structured-data-extraction

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

• Third-party content exposure detected (medium risk: 0.30). In SKILL.md the required workflow has the operator provide arbitrary page URLs (e.g., step 1 python3 scripts/extract.py https://site.com/product), and scripts/extract.py then fetches that page and ingests outsider-authored free text from its HTML—specifically contents of application/ld+json / application/json script blocks, Open Graph/meta tags, microdata attributes/text, and tables via the HTML parsers at runtime.