hugging-face-evaluation

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

• Third-party content exposure detected (high risk: 0.85). The extract-readme/inspect-tables runtime path loads a Hugging Face model card via ModelCard.load(repo_id) and then parses card.content (outsider-authored README text from the target repo) into tables that are converted into YAML/printed to stdout and optionally written back to the model card.