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
Risk Level
SAFE
Analyzed
Oct 23, 2026, 11:01 PM