Firecrawl Web Skill
Scrape, search, and crawl the web using Firecrawl.
Setup
- Get your API key from firecrawl.dev/app/api-keys
- Set the environment variable:
export FIRECRAWL_API_KEY=fc-your-key-here - Install the SDK:
pip3 install firecrawl
Usage
All commands use the bundled fc.py script in this skill's directory.
Get Page as Markdown
Fetch any URL and convert to clean markdown. Handles JavaScript-rendered content.
python3 fc.py markdown "https://example.com"
python3 fc.py markdown "https://example.com" --main-only # skip nav/footer
Take Screenshot
Capture a full-page screenshot of any URL.
python3 fc.py screenshot "https://example.com" -o screenshot.png
Extract Structured Data
Pull specific fields from a page using a JSON schema.
Schema example (schema.json):
{
"type": "object",
"properties": {
"title": { "type": "string" },
"price": { "type": "number" },
"features": { "type": "array", "items": { "type": "string" } }
}
}
python3 fc.py extract "https://example.com/product" --schema schema.json
python3 fc.py extract "https://example.com/product" --schema schema.json --prompt "Extract the main product details"
Web Search
Search the web and get content from results (may require paid tier).
python3 fc.py search "Python 3.13 new features" --limit 5
Crawl Documentation
Crawl an entire documentation site. Great for learning new frameworks.
python3 fc.py crawl "https://docs.example.com" --limit 30
python3 fc.py crawl "https://docs.example.com" --limit 50 --output ./docs
Note: Each page costs 1 credit. Set reasonable limits.
Map Site URLs
Discover all URLs on a website before deciding what to scrape.
python3 fc.py map "https://example.com" --limit 100
python3 fc.py map "https://example.com" --search "api"
Example Prompts
- "Scrape https://blog.example.com/post and summarize it"
- "Take a screenshot of stripe.com"
- "Extract the name, price, and features from this product page"
- "Crawl the Astro docs so you can help me build a site"
- "Map all the URLs on docs.stripe.com"
Pricing
Free tier includes 500 credits. 1 credit = 1 page/screenshot/search query.