testing

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

• Third-party content exposure detected (high risk: 0.72). The skill’s runtime workflow ingests untrusted, outsider-authored free text from “failing validation log”/artifact outputs (e.g., artifacts/test/test-output-e2e.log in the pressure-prompt-injection case), which can contain prompt-injection strings that the agent may include in its LLM context for analysis.