baseline-establishment

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow explicitly includes “Web Searches” (e.g., via brave-search, apify/rag-web-browser, google-scholar-scraper, semantic-scholar) and “leaderboard-harvesting,” which will fetch and ingest outsider-authored public web/page/paper text into the agent’s LLM context for extraction/assembly.