website-launch-kit

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

• Third-party content exposure detected (high risk: 0.90). The skill's mandatory "Design Inspiration" and "Research & Design Extraction" steps (Phase 2 and Phase 3 in SKILL.md and references/02-research-guide.md) require the agent to accept a user-provided public URL and use WebFetch/browser_subagent to fetch, screenshot, and analyze that external website, then directly use extracted colors, typography and layout in the generated design and code—exposing the agent to untrusted third-party content that can influence actions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs the agent to use WebFetch or browser_subagent to fetch and analyze the user-provided inspiration URL at runtime (e.g., sites like https://www.awwwards.com/ listed in Phase 2), and that fetched site content is injected into the agent's design extraction and prompts — meaning external site content directly controls the agent's behavior.