Noroboto

A proof-of-concept Unicode obfuscation tool for .docx and .pdf documents. Every glyph in the body is recoded so text extractors see only Private Use Area characters; the rendered page is unchanged.

Background

• Noroboto and Legal Tech's Mythos Moment
• Noroboto and the PDF That Lied Twice

Setup

pip install -r requirements.txt

Command line

python noroboto.py input.[docx|pdf] output.[docx|pdf]

Run the tests

python -m unittest discover tests

The corpus test runs the CLI over every .docx and .pdf it finds under docs/.

Run the server

python app.py

Then open http://127.0.0.1:5000. The upload accepts both .docx and .pdf.