recipe-add-integration-tests

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow reads outsider-authored Design Doc/UI Spec text provided via user-supplied $ARGUMENTS in Step 1 (and then forwards those documents into Step 2’s spawned acceptance-test-generator prompt), creating an indirect prompt-injection path from external document content into the LLM context.