omnistudio-dependencies-analyze

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

• Third-party content exposure detected (high risk: 0.85). High: at runtime the skill queries outsider-authored OmniStudio component metadata (e.g., OmniProcessElement.PropertySetConfig and OmniUiCard.DataSourceConfig / OmniUiCard.Definition JSON) via sf data query, then parses that JSON into LLM-readable text for dependency extraction and reporting—this content is authored by other users in the target org, not by the operating user.