atai-anomaly-discovery-agent

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime path ingests outsider-authored free text via the user-provided input CSV uploaded to the Files API and then fed into the Agents API pipeline (the text content of CSV cells is LLM-readable at runtime if parsed/used by the agent).