web-crawler

Fully recursive web crawler powered by Firecrawl open-source API. Downloads everything from a URL into typed local subfolders: HTML/markdown, images, PDFs, videos, audio, documents. Supports Firecrawl cloud (api.firecrawl.dev) and self-hosted instances on Coolify/Docker. Handles JS-heavy sites natively. Saves YouTube/Vimeo embed URLs separately. Enforces file size limits. Produces a full crawl manifest. ALWAYS trigger this skill when a user says: "crawl this site", "download everything from", "scrape all assets", "mirror a website", "download all PDFs/images/videos from a URL", "web scrape", "archive this site", "download all files from", "recursive download", or any request to bulk-download or mirror content from a website.