Security audit
Security checks for vulnerabilities and agentic risk
The package appears to be an incomplete PowerShell sandbox with unsupported safety claims, so it should be reviewed before use with untrusted scripts.
Do not rely on this package as a safe sandbox for untrusted PowerShell scripts until the actual src/sandbox.ps1 implementation, tests, and security controls are present and independently reviewed. Treat the current artifact as documentation for a proposed tool, not a working containment mechanism.
This markdown file presents the skill name, instructions, examples, warnings, and operational guidance entirely in Chinese, with no indication that users may choose another language. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is explicitly justified, which is not documented here.
The document makes strong security assurances such as 'safe for untrusted scripts' and 'ready for production use' even though the listed next steps show key security properties remain untested, including file path isolation, output limits, and .NET type restrictions. This can mislead operators into deploying the sandbox in high-trust scenarios before essential containment controls are validated, increasing the chance of sandbox escape, data access, or denial-of-service from untrusted scripts.
The package description is written entirely in Chinese and does not indicate any language choice or opt-in. Under the natural-language policy rule, this can be treated as a locale/language constraint because the skill metadata presents itself in a single language without offering alternatives or documenting that the skill is region-specific.
No suspicious patterns detected.
