Scrybe — web pages, transcribed

Scrybe converts any webpage into clean, structured Markdown — built specifically for LLM workflows, AI agents, and knowledge bases. Most tools give you broken  image links that LLMs can't read. Scrybe describes every image as structured text: alt text, inferred type (Chart, Diagram, Logo, Screenshot), dimensions, and source URL. LLMs can actually reason about the content. ────────────────────────────── WHAT SCRYBE GIVES YOU ────────────────────────────── ▸ Token counter — See the estimated token count before pasting into Claude, ChatGPT, or Gemini. Know if your content fits the context window. ▸ LLM-ready frontmatter — Auto-extracts title, author, published date, source URL, reading time, word count, and token count as structured metadata — pulled from JSON-LD and Open Graph. ▸ Image descriptions — Every image becomes readable text instead of a broken link. Scrybe infers the image type and captures all available metadata. ▸ Three output formats — Save as .md, save as .txt, or copy to clipboard. Works for any LLM or knowledge base workflow. ▸ Right-click capture — Right-click any page → "Scrybe this page" → clean .md file downloaded instantly. No popup needed. ▸ Smart noise removal — Strips navigation, ads, cookie banners, sidebars, and author bylines. Leaves only the article content. ▸ Auto table of contents — Generates a linked TOC when the page has 3 or more headings. ▸ Paywall detection — Warns you when content may be truncated so you don't paste incomplete data into your LLM session. ▸ Works offline — No accounts, no API keys, no servers. Everything runs inside your browser. ────────────────────────────── PERFECT FOR ────────────────────────────── • Feeding articles and docs into Claude, ChatGPT, Gemini, or any LLM • Building RAG pipelines and AI knowledge bases • Saving research to Obsidian or Notion • Archiving web content as clean, portable Markdown ────────────────────────────── PRIVACY ────────────────────────────── Scrybe collects zero data. No tracking, no analytics, no external requests. Your content never leaves your browser. Full privacy policy: https://xnomad-dev.github.io/Scrybe_page/privacy.html Support: https://xnomad-dev.github.io/Scrybe_page/support.html