MCP HubMCP Hub
Volver a habilidades

web-scraper

agentbay-ai
Actualizado 2 days ago
2 vistas
40
2
40
Ver en GitHub
Documentacióngeneral

Acerca de

Esta habilidad extrae páginas web a HTML o Markdown con texto e imágenes utilizando dependencias mínimas (requests y beautifulsoup4). Está diseñada para descargar y archivar contenido localmente cuando se proporciona una URL. Las características clave incluyen el guardado automático de imágenes y la opción de seguir enlaces de forma recursiva.

Instalación rápida

Claude Code

Recomendado
Principal
npx skills add agentbay-ai/agentbay-skills -a claude-code
Comando PluginAlternativo
/plugin add https://github.com/agentbay-ai/agentbay-skills
Git CloneAlternativo
git clone https://github.com/agentbay-ai/agentbay-skills.git ~/.claude/skills/web-scraper

Copia y pega este comando en Claude Code para instalar esta habilidad

Documentación


name: web-scraper description: Scrape web pages and save as HTML or Markdown (with text and images). Minimal dependencies - only requests and beautifulsoup4. Use when the user provides a URL and wants to download/archive the content locally. homepage: https://requests.readthedocs.io/ metadata: { "openclaw": { "emoji": "🕷️", "requires": { "bins": ["python3"], "env": [] }, }, }

Web Scraper

Fetch web page content (text + images) and save as HTML or Markdown locally.

Minimal dependencies: Only requires requests and beautifulsoup4 - no browser automation.

Default behavior: Downloads images to local images/ directory automatically.

Quick start

Single page

{baseDir}/scripts/scrape.py --url "https://example.com" --format html --output /tmp/page.html
{baseDir}/scripts/scrape.py --url "https://example.com" --format md --output /tmp/page.md

Recursive (follow links)

{baseDir}/scripts/scrape.py --url "https://docs.example.com" --format md --recursive --max-depth 2 --output ~/Downloads/docs-archive

Setup

Requires Python 3.8+ and minimal dependencies:

cd {baseDir}
pip install -r requirements.txt

Or install manually:

pip install requests beautifulsoup4

Note: No browser or driver needed - uses pure HTTP requests.

Inputs to collect

Single page mode

  • URL: The web page to scrape (required)
  • Format: html or md (default: html)
  • Output path: Where to save the file (default: current directory with auto-generated name)
  • Images: Downloads images by default (use --no-download-images to disable)

Recursive mode (--recursive)

  • URL: Starting point for recursive scraping
  • Format: html or md
  • Output directory: Where to save all scraped pages
  • Max depth: How many levels deep to follow links (default: 2)
  • Max pages: Maximum total pages to scrape (default: 50)
  • Domain filter: Whether to stay within same domain (default: yes)
  • Images: Downloads images by default

Conversation Flow

  1. Ask user for the URL to scrape
  2. Ask preferred output format (HTML or Markdown)
    • Note: Both formats include text and images by default
    • HTML: Preserves original structure with downloaded images
    • Markdown: Clean text format with downloaded images in images/ folder
  3. For recursive mode: Ask max depth and max pages (optional, has sensible defaults)
  4. Ask where to save (or suggest a default path like /tmp/ or ~/Downloads/)
  5. Run the script and confirm success
  6. Show the saved file/directory path

Examples

Single Page Scraping

Save as HTML

{baseDir}/scripts/scrape.py --url "https://docs.openclaw.ai/start/quickstart" --format html --output ~/Downloads/openclaw-quickstart.html

Save as Markdown (with images, default)

{baseDir}/scripts/scrape.py --url "https://en.wikipedia.org/wiki/Web_scraping" --format md --output ~/Documents/web-scraping.md

Result: Creates web-scraping.md + images/ folder with all downloaded images (text + images).

Without downloading images (optional)

{baseDir}/scripts/scrape.py --url "https://example.com" --format md --no-download-images

Result: Only text + image URLs (not downloaded locally).

Auto-generate filename

{baseDir}/scripts/scrape.py --url "https://example.com" --format html
# Saves to: example-com-{timestamp}.html

Recursive Scraping

Basic recursive crawl (depth 2, same domain, with images)

{baseDir}/scripts/scrape.py --url "https://docs.example.com" --format md --recursive --output ~/Downloads/docs-archive

Output structure (text + images for all pages):

docs-archive/
├── index.md
├── getting-started.md
├── api/
│   ├── authentication.md
│   └── endpoints.md
└── images/              # Shared images from all pages
    ├── logo.png
    └── diagram.svg

Deep crawl with custom limits

{baseDir}/scripts/scrape.py \
  --url "https://blog.example.com" \
  --format html \
  --recursive \
  --max-depth 3 \
  --max-pages 100 \
  --output ~/Archives/blog-backup

Ignore robots.txt (use with caution)

