ai-ecommerce-expert-kuaishou-ecommerce-content

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The code launches an npm package via npx@latest at runtime while forwarding a sensitive MCP_TOKEN in the environment, creating a high-risk supply-chain / remote-code-execution and credential-exfiltration vector if the fetched package is malicious or compromised.

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

• Third-party content exposure detected (medium risk: 0.30). 运行时在 scripts/imiva_mcp.py 的 invoke(...)->tools/call 中把代理用户/调用方提供的 --args JSON（如 prompt、素材 HTTPS URL 等）直接传给 IMIVA MCP（@infimind/ecom-content-cli），属于对外部自由文本的直接摄取路径。

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

• Potentially malicious external URL detected (high risk: 0.90). The skill invokes npx to run the remote npm package "@infimind/ecom-content-cli@latest" at runtime (scripts/imiva_mcp.py uses subprocess to run npx with that package), which causes fetching and execution of remote code: "@infimind/ecom-content-cli@latest".