meigen-ai-design-mcp

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly ingests a public, user-generated prompt library (the nanobanana-trending-prompts GitHub prompt collection referenced in SKILL.md) via tools like search_gallery and get_inspiration which the agent reads and uses to craft prompts and drive image-generation/sub-agent actions, so those external prompts could materially influence tool use and behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The MCP configuration and install instructions run remote packages (e.g., "npx meigen@1.3.1" / "npx meigen ...") which causes npx to fetch and execute remote code at runtime, so the meigen npm package (invoked via npx) is a runtime external dependency that can execute code.