harvard-artifacts-collection-analytics-etl

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

• Third-party content exposure detected (high risk: 0.75). At runtime the skill calls the Harvard Art Museums public API (requests.get(...).json() in extract_artifacts), and the returned JSON fields (e.g., title, description, etc.) are converted into LLM-readable strings via transform_artifacts and then used downstream; this is outsider-authored free text from a public web source.