writing-staffing

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

• Third-party content exposure detected (medium risk: 0.65). Outsider free text can enter the LLM context via the runtime web-research path: the skill’s researcher workflow uses the base @web-researcher for “web research” and that agent typically fetches public web pages and ingests their readable text (e.g., articles/scraped pages) into the model for report generation.