detect-ai

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s /detect-ai workflow extracts the caller-provided $ARGUMENTS text and sends it verbatim to the third-party endpoint https://humanizerai.com/api/v1/detect as the JSON "text" field, so outsider-authored free text is directly ingested at runtime (via user input).