skills.rest
eylulsenakumral
eylulsenakumralCOMMUNITY
Explore all skills by this author →
GitHub

websh

Navigate and query web pages with Unix-like shell commands over cached content.

Updated Jun 10, 2026
Search Tags:#caching#web scraping#content extraction#shell#crawling#web navigation#unix commands
One-click install
npx skills add https://github.com/eylulsenakumral/auto-company-clean --skill websh-eylulsenakumral
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: websh
Source: https://github.com/eylulsenakumral/auto-company-clean/tree/main/.claude/skills/websh
Command: npx skills add https://github.com/eylulsenakumral/auto-company-clean --skill websh-eylulsenakumral

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Browsing and extracting information from websites typically requires manual clicking, copying, and re-fetching pages. This Skill turns the web into a filesystem-like environment where URLs are paths and pages are cached locally, so you can query, filter, and navigate web content with familiar shell commands without repeated network requests. ## Core Features & Use Cases - Shell-style navigation: Use cd, ls, cat, grep, follow, and back to move between URLs and extract page content from a local cache. - Asynchronous background processing: Fetches, HTML-to-markdown extraction, and eager link prefetching run in background subagents so the prompt never blocks. - Search, diff, and monitoring: Search across all cached pages with locate, compare page versions with diff, and poll sites for changes with watch. - Use Case: A researcher opens Hacker News with cd https://news.ycombinator.com, filters stories with grep "AI", follows a discussion, extracts comments with cat .comment, and archives the session with tar. ## Quick Start Start the websh shell and navigate to https://news.ycombinator.com, then list the first five links on the page.

Frequently Asked Questions about websh

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

FAQPage Schema
How do I browse a website using shell commands?▼

Start websh and run cd followed by a URL, such as cd https://news.ycombinator.com. The page is fetched and cached, then commands like ls, grep, and cat operate on the cached content without refetching.

How do I extract specific content from a web page?▼

Use cat with a CSS selector, such as cat .title or cat article, to pull text from matching elements. You can pipe results through grep, head, sort, and other text-processing commands for filtering.

Does websh work with JavaScript-rendered websites?▼

JavaScript-heavy sites are a known limitation since fetching retrieves raw HTML without executing scripts. The skill fails gracefully with helpful messages, and click or scroll interactions are best-effort without a full browser.

Can I search across multiple cached pages at once?▼

Yes, the locate command searches all cached pages instantly by text or regex. Use locate -i for case-insensitive search or --urls and --titles to restrict matching to specific fields.

How do I monitor a web page for changes?▼

Use watch followed by a URL to poll the page at a default 60-second interval, or set a custom interval with -n. Add --notify for system notifications or -d to highlight differences between checks.

Why does a command show fetching instead of results?▼

Commands run asynchronously, so fetching and extraction happen in background subagents while the prompt stays responsive. Commands degrade gracefully using raw HTML until extraction completes, and ps or jobs shows background task status.

Related Skills

Explore More Software Engineering Skills→
TeamWiseFlowTeamWiseFlow
8.4k

rss-reader

Convert RSS/Atom feed URLs into markdown lists of recent entries.

Official
Intermediate
zengbaochengzengbaocheng

ocr-and-documents

Extract text from PDFs and scanned documents using pymupdf or marker-pdf OCR.

Community
Advanced
Yog-SothoYog-Sotho
4

data-scraper-agent

Schedule public data crawls with AI enrichment and storage to Notion, Sheets, or Supabase.

Community
Advanced
brixtonphambrixtonpham

cloudflare-browser-rendering

Automate headless browser tasks for screenshots, PDFs, and content extraction.

Community
Advanced