Security audit
Security checks for vulnerabilities and agentic risk
This skill is a simple local model-card generator with some documentation quality issues but no evidence of hidden, destructive, persistent, or exfiltrating behavior.
Installers should treat this as a benign local documentation helper. Before relying on it operationally, fix the CLI versus JSON-RPC documentation mismatch and align the allowed-tools/permissions metadata with the actual script behavior; expect generated output to be Chinese-labeled unless the script is localized.
The heading and introductory instructions are written as a Chinese-only agent manual, directing both AI agents and maintainers to use this document without indicating any language opt-in or alternative locale. Under the policy, forcing a specific language without user choice is a natural-language policy violation unless the locale constraint is explicitly justified.
The README includes a full Chinese section alongside English, but the policy concern is not the presence of Chinese itself. However, the document presents bilingual content rather than forcing a single language, so this does not constitute a violation. No opt-in is required because both languages are offered directly.
The manifest presents the title and display name only in Chinese, and the rest of the document is bilingual but does not offer a user choice of preferred language. This can violate a language/locale policy when a skill imposes a specific language presentation without explicit opt-in or justification.
The manifest interface section defines the skill as a CLI with call_pattern: args, while the documentation instructs users to pass task configuration via JSON-RPC. That is an active contradiction about how the skill is actually invoked, not just an omitted detail.
The Chinese documentation repeats that the skill can be used through JSON-RPC, but the manifest specifies a CLI entry point with argument-based invocation. This is a clear intent/usage contradiction that could mislead integrators about the skill's real behavior.
The example usage includes the limitation "只支持中文" as a built-in model constraint. Under the policy, forcing a specific language without opt-in or a clearly justified region-specific reason is a natural-language policy concern, and no alternative language choice or justification is provided in this file.
The natural-language content of the file is entirely in Chinese and does not indicate that users or agents may choose another language. Under the policy, forcing a specific language without opt-in can be a locale-policy violation unless the constraint is explicitly justified.
