Security audit
Security checks for vulnerabilities and agentic risk
This is a Markdown-only Chinese guide for designing AI agents, with no executable code or hidden behavior found.
Installers should treat this as general architecture guidance. If using its examples in a real enterprise agent, add explicit user consent, data minimization, retention limits, and confirmation before storing customer preferences or writing to CRM/Feishu-style systems.
The Usage section lists open-ended prompts such as “帮我设计...” and “如何给智能体加上长期记忆...”, which are common natural-language requests rather than narrowly scoped activation phrases. The file does not provide explicit trigger constraints, exclusions, or negative examples to clarify when this skill should or should not activate.
The skill explicitly recommends long-term memory of user preferences and automated writes to external documents, but it does not mention consent, data minimization, retention limits, or confirmation before persistence. In an enterprise agent context, this can lead to unauthorized storage of personal or sensitive business data and unintended modification of systems like Feishu documents.
The skill name and description are explicitly Chinese-focused, and the usage/examples are entirely in Chinese, which can imply a fixed language experience. There is no statement that users may interact in another language or opt into Chinese output.
No suspicious patterns detected.
