deep-research-skills

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

• Third-party content exposure detected (high risk: 0.90). High risk: the required runtime workflow for these deep-research skills involves fetching/reading external web/academic sources (e.g., via “web search engines” and scraping/reading retrieved documents), and that retrieved outsider-authored prose is then fed back into the agent’s LLM for analysis/synthesis.