housekeep-design
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes shell commands such as
rg(ripgrep) for repository-wide searches andeslint --fixfor mechanical refactoring of tokens and components. It also employsplaywright-clifor visual regression testing (screenshots). These operations are well-scoped to the skill's primary purpose of design system housekeeping. - [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted codebase content (CSS, TSX, JSX files) and possesses the capability to modify files and execute shell commands. However, no specific patterns attempting to exploit this surface or bypass agent safety guidelines were identified.
- [DYNAMIC_EXECUTION]: The skill mentions using codemods and scripted replacements to migrate deprecated tokens. This is a standard practice for large-scale refactoring and does not involve runtime execution of untrusted remote code.
- [DATA_EXPOSURE]: The regex patterns used by the tool specifically target visual design elements like hex codes, RGB/HSL values, and Tailwind CSS arbitrary values. There are no patterns targeting sensitive files (e.g.,
.env,.ssh) or credentials.
Audit Metadata