space-brand-ui-design

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

• Third-party content exposure detected (high risk: 0.85). Runtime path: the skill runs npx getdesign@latest add <slug> and then cat /tmp/design-md-tmp/DESIGN.md, so the LLM ingests free-form text from the externally published getdesign.md package/repo (public, non-user-authored content), which is an outsider source.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly runs "npx getdesign@latest add " at runtime to fetch DESIGN.md from getdesign.md and then reads/injects that file as the design context for generation, so https://getdesign.md (and the npx getdesign package) is a runtime external dependency that directly controls prompts.