scenario-ideogram

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

• Third-party content exposure detected (medium risk: 0.30). In SKILL.md’s required workflow for selecting and running Ideogram models on Scenario via MCP, the runtime path model_run/model_schema_get consumes the caller-supplied parameters.prompt (and other free-form text fields like negativePrompt/expandPrompt), which is outsider-authored free text.