product_id)	Walmart product detail
client.walmart.search(keyword)	Walmart keyword search
client.login()	Force (re)authentication
client.close()	Close the HTTP connection pool

AsyncPangolinfoClient

Async mirror of PangolinfoClient — every method is a coroutine and supports async with.

ScrapeResult

Attribute	Type	Description
result.json_data	dict / list / None	Parsed JSON payload
result.html	str / None	Raw HTML (when rawHtml requested)
result.markdown	str / None	Markdown (when markdown requested)
result.success	bool	True when code == 0
result.code	int	API response code
result.raw	dict	Full unmodified response

Exceptions

Exception	When
PangolinfoError	Base class for all SDK errors
AuthenticationError	Login failed or credentials missing
APIError	API returned non-zero code
TimeoutError	Request exceeded the timeout
ValidationError	Invalid arguments supplied

Supported parsers

Key	Parser name	Platform
amazon_product_detail	amzProductDetail	Amazon
amazon_keyword_search	amzKeywordSearch	Amazon
amazon_best_sellers	amzBestSellers	Amazon
amazon_reviews	amzReview	Amazon
walmart_product_detail	wmtProductDetail	Walmart
walmart_keyword_search	wmtKeywordSearch	Walmart

amzProductDetail is confirmed in the official example. Other parsers follow the Pangolinfo documentation naming — pass any raw parser name to client.scrape() if yours isn't listed.

Links

• Homepage: pangolinfo.com
• Documentation: docs.pangolinfo.com
• Get a free API key: tool.pangolinfo.com
• Source code: github.com/Pangolin-spg/pangolinfo-amazon-scraper
• Support: support@pangolinfo.com

Development

# Install with dev dependencies
pip install -e ".[dev]"

# Run tests
pytest

# Build distributions
python -m build

# Upload to PyPI
twine upload dist/*

License

MIT — © PANGOLIN INFO TECH PTE. LTD.