Smart Scraper
by @yadanzheng68-cmyk
AI-powered web scraper with intelligent structure recognition. Extracts lists, articles, and tables from any website with automatic type detection.
clawhub install smart-scraperπ About This Skill
name: smart-scraper description: AI-powered web scraper with intelligent structure recognition. Extracts lists, articles, and tables from any website with automatic type detection. metadata: openclaw: emoji: "π·οΈ" requires: bins: [node, npm] install: - kind: node package: "playwright" bins: [npx]
Smart Scraper
Intelligent web scraping that understands page structure.
Features
Usage
# Extract product listings
smart-scraper --url "https://example.com/products" --type listExtract article content
smart-scraper --url "https://example.com/article" --type article --format markdownExtract table data
smart-scraper --url "https://example.com/data" --type table --format csv
Options
| Option | Description | Default |
|--------|-------------|---------|
| --url, -u | Target URL (required) | - |
| --type, -t | Extraction type: list, article, table, auto | auto |
| --format, -f | Output format: json, csv, markdown | json |
| --max, -m | Maximum items to extract | 100 |
| --scroll | Enable auto-scroll for lazy-loaded content | false |
Examples
Extract Hacker News
smart-scraper -u https://news.ycombinator.com -t list -m 10
Save article as Markdown
smart-scraper -u https://blog.example.com/post -t article -f markdown > article.md
Export table to CSV
smart-scraper -u https://example.com/prices -t table -f csv > prices.csv
π‘ Examples
Extract Hacker News
smart-scraper -u https://news.ycombinator.com -t list -m 10
Save article as Markdown
smart-scraper -u https://blog.example.com/post -t article -f markdown > article.md
Export table to CSV
smart-scraper -u https://example.com/prices -t table -f csv > prices.csv
βοΈ Configuration
| Option | Description | Default |
|--------|-------------|---------|
| --url, -u | Target URL (required) | - |
| --type, -t | Extraction type: list, article, table, auto | auto |
| --format, -f | Output format: json, csv, markdown | json |
| --max, -m | Maximum items to extract | 100 |
| --scroll | Enable auto-scroll for lazy-loaded content | false |