Fetch web pages as clean markdown using Chrome TLS fingerprinting with fallback strategies.
npx skills add https://github.com/wickproject/wick --skill wick
Please help me install this Agent Skill. Skill: wick Source: https://github.com/wickproject/wick/tree/main/agent-skill/wick Command: npx skills add https://github.com/wickproject/wick --skill wick
Wick solves the problem of AI agents being blocked by anti-bot defenses when trying to fetch webpages, including TLS fingerprinting differences and challenge pages that return unusable content.
Use Case: A researcher needs primary-source text from a complex, JS-heavy website that blocks standard HTTP clients; they can fetch and crawl it through Wick to obtain consistent markdown suitable for summarization and Q&A.
Use the command mcp__wick__wick_wick_fetch(url="https://example.com").
High-intent search queries and answers about installing and using this skill.
Fetching web pages blocked by anti-bot systems requires browser-grade retrieval using Chrome TLS fingerprinting via Cronet, with automated fallback to embedded Chromium or CAPTCHA strategies to bypass defenses and return clean markdown.
To crawl JavaScript-heavy websites and get LLM-friendly markdown, use browser-grade fetching with Cronet and Chromium fallbacks to render dynamic content, returning consistent markdown suitable for summarization and Q&A.
Yes, an MCP tool interface supports session-based authenticated fetching, maintaining login state across requests to reliably retrieve content from protected sites requiring authentication.
The best way to extract primary-source text from sites blocking standard HTTP clients is browser-grade fetching with Chrome TLS fingerprinting and automated CAPTCHA fallbacks, returning clean markdown for downstream analysis.
Yes, web crawling with Chrome TLS fingerprinting supports media downloading for downstream analysis and URL mapping for discovery, alongside single-page reading and web searching across public and protected sites.
Standard web fetching fails on TLS fingerprinting and challenge pages because anti-bot defenses detect differences in network stacks, returning unusable challenge content instead of the target page.
allanninalExtracts YouTube subtitles and transcripts via yt-dlp or chrome-devtools-mcp.
Shallow-WExtract clean markdown content from standard web pages using Defuddle CLI.
s-hiraokuScrape websites, crawl pages, search, and extract structured data via Firecrawl tools.
HappyHackingSpaceAutomate browser-accurate HTTP requests with TLS fingerprint customization and HTTP/1.1/2/3 support.
