speckle-impl-blender

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

• Third-party content exposure detected (high risk: 0.80). The skill's required "Load from Model" workflow (references/methods.md and SKILL.md) explicitly fetches Speckle project/model content from a Speckle server via a project/model URL or selection, and converts that user-generated/untrusted model and its material/property data into Blender objects—content that the connector must parse and which directly determines object creation and material/instancing behavior.