name-audition

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow includes Stage 3b “Collision research,” which uses the firecrawl skill or web search to fetch and read outsider-authored public web pages (e.g., SaaS/startup listings, GitHub repos, PyPI/npm package pages) and then feeds the extracted page text/URLs into the agent’s LLM context for summarization/table building.