Security audit
Security checks for vulnerabilities and agentic risk
This skill is a Chinese-language code review and scoring workflow with a small local scorer script, and I found no hidden persistence, credential access, network exfiltration, or destructive behavior.
Install this only if you want a Chinese-language review workflow and are comfortable with the agent or bundled script reading code files you explicitly provide for scoring. Treat the scorer as a lightweight heuristic, not a complete security or quality audit.
The skill’s declared purpose and the described/observed behavior do not align, which can mislead users and reviewers about what the skill actually does and what risks it introduces. Description-behavior mismatch is dangerous because it undermines informed consent, complicates security review, and can conceal unexpected local file processing or other operational behavior behind an innocuous workflow description.
Without declared permissions the skill's intent is opaque and cannot be validated.
Overly broad trigger phrases can cause the skill to activate in contexts the user did not intend, increasing the chance of inappropriate task interception or unneeded analysis. In an agent environment, vague activation rules can become a security issue because they expand the skill’s effective attack surface and may route sensitive prompts into an unexpected workflow.
Forcing Chinese-language behavior without user opt-in can create misunderstanding of outputs, reduce transparency, and increase the likelihood that users approve or run a workflow they do not fully understand. This is mainly a trust and usability security issue rather than direct code execution risk, but it becomes more concerning when the skill is used for review, safety, or policy-relevant decisions.
This code contains multiple natural-language strings, comments, and CLI messages in Chinese, including the module docstring, class/doc method descriptions, scoring labels, and error/usage output. Under the policy rule, forcing a specific language without user opt-in is a locale/language policy violation unless it is clearly justified as region-specific, which is not documented here.
This markdown file contains all user-facing guidance in Chinese, beginning with the title at L001 and continuing throughout the document. Under the policy rule for natural-language violations, forcing a specific language without user opt-in can be a locale policy issue when no alternative language option or justification is provided.
No suspicious patterns detected.
