Content Type
Static (Markdown)
Unlocks Tools
These tools become available when this skill is activated:
execute_shell
Trigger Keywords
scrapewebextracthtmlparsecrawldata extractionwebsite
Required Packages
beautifulsoup4requestslxmlpandas
Instructions Preview
# Web Scraping Skill
Extract structured data from websites using Python. Use the helper script or write custom code.
Scraping runs in the sandbox: fetch and parse with `execute_shell`, using the helper script
below or your own Python when a page needs custom handling. Write results under
`/home/daytona/out/` so they sync back as artifacts.
## Using the Helper Script
```bash
# Scrape a single page and extract all links
python /home/daytona/skills/web-scraping/scraper.py https://example.com --links
# Extract text content
python /home/daytona/skills/web-scraping/scraper.py https://example.com --text
# Extract data using CSS selectors
python /home/daytona/skills/web-scraping/scraper.py https://example.com --selector "h1,h2,h3" --output /home/daytona/out/headings.json
# Extract table data to CSV
python /home/daytona/skills/web-scraping/scraper.py https://example.com --tables --output /home/daytona/out/tables.csv
# Full page analysis
python /home/daytona/skills/web-scraping/scraper....