The template utilizes a multi-layer steganographic approach to assert provenance.
Configurable text strings (such as "CONFIDENTIAL", "DRAFT", or the Author Name) can be diagonally overlaid across every page of the manuscript.
By default, the overlay_opacity is set to 0.08 (8% visibility). At this setting, the text is minimally intrusive to human readers but profoundly disrupts naive OCR scanners and automated data-scraping neural networks unless they deploy advanced de-noising algorithms.
If barcodes_enabled is true, the SteganographyProcessor will automatically extract the authors array from your projects/<name>/manuscript/config.yaml.
If it finds email keys for the authors, the processor dynamically generates an actionable mailto: URI bridging all authors together in the to field, and embeds it as a High Error-Correction (Q Level) QR code on every page margin.
The processor bypasses the visual later completely to rewrite the PDF metadata dictionaries (XMP and standard PDF info keys), unconditionally branding the Author, Subject, and specific identifiers (like a Doc-ID tracking UUID) into the file internals.
Steganography is controlled by infrastructure/config/secure_config.yaml.
If you delete this file, the pipeline will fallback to conservative hard-coded defaults (Text overlay: CONFIDENTIAL, Opacity: 8%, Hash and Metadata enabled, QR barcodes enabled).
steganography:
# Mode: text, qr, or none
overlay_mode: text
# The text to display
overlay_text: 'PROVENANCE PROTECTED'
# 0.0 to 1.0 (0.08 is recommended for disruption without visual clutter)
overlay_opacity: 0.08
# RGB values (e.g. Red is [255, 0, 0])
overlay_color_rgb: [128, 128, 128]
# Generate and append margin QR code emails
barcodes_enabled: true
# Inject cryptographic SHA manifests
hashing_enabled: true