ChatGPT’s business and Team accounts often have no “Export data” button — the one-click archive that personal accounts get is simply missing, and there’s no official way to take your conversation history with you when the workspace is closed or your subscription is cancelled. scrapemychats fills that gap: it exports every conversation — text and attached files — to your own computer, then builds a beautiful, searchable, offline archive you can keep forever. (It works on personal accounts too.)
Your data never leaves your machine. The tool drives a real Chrome window on your computer using your own logged-in ChatGPT session. It never sees your password, and nothing is sent anywhere except to chatgpt.com itself.
The offline viewer: categories on the left, your chats in the middle, the full conversation — tables, files, images — on the right. (Sample data shown.)
What you get
export/
├── viewer.html ← open this: search + browse everything
├── manifest.csv ← one row per chat: status, counts
├── errors.log ← anything that couldn't be fetched
└── 001_Some Chat Title_6a5f6592/
├── conversation.json ← complete raw data (every message, tool call)
├── conversation.md ← readable transcript
└── files/ ← attachments, images & generated files
The viewer is a single self-contained HTML file: instant full-text search across all chats, optional Personal/Work categorisation, inline image thumbnails, keyboard navigation. No server, no internet, no dependencies — it works from a USB stick.
Full-text search is instant and highlights matches in the results and inside the open chat. (Sample data shown.)
Quick start (if you’re comfortable with a terminal)
Requirements: Python 3.10+, Google Chrome (or Microsoft Edge with --browser-channel msedge).
pip install -r requirements.txt # 1. Test with a few chats first (a Chrome window opens — log into ChatGPT # in it; if you have multiple workspaces, switch to the right one) python export_chats.py --limit 3 # 2. Check the export/ folder looks right, then run the full export python export_chats.py # 3. Build the viewer, then open export/viewer.html in your browser python build_viewer.py
The chat list is discovered automatically (including chats inside Projects) and saved to chats.csv. Alternatively supply your own list with --csv mylist.csv (first column: chat URL, second: title).