scientific-writing

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime workflow explicitly relies on the separate research-lookup step to gather literature and extract “bullet points” for later paragraph conversion; since that lookup commonly ingests public web/third-party authored text (e.g., scraped pages or fetched articles) into the agent context, this creates an indirect prompt-injection exposure path.