speckle-impl-rhino-grasshopper

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly ingests models from third-party Speckle servers via the "Speckle Model URL" and "Load" components (see SKILL.md and references/examples.md), and then reads and interprets that untrusted/user-generated model data (Query Objects, Filter Objects, Deconstruct) as part of its workflow, which can materially influence filtering, object mutation, and subsequent publish actions.