continuous-llm-red-teaming-with-promptfoo

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

• Third-party content exposure detected (low risk: 0.10). In scripts/agent.py, the runtime workflow runs promptfoo redteam run -c ..., and the resulting prompts/responses (including any injection text generated by Promptfoo) are processed locally by the script only via the model-generated results.json file rather than any outsider-authored free-text feed/queue.