{baseDir}/scripts/scrape.py \
  --url "https://example.com" \
  --format md \
  --recursive \
  --no-respect-robots \
  --rate-limit 1.0

Faster scraping (reduced rate limit)

{baseDir}/scripts/scrape.py \
  --url "https://yoursite.com" \
  --format md \
  --recursive \
  --rate-limit 0.2

Features

Single Page Mode

  • HTML output: Preserves original page structure
    • ✅ Clean, readable HTML document
    • ✅ All images downloaded to images/ folder
    • ✅ Suitable for offline viewing
  • Markdown output: Extracts clean text content
    • Auto-downloads images to local images/ directory (default)
    • ✅ Converts image URLs to relative paths
    • ✅ Clean, readable format for archiving
    • ✅ Fallback to original URLs if download fails
    • Use --no-download-images flag to keep original URLs only
  • Simple and fast: Pure HTTP requests, no browser needed
  • Auto filename: Generates safe filename from URL if not specified

Recursive Mode (--recursive)

  • ✅ Intelligent link discovery: Automatically follows all links on crawled pages
  • ✅ Depth control: --max-depth limits how many levels deep to crawl (default: 2)
  • ✅ Page limit: --max-pages caps total pages to prevent runaway crawls (default: 50)
  • ✅ Domain filtering: --same-domain keeps crawl within starting domain (default: on)
  • ✅ robots.txt compliance: Respects site's crawling rules by default
  • ✅ Rate limiting: --rate-limit adds delay between requests (default: 0.5s)
  • ✅ Smart URL filtering: Skips images, scripts, CSS, and duplicate URLs
  • ✅ Progress tracking: Real-time console output with success/fail/skip counts
  • ✅ Organized output: Preserves URL structure in directory hierarchy
  • ✅ Efficient crawling: Sequential with rate limiting to respect servers

Guardrails

Single Page Mode

  • Respect robots.txt and site terms of service
  • Some sites may block automated access; this tool uses standard HTTP requests
  • Large pages with many images may take time to download

Recursive Mode

  • Start small: Test with --max-depth 1 --max-pages 10 first
  • Respect robots.txt: Default is on; only use --no-respect-robots for your own sites
  • Rate limiting: Default 0.5s is polite; don't go below 0.2s for public sites
  • Same domain: Strongly recommended to keep --same-domain enabled
  • Monitor progress: Watch for high fail rates (may indicate blocking)
  • Storage: Recursive crawls can generate many files; ensure sufficient disk space
  • Legal: Ensure you have permission to crawl and archive the target site

Troubleshooting

  • Connection errors: Check your internet connection and URL validity
  • 403/blocked: Some sites block scrapers; the tool uses realistic User-Agent headers
  • Timeout: Increase --timeout flag for slow-loading pages (value in seconds)
  • Image download fails: Images will fall back to original URLs
  • Missing images: Some sites use JavaScript to load images dynamically (not supported)

Repositorio GitHub

agentbay-ai/agentbay-skills
Ruta: web-scraper
0

Habilidades relacionadas

railway-docs

Documentación

Esta habilidad obtiene la documentación actual de Railway para responder preguntas sobre características, funcionalidad o URLs específicas de documentación. Garantiza que los desarrolladores reciban información precisa y actualizada directamente de las fuentes oficiales de Railway. Úsala cuando los usuarios pregunten cómo funciona Railway o hagan referencia a la documentación de Railway.

Ver habilidad

n8n-code-python

Documentación

Esta Skill de Claude proporciona orientación experta para escribir código Python en los nodos Code de n8n, específicamente para usar la biblioteca estándar de Python y trabajar con la sintaxis especial de n8n como `_input`, `_json` y `_node`. Ayuda a los desarrolladores a comprender las limitaciones de Python dentro de n8n y recomienda usar JavaScript para la mayoría de los flujos de trabajo, mientras ofrece soluciones en Python para necesidades específicas de transformación de datos.

Ver habilidad

archon

Documentación

La habilidad Archon proporciona búsqueda semántica con tecnología RAG y gestión de proyectos a través de una API REST. Úsala para consultar documentación, gestionar proyectos/tareas jerárquicos y realizar recuperación de conocimiento con capacidades de carga de documentos. Prioriza siempre a Archon en primer lugar al buscar en documentación externa antes de utilizar otras fuentes.

Ver habilidad

n8n-code-javascript

Documentación

Esta habilidad de Claude proporciona orientación experta para escribir código JavaScript en los nodos de Código de n8n. Cubre sintaxis esencial específica de n8n como las variables `$input`/`$json`, ayudantes HTTP y manejo de DateTime, mientras soluciona errores comunes. Úsela al desarrollar flujos de trabajo en n8n que requieran procesamiento personalizado de JavaScript en los nodos de Código.

Ver habilidad