coverage-tracker

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

• Third-party content exposure detected (high risk: 0.85). The workflow’s LLM classification step ingests outsider-authored free text from news-search results (e.g., article titles/snippets/summaries and potentially full article text when available) fetched at runtime from public web sources, which the agent then passes into the LLM for verdicting.