File indexing
How Studycamp reads your files so AI tools can answer from them, what the statuses mean, and how to retry.
This page is not translated yet. Showing the English version.
Before an AI tool can use a file, Studycamp reads it and splits it into searchable pieces. This is called indexing. It starts automatically after upload.
What happens
- The file is downloaded and its text is extracted. PDF text keeps its page numbers.
- The text is split into chunks, one page at a time.
- Each chunk is embedded so it can be searched by meaning.
Because pages are kept, chat can cite the page an answer came from.
File status
| Status | Meaning |
|---|---|
| Pending | Waiting to start |
| Processing | Being read. Large uploads show chunk progress |
| Ready | Usable by chat, summaries, flashcards and quizzes |
| Error | Reading failed |
Files in a workspace show the status as a badge, and the list refreshes by itself while files are being read. Tools wait for files that are still processing and ask you to try again in a moment.
Which files can be indexed
PDF, Word (.docx), PowerPoint (.pptx), text and Markdown. A file with no extractable text, such as a scan of images with no text layer, produces nothing to search and ends in an error.
Retry
If a file shows an error, open it in the study session file list and click it to start indexing again. In the upload progress window, click Retry. If it fails again, re-export the file as a standard PDF or DOCX and upload it again.
Good to know
- Chat can only cite files that are Ready.
- Scanned PDFs are not supported by Agent either, which reads documents separately. See Syllabus extraction.