bmad-create-architecture

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

• Third-party content exposure detected (high risk: 1.00). The workflow steps (e.g., steps/step-03-starter.md and steps/step-04-decisions.md) explicitly instruct the agent to "Search the web" and verify current starter templates and technology versions, meaning the agent will fetch and act on arbitrary public web content which can directly influence tool use and decisions.