data-context-extractor

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

• Third-party content exposure detected (high risk: 0.75). The required workflow ingests outsider-authored free text when the user uploads an existing skill/zip in “Iteration Mode” and the agent reads the included reference files and SKILL.md (those files are not authored by the operating user by definition, and their contents become LLM-readable context).