vue-best-practices
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and modify user-supplied source code, creating a surface for indirect prompt injection attacks.\n
- Ingestion points: Processes user-provided .vue components and TypeScript source files as part of its primary function (refactoring and review).\n
- Boundary markers: The skill lacks explicit instructions or markers to help the agent differentiate between its own operational instructions and potentially malicious instructions embedded in the code being analyzed.\n
- Capability inventory: The agent is empowered to refactor and generate new code based on the analyzed input, which could lead to the introduction of backdoors or vulnerabilities if an injection is successful.\n
- Sanitization: No input sanitization or validation mechanisms are defined for the code content being processed.\n- [NO_CODE]: The skill consists exclusively of markdown documentation and configuration files. It does not include any executable scripts, binary files, or automated installation processes.
Audit Metadata