design-review

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

• Third-party content exposure detected (medium risk: 0.30). Source: /design-review runtime path includes navigating to a user-supplied Target URL and extracting page content via the browse/screenshot/snapshot steps (e.g., $B goto & Phase 1-6), which means outsider-authored text on those pages can be ingested.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill runs a runtime shell command that fetches and executes remote code via "curl -fsSL https://bun.sh/install | bash", which downloads and runs code from https://bun.sh/install during setup.