vue-expert
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and refactor external code provided by users, which serves as a primary ingestion point for untrusted data. It lacks explicit instructions to differentiate between the code being reviewed and instructions that might be embedded within that code.
- Ingestion points: User-provided source code files and project structures (SKILL.md).
- Boundary markers: The instructions do not define delimiters or specific safety guidelines to ignore commands hidden in comments or strings within the reviewed code.
- Capability inventory: The skill frontmatter allows access to powerful tools such as
Bash,Write, andEdit. - Sanitization: There is no evidence of input validation or sanitization routines for the data processed by the agent.
- [COMMAND_EXECUTION]: The "Memory Protocol" in the skill documentation instructs the agent to perform shell operations for state management, specifically using
catto read from a local path (.claude/context/memory/learnings.md). This demonstrates an operational pattern of using system commands to handle internal agent data.
Audit Metadata