iblai-api-agent-eval

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

• Third-party content exposure detected (medium risk: 0.65). SKILL.md describes creating/adding dataset items and starting an experiment where the service sends each question to the agent and can grade via LLM-as-Judge, so outsider-authored free text could enter the runtime LLM context via dataset items derived from externally sourced chat traces/linked uploads.