Skip to main content

My upload became a title-only memory

A PDF or image stored without a reading has no facts to recall. Here's how that happens and how to fix it.

A
Written by Aaron Wittman

JustOnce runs no OCR. When a PDF, scan or photo is stored without its text — your AI tool saved the file but didn't pass a transcription, or the upload link was used without one — the result is a memory with a title, the attached file, and no content for recall to use. The AI tool's result reports this as textSource: none.

Fix it

  • Ask your AI tool to read it: "Open the policy document you stored yesterday, read it, and update the memory with the insurer, policy number, excess and renewal date." It fetches the file with fetch_document and corrects the memory with update_memory.

  • Or type the facts in: open the memory in the app, click Edit, and paste what matters.

Avoid it next time

Ask for the reading and the storing in one instruction: "Read this, save the key facts to JustOnce, and keep the original attached."

On a paid plan, you can also save your own AI key under Settings → Email forwarding. JustOnce then uses that key to read PDFs and images that arrive without text — uploaded, stored by an AI tool or forwarded by email.

Did this answer your question?