ai-code-review

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

• Third-party content exposure detected (high risk: 0.85). The required workflow runs external reviewer CLIs (CodeRabbit cr review --agent and Greptile greptile review), which at runtime send the repository diff/code to third-party services and ingest their returned findings as free-text into the agent context; those findings are outsider-authored (by the external tools/services) and thus indirect prompt-injection risk.