humanize-text

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

• Third-party content exposure detected (medium risk: 0.30). In the rewrite workflow, the agent sends the user-supplied “complete text” directly to the hosted Humanizer PRO MCP tool humanize_text (and similarly scan_ai_detection) via https://texthumanizer.pro/mcp, so outsiders can inject free text through any channel that provides the text to be edited/analyzed.