An upload is validated before it is stored and processed. Depending on the format, the pipeline extracts native text or uses OCR, records processing information, and prepares searchable content. The status shown in Documents Hub is the source for whether a file is ready, still processing, or needs attention.

Safe previews and extracted text help you inspect the content that is available to Query. OCR output, text extraction, embeddings, and search quality vary by document and runtime configuration. Confirm important results against the original page or file.

The accepted formats are determined by the upload interface and backend extractor registry. Legacy Office conversion and some OCR paths require their configured worker or converter. Malware scanning and private object storage are required parts of the production upload path; a deployment that lacks required services should not be assumed to be production-ready.