l2

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

• Third-party content exposure detected (low risk: 0.10). The required workflow fetches and reads a remote markdown file from a specific URL provided in the skill documentation.

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

• Potentially malicious external URL detected (high risk: 0.70). An operational dependency on an external URL (ethskills.com) fetches runtime skill instructions from an arbitrary third-party domain of unknown provenance, representing a weak and unverifiable runtime dependency.