gslides-parser

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). This skill directly reads arbitrary Google Slides content (parsePresentation in gas/parseSlides.js, invoked by scripts/gslides_parser.py analyze) — extracting slide text, image source URLs, links, and video IDs — and then semantically analyzes that user-provided slide content in scripts/gas_to_yaml.py to assign roles/placeholders that can drive downstream actions, so untrusted slide content can materially influence behavior.