skeleton-ui

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

• Third-party content exposure detected (high risk: 0.70). This skill's SKILL.md prerequisites explicitly tell the agent to consult https://www.skeleton.dev/llms.txt and to use the webfetch tool to retrieve the latest component specs, so the agent will fetch and interpret public third‑party documentation that can influence its actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs using the webfetch tool at runtime to retrieve https://www.skeleton.dev/llms.txt, which the agent would use as authoritative component specs to shape its prompts/output, so the external URL directly controls agent instructions.