research-knowledge-base

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs the agent at runtime to "Read https://roy-tong.github.io/llms.txt" (with fallbacks "https://roy-tong.github.io/sitemap.xml" and "https://roy-tong.github.io/feed.xml"), so externally fetched content would directly control which pages the agent selects and how it composes responses.