bonsai-impl-classification

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md and examples explicitly use the bsdd.Client and Blender operators (e.g., LoadBSDDDictionaries, SearchBSDDClassifications, client.search_class and client.get_class in references/examples.md) to fetch live classification data from the public buildingSMART bSDD API (identifier.buildingsmart.org / api.bsdd.buildingsmart.org) and then apply those returned class data and properties to IFC models, meaning external third‑party content is read and directly drives tool actions and decisions.