How do I install the Webclaw skill?

Run npx skillmds@latest add javimosch/webclaw in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Webclaw safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Webclaw?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Webclaw free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Webclaw?

javimosch (@javimosch) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Webclaw by javimosch — Agent Skill

Name: Webclaw
Author: javimosch

File contents

webclaw Plugin

Fast, local-first web content extraction for LLMs. webclaw scrapes single pages, crawls entire sites, and outputs clean structured data in markdown, JSON, text, or LLM-optimized formats. Zero browser overhead.

Commands

Extraction

webclaw page extract — Extract content from a single web page
webclaw site crawl — Crawl a website and extract from multiple pages
webclaw page diff — Compare a page against a previous snapshot
webclaw self version — Print webclaw version
webclaw _ _ — Passthrough to webclaw CLI

Usage Examples

"Extract the content from https://example.com as markdown"
"Crawl https://docs.example.com to depth 2"
"Scrape a page and output JSON"
"Extract brand information from a website"

Installation

brew tap 0xMassi/webclaw && brew install webclaw

Examples

# Extract a single page as markdown
webclaw page extract https://stripe.com -f markdown

# Extract main content only
webclaw page extract https://example.com --only-main-content

# Extract with specific CSS selectors
webclaw page extract https://example.com --include "article, .content" --exclude "nav, footer"

# Extract brand assets
webclaw page extract https://github.com --brand

# Crawl a docs site to depth 2, max 50 pages
webclaw site crawl https://docs.rust-lang.org --depth 2 --max-pages 50 -f markdown

# Crawl from sitemap
webclaw site crawl https://example.com --sitemap -f json

# Discover URLs only
webclaw site crawl https://example.com --map

# Extract as LLM-optimized format
webclaw page extract https://example.com -f llm

# Compare page against a snapshot
webclaw page extract https://example.com -f json > snap.json
webclaw page diff snap.json https://example.com

Key Features

Sub-millisecond extraction with zero browser overhead
5 output formats: markdown, text, JSON, LLM-optimized, HTML
CSS selector filtering (--include, --exclude)
Auto-detect main content with --only-main-content
Same-origin BFS crawling with depth and max-pages limits
Sitemap-based crawling
Brand extraction (colors, fonts, logos)
Change tracking with snapshot diffing
PDF and YouTube metadata extraction
Readability scoring and noise filtering

---name: webclawdescription: Use this skill when the user wants to scrape web pages, crawl websites, or extract structured content from URLs for LLM consumption.---# webclaw PluginFast, local-first web content extraction for LLMs. webclaw scrapes single pages, crawls entire sites, and outputs clean structured data in markdown, JSON, text, or LLM-optimized formats. Zero browser overhead.## Commands### Extraction- `webclaw page extract` — Extract content from a single web page- `webclaw site crawl` — Crawl a website and extract from multiple pages- `webclaw page diff` — Compare a page against a previous snapshot- `webclaw self version` — Print webclaw version- `webclaw _ _` — Passthrough to webclaw CLI## Usage Examples- "Extract the content from https://example.com as markdown"- "Crawl https://docs.example.com to depth 2"- "Scrape a page and output JSON"- "Extract brand information from a website"## Installation```bashbrew tap 0xMassi/webclaw && brew install webclaw```## Examples```bash# Extract a single page as markdownwebclaw page extract https://stripe.com -f markdown# Extract main content onlywebclaw page extract https://example.com --only-main-content# Extract with specific CSS selectorswebclaw page extract https://example.com --include "article, .content" --exclude "nav, footer"# Extract brand assetswebclaw page extract https://github.com --brand# Crawl a docs site to depth 2, max 50 pageswebclaw site crawl https://docs.rust-lang.org --depth 2 --max-pages 50 -f markdown# Crawl from sitemapwebclaw site crawl https://example.com --sitemap -f json# Discover URLs onlywebclaw site crawl https://example.com --map# Extract as LLM-optimized formatwebclaw page extract https://example.com -f llm# Compare page against a snapshotwebclaw page extract https://example.com -f json > snap.jsonwebclaw page diff snap.json https://example.com```## Key Features- Sub-millisecond extraction with zero browser overhead- 5 output formats: markdown, text, JSON, LLM-optimized, HTML- CSS selector filtering (--include, --exclude)- Auto-detect main content with --only-main-content- Same-origin BFS crawling with depth and max-pages limits- Sitemap-based crawling- Brand extraction (colors, fonts, logos)- Change tracking with snapshot diffing- PDF and YouTube metadata extraction- Readability scoring and noise filtering

Frequently asked questions

Webclaw