Attaching documents
Through the paperclip in the input area you attach a document. ChEddi extracts the text from it and works with that, for instance to create content elements from a PDF.
Text-based documents are read: PDF, Word, spreadsheets and plain text files. The answer to the upload tells you per file whether it can be read at all. So you learn that before you send the message. Possible notes are “Too large to read”, “This format cannot be read in this installation” or “No text can be read from this file”.
Images as attachments
Images are not yet supported as attachments and are rejected on upload. The background is technical: messages currently travel to the server as plain text, so image content would not reach the model at all. Image formats are therefore not on the list of permitted file types for now.