one-prompt-website

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

• Third-party content exposure detected (low risk: 0.10). The workflow involves fetching a live web page via WebFetch (SKILL.md line 48), which exposes the agent to arbitrary outsider-authored web content.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill makes programmatic requests to Leonardo AI (cloud.leonardo.ai) to generate assets and video loops used directly in the built website, constituting a third-party operational dependency without strong official provenance.