social-media-posts

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

• Third-party content exposure detected (high risk: 0.85). The required workflow explicitly allows the user to provide a “path to blog post” or “URL to existing content,” which the skill then “read[s]…and extract[s] the key messages,” meaning runtime-fetched outsider-authored free text (public web pages or downloaded/forwarded documents) can be ingested into the LLM context for repurposing.