receiving-code-review

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

• Third-party content exposure detected (high risk: 0.75). The required workflow is triggered by “review comments” from human review channels (e.g., code-review / pr-multi-perspective-review), which are outsider-authored free text that the agent would need to read and restate before verification—creating an indirect prompt-injection risk.