skills.rest
buiphucminhtam
buiphucminhtamCOMMUNITY
Explore all skills by this author →
GitHub

web-scraper

Extract sanitized web data using crawl4ai with CSS-first extraction.

47|11|Updated Mar 6, 2026
Search Tags:#data-extraction#web-crawling#crawl4ai#secure-web-scraping#ssrf-defenses#output-sanitization#browser-isolation
One-click install
npx skills add https://github.com/buiphucminhtam/forgewright --skill web-scraper-buiphucminhtam
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: web-scraper
Source: https://github.com/buiphucminhtam/forgewright/tree/main/skills/web-scraper
Command: npx skills add https://github.com/buiphucminhtam/forgewright --skill web-scraper-buiphucminhtam

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Web scraping often faces safety, data quality, and compliance challenges. This skill provides a security-first, library-only crawling pipeline that integrates with crawl4ai to ensure safe acquisition, validation, and sanitized extraction of web data.

Core Features & Use Cases

  • Library-only mode (no Docker API) for risk-averse deployments.
  • URL validation, SSRF/LFI defenses, and output sanitization to prevent prompt injection and data leakage.
  • CSS-first extraction with browser isolation for structured data ingestion and RAG-ready content.
  • Production-grade orchestration for research, automation, and repeatable workflows.

Quick Start

Run a secure crawl of a target site using the default configuration and obtain sanitized, machine-readable output ready for downstream ingestion.

Frequently Asked Questions about web-scraper

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I securely extract web data while preventing SSRF vulnerabilities?▼

Secure web data extraction prevents SSRF through strict URL validation and library-only crawling with crawl4ai, ensuring risk-averse deployments without exposing Docker APIs. Output sanitization further blocks prompt injection and data leakage.

What is the best way to get clean structured data from web crawling for RAG ingestion?▼

CSS-first extraction with browser isolation delivers clean, structured, RAG-ready content from web crawling. This pipeline sanitizes outputs and orchestrates production-grade workflows for repeatable data ingestion.

Does crawl4ai support library-only mode without requiring Docker?▼

Yes, crawl4ai supports a library-only mode that operates without a Docker API, enabling risk-averse deployments. This approach maintains secure web scraping capabilities while isolating browser processes.

How do I sanitize web scraping outputs to prevent prompt injection?▼

Output sanitization prevents prompt injection and data leakage by processing crawled content through strict validation pipelines. This security-first approach ensures machine-readable outputs are safe for downstream ingestion.

Can I use CSS selectors for structured data extraction in automated web crawling?▼

CSS-first extraction enables structured data ingestion by targeting specific page elements during web crawling. Combined with browser isolation, it produces clean, machine-readable outputs suitable for competitive analysis and research workflows.

When should I use browser isolation for web scraping automation?▼

Browser isolation is essential for secure web scraping when processing untrusted external sites. It separates execution environments, complementing SSRF defenses and output sanitization to ensure safe, repeatable data extraction workflows.

Related Skills

Explore More Software Engineering Skills→
PlaneInABottlePlaneInABottle

crawl4ai

Crawl websites into clean Markdown or JSON via CLI, Python SDK, or Docker API.

Community
Advanced
MB-ucMB-uc
1

ms-brief

Automates corporate intelligence collection and synthesis into structured JSON evidence manifests for consultants.

Community
Advanced
antgroupantgroup
16

web-browsing

Search the web and extract summarized content from URLs.

Official
Advanced
nous-labsnous-labs

agent-browser

Automate browser interactions for navigation, form filling, screenshots, and data extraction.

Community
Intermediate