baoyu-cover-image

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

• Third-party content exposure detected (medium risk: 0.30). The workflow ingests outsider-authored free text via user-pasted article content in Step 1 (“Save source content … if pasted, save to source.md” then “Analyze content” and later includes the result in the generated prompt/cover), making it directly available to the LLM at runtime.