clearpass

Crawl a website — or a published presentation deck — and save what you find as PNG, JPEG, PPTX, DOCX, PDF, or interactive MHTML archives. Includes a beautiful TUI (Terminal UI) for ease of use.

Install

npm install -g clearpass

Features

• Universal Auto-Detect: Simply paste a URL. The engine will automatically detect if it's a website or a presentation (Google Slides, Canva, Pitch, SlideShare, etc.) and engage the correct crawler automatically!
• TUI & CLI: Drop into an interactive shell by running clearpass, or use it strictly as a CLI tool.
• Universal Presentation Crawler: Captures generic presentations using smart visual hashing to detect slide changes.
• Google Slides Native Interactivity: Intercepts Google Slides and downloads perfect PDFs with 100% interactivity, falling back to the visual crawler automatically if the native export fails or takes longer than 4 seconds.
• Reconstructed Interactivity: The visual crawler automatically extracts hyperlink coordinates from presentation slides and injects them as clickable links into your final PDFs and PPTXs.
• Intelligent Naming System: Your output files will be given extremely short, readable, and 100% unique names (e.g., canva-4f8a.pdf), completely preventing accidental overwriting.
• Visual Stability Engine: It doesn't blindly screenshot. It compares image hashes to ensure animations and lazy-loaded elements are finished rendering.
• Smart Security Bypass: Detects Cloudflare and reCAPTCHA. New: It now actively waits for Cloudflare's "Just a moment..." challenges to automatically resolve instead of blindly snapshotting the security clearance screen!
• Safe Interrupts: Press Ctrl+Q at any time to safely abort a crawl without losing the progress you've already made!

Interactive Shell (TUI)

Run the tool with no arguments to enter the shell:

clearpass

Commands & Auto-Detect

If you simply type or paste a URL (e.g., https://example.com), the engine will auto-detect the type of content and capture it appropriately.

The explicit commands are still there so you can override the auto-detection if it guesses wrong, or if you want to perform a specific action:

• /snap <url> (Force Screenshots): Use this if the engine accidentally detects a website as a presentation and you want to force it to just take raw screenshots of the page.
• /scrape <url> (Extract Text): Auto-detect never scrapes text by default. If you want to extract the raw markdown text from a webpage instead of taking images, you must explicitly use this command. Note: Scraping strictly outputs to PDF, DOCX, or MHTML formats only. If another format is set, it will safely override to PDF.
• /slides <url> (Force Presentation): Use this if you have a presentation link from a niche platform that the engine doesn't recognize (not Canva, Pitch, etc.). This forces the engine to use the Presentation Crawler logic.
• /bulk: Enter bulk-URL mode (paste multiple URLs, hit enter on an empty line to finish).
• /format <fmt>: Set default output format (png, jpeg, pptx, docx, pdf, mhtml).
• /max <number>: Set max pages/slides (default 300).
• /out [dir]: Set output folder. Leave [dir] blank to open a native OS folder picker!
• ? or /help: Show all commands.
• q or /quit: Exit.

CLI Usage

clearpass -u https://example.com -o ./out -f pdf

CLI Options

• -u, --url <url>: The starting URL to crawl.
• -o, --out <dir>: The directory to save the output files.
• -f, --format <format>: Format for output (png, jpeg, pptx, docx, pdf, mhtml).
• -m, --max <number>: Max pages to crawl.
• -d, --depth <number>: Max link depth for website crawls.
• --scrape: Extract text instead of taking screenshots.
• --slides: Treat as a generic presentation (Canva, Pitch, etc.).
• --all-domains: Follow links to other domains (default: same-domain only).