pawgrab 0.0.3

Web scraping API — clean output from any URL

Pawgrab

Web scraping API. Returns clean Markdown, HTML, text, or structured JSON from any URL.

Features

• Single URL scraping with multiple output formats
• Async site crawling (BFS, depth/page limits, Redis job queue)
• Structured extraction via OpenAI, CSS selectors, XPath, or regex
• Auto JS detection - curl_cffi first, Playwright fallback for JS-heavy pages
• Anti-bot evasion - TLS fingerprint impersonation, stealth browser profiles
• Robots.txt compliance
• Per-domain rate limiting
• Proxy rotation with health checking
• Docker Compose deployment (API + worker + Redis)

Install

pip install pawgrab
patchright install chromium

Quickstart

# Start Redis (needed for /crawl)
docker run -d -p 6379:6379 redis:7-alpine

# Configure
cp .env.example .env
# Set PAWGRAB_OPENAI_API_KEY if you need /extract

# Run
pawgrab serve

Or with Docker:

cp .env.example .env
docker compose up

API

All endpoints under /v1.

POST /v1/scrape

curl -X POST http://localhost:8000/v1/scrape \
  -H 'Content-Type: application/json' \
  -d '{"url": "https://example.com"}'

Field	Type	Default	Description
url	string	required	URL to scrape
formats	array	["markdown"]	markdown, html, text, json
wait_for_js	bool/null	null	Force JS (true), skip (false), auto (null)
timeout	int	30000	Timeout in ms

POST /v1/crawl

Returns job ID (HTTP 202). Poll with GET /v1/crawl/{job_id}.

curl -X POST http://localhost:8000/v1/crawl \
  -H 'Content-Type: application/json' \
  -d '{"url": "https://example.com", "max_pages": 5}'

POST /v1/extract

Requires PAWGRAB_OPENAI_API_KEY.

curl -X POST http://localhost:8000/v1/extract \
  -H 'Content-Type: application/json' \
  -d '{"url": "https://example.com", "prompt": "Extract the main heading"}'

GET /health

curl http://localhost:8000/health

CLI

pawgrab scrape https://example.com
pawgrab scrape https://example.com --format text
pawgrab extract https://example.com --prompt "Extract the main heading"
pawgrab serve --port 8000 --reload

Configuration

All settings via env vars with PAWGRAB_ prefix. See .env.example for the full list.

License

DBaJ-GPL v69.420