validate

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly directs the agent to research and summarize public, user-generated sources—naming Reddit, forums, Twitter/X, competitor reviews and online communities—and to find and interpret threads/links as part of its required workflow, which means it ingests untrusted third-party content that can materially influence decisions.