403 Whisperer — Docs to Markdown for AI

About

Tired of pointing your AI tools at documentation they can't actually read? Cloudflare 403s, login walls, and JavaScript-heavy pages mean your assistant often grabs the wrong version, an outdated cache, or nothing at all — and then confidently gives you answers based on docs it never saw.

403 Whisperer fixes that. With one click it captures the docs page you're actually looking at, splits it into clean Markdown sections, and downloads the whole thing as a ZIP you can drop straight into your AI workflow. Think of it as a quick way to build your own RAG source from any docs — so your assistant works from the real, current docs instead of guessing.

THE PROBLEM IT SOLVES
• AI tools hit 403 / Cloudflare blocks and can't fetch the page
• They read a stale or wrong version from their training data or cache
• They can't find the docs for a niche or internal library at all
• You end up copy-pasting docs by hand to give the model context

HOW IT HELPS
You're already authenticated and looking at the right page in your browser — so the extension reads exactly what you see and hands you clean, current Markdown. Feed those files to ChatGPT, Claude, Cursor, a local model, or any RAG pipeline for accurate, grounded answers about the libraries and APIs you actually use.

WHAT YOU GET
• index.md — an overview with a linked table of contents and a one-line summary of each section
• One clean .md file per section (headings, code blocks, lists, tables, links, and images preserved)
• Optionally, the original HTML of each section

WHY YOU'LL LIKE IT
• One click — no setup, no copy-paste
• Works on pages AI fetchers can't reach — you've already passed the Cloudflare check and any login
• Smart content detection — strips navigation, sidebars, and footers automatically
• Works on JavaScript-rendered docs (reads the live page)
• 100% local — no accounts, no servers, no tracking, no network requests
• No external libraries; everything runs inside the extension

PERFECT FOR
• Giving AI assistants accurate, current context for the docs you rely on
• Building a personal RAG knowledge base from custom, internal, or niche docs
• Capturing docs behind Cloudflare, logins, or heavy JavaScript
• Offline reading and archiving

HOW TO USE
1. Open a documentation page
2. Click the 403 Whisperer icon
3. (Optional) name the output folder
4. Click "Ingest this page → ZIP"
5. Unzip the file from your Downloads folder

Your data never leaves your computer.