contributing
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted code and pull request content while maintaining command execution capabilities. 1. Ingestion points: The agent reads Vue SFC files, MDC documentation files, and pull request data as part of the development workflow. 2. Boundary markers: Standard Markdown and Vue syntax are used without specific isolation markers to prevent the agent from obeying instructions embedded in the processed data. 3. Capability inventory: The skill is authorized to execute local development commands like 'pnpm run' and the 'nuxt-ui' CLI for scaffolding. 4. Sanitization: No explicit sanitization or validation of the input code or documentation is mentioned.
- [REMOTE_CODE_EXECUTION]: The skill references several standard Node.js packages and a local CLI tool used for project development. These resources are part of the official Nuxt ecosystem and are considered safe, with no unverified remote execution patterns detected.
Audit Metadata