wkp

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

• Third-party content exposure detected (high risk: 0.80). This skill directly fetches and extracts content from Wikipedia (see SKILL.md commands like "x wkp extract OpenAI" and "x wkp hop 'Rust Programming'"), which is an open, user-editable third-party source the agent reads and could use to influence subsequent decisions.