ccc-design

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

• Third-party content exposure detected (high risk: 1.00). The design-consultation skill explicitly uses WebSearch and a headless "browse" binary in Phase 2 to visit public websites, capture screenshots/snapshots, and synthesize findings to inform design recommendations (see design-consultation SKILL.md Phase 2), so untrusted third‑party web content is fetched and read and can materially influence the agent's decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's setup instructions instruct installing a required runtime dependency with a direct shell pipeline that fetches and executes remote code (curl -fsSL https://bun.sh/install | bash), so the URL https://bun.sh/install is used at runtime to retrieve and run remote code.