generate-tasks

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow reads and analyzes a PRD/feature/requirements document (“Break down the feature/PRD”), and since those requirements are provided via the create-prd integration step by an outsider source (user did not author/choose it), that free text can enter the LLM context indirectly through the requirements-analysis step.