anomaly-hunter

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

• Third-party content exposure detected (high risk: 0.90). The skill's discovery-mode and implementation guide explicitly require performing web literature checks and reading external sources (e.g., "Use web search tools" and check Google Scholar/arXiv, GitHub issues, Stack Overflow, Wikipedia) as described in references/discovery-mode.md and references/implementation-guide.md, meaning the agent ingests untrusted public content that can change its hypotheses and experiment actions.