motivation-diagnostic-task-redesign

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime LLM context is populated by the teacher-provided free-text fields (e.g., task_description, learning_objective, motivation_concern, student_profiles), which are outsider-authored relative to the agent (the operating user did not author the content being analyzed), so arbitrary free text can be injected into the prompt via these template variables.