PromptDock — PDF Crop & Extract for ChatGPT

Crop questions from any PDF into ChatGPT, extract them as text, and auto-send them one by one.

romptDock docks a clean panel onto ChatGPT so you can work through PDF question banks without ever taking a screenshot. WHAT IT DOES 📄 Upload once, reuse anytime Load a PDF and it's automatically saved to your local library — reopen it instantly next time, no re-uploading. ✂️ Crop straight into the chat box Drag over any question in the PDF preview and it's sent directly into ChatGPT as an image — no screenshot tool, no clipboard juggling. Works cleanly at any zoom level, and auto-scrolls as you drag near an edge so you can select something bigger than what's on screen. 🔍 Auto-extract numbered questions PromptDock scans the PDF text and detects numbered questions automatically (1., 2), Q1., or plain table-style numbering), resetting the count at each new unit/section. Click any question to send it to the chat as text, or copy it. ▶ Autosolve Point it at your extracted question list and it sends them to ChatGPT one after another automatically — waiting for each answer to finish, cooling down 5 seconds, then moving to the next. Stop it anytime. If one question fails to send, it's skipped so the rest keep going. 🔗 Save your chat links Bookmark the current ChatGPT conversation, tagged to the PDF and page you were working from, so you can find your way back to a module and its answers later. PRIVACY Everything runs locally in your browser. PromptDock has no backend server — your PDFs, crops, and chat links never leave your device. Full privacy policy: [YOUR PRIVACY POLICY URL HERE] WORKS ON chatgpt.com and chat.openai.com