Fast review
The layout is built for quick scanning, with file details and metadata sections grouped into a readable structure.
The platform is designed to keep the workflow fast, readable, and privacy-aware from the first upload to the final report.
When a DOCX or PDF file is submitted, the site reads the available metadata blocks directly from the document structure. For DOCX files, this includes core, extended, and custom properties. For PDF files, it inspects the Info dictionary and any available XMP metadata.
The AI Writing Risk Indicators page uses extracted body text, language detection, and language-specific rules for English and Bulgarian. It produces separate subscores for structural regularity, formulaic framing, personal voice, and optional baseline drift, then rolls them into one aggregate score.
If a PDF exposes little or no embedded text, the extractor can fall back to OCR so scan-heavy documents can still be reviewed. Results are presented in grouped sections so the output is easier to review during audits, editorial checks, due diligence, or technical inspections.
The layout is built for quick scanning, with file details and metadata sections grouped into a readable structure.
The site is intended for real workflows, not a prototype demo, which is why the interface, privacy messaging, and report structure are all consistent.
The separate AI Writing Risk Indicators page analyzes extracted body text and returns heuristic indicators instead of making a final authorship claim.