Feature

Document Parsing

Extract structured text from PDFs, DOCX, images, and more.

Supported formats

FormatExtensions
PDF.pdf
Word.docx, .doc
PowerPoint.pptx, .ppt
Excel.xlsx, .xls
Web.html, .htm
Images.png, .jpg, .jpeg, .tiff, .bmp

How it works

Jetro converts documents to structured, searchable text. Tables are preserved, images are extracted, and the output is saved as a note file in your project. The agent can then reference this parsed content for analysis. For example: “What was the R&D spend mentioned in the annual report?”

Project-scoped notes

Parsed documents are saved to your project’s notes/ folder. If no project is active, they go to .berrywise/notes/. The agent can search and reference these notes across conversations.