Security audit
Security checks for vulnerabilities and agentic risk
This is a local Chinese-language prompt/avatar generator that prints output for manual use and does not show hidden network, file-writing, persistence, or credential behavior.
Installers should understand this is a Chinese-language local CLI and should manually review any generated SOUL.md/System Prompt text before pasting it into an agent, especially any style guidance that discourages normal safety or identity wording.
Line L22 explicitly says the SKILL name and all CHANGELOG descriptions were changed to Chinese for domestic developers. This is a natural-language locale policy constraint and does not mention any user choice, opt-in, or justified region-specific limitation.
The title and opening description explicitly label the skill as a Chinese version and present it as the interaction mode, with no indication that users may choose another language. This is a natural-language locale policy concern because it imposes a language constraint without documented user opt-in or justification as a region-specific tool.
The title and description identify this as a Chinese-only version ("中文版") and all user-facing instructions are presented only in Chinese. This imposes a language choice without stating that users can opt into another language or select their preferred locale, which matches the language/locale policy violation criteria.
该白皮书声明 AI 智能体的具身形象“必须作为 SOUL.md 文件的原生组成部分，以纯文本描述或代码化定义的方式存在”，属于自然语言层面的强制格式/语言要求。文件未说明用户可选择其他语言或本地化表达方式，也未给出地区合规等正当限定，因此存在未获用户选择即施加特定语言/文本格式约束的政策风险。
The manifest name, author, description, and tags are entirely in Chinese, and there is no indication that users may choose another language or that the skill is intentionally limited to a China-specific/regional compliance context. Under the policy for natural-language issues, forcing a specific language without user opt-in can be a locale-policy violation.
The anti-pattern rules explicitly discourage AI self-identification and standard cautionary framing, which can weaken safety signaling and make downstream prompt-injection or policy-evasion attempts more effective. In this skill, the strings are configuration content rather than executed control flow, so the danger is contextual and moderate, but they still normalize suppression of protective behavior in generated outputs.
The user-facing banner and prompts are entirely in Chinese, and the skill name itself indicates a CN-specific variant, but the code does not present any opt-in, fallback, or language-selection mechanism. Under the stated policy, forcing a specific language without user choice is a natural-language policy violation unless the locale restriction is clearly documented and justified.
No suspicious patterns detected.
