pinia
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: LOW
Full Analysis
- [PROMPT_INJECTION] (SAFE): The content consists entirely of technical documentation and code examples. There are no instructions that attempt to override the AI agent's safety protocols or system prompts.
- [DATA_EXFILTRATION] (SAFE): No commands or patterns were found that attempt to access sensitive local files (like SSH keys or environment variables) or send data to unauthorized external domains.
- [CREDENTIALS_UNSAFE] (SAFE): No hardcoded API keys, tokens, or other credentials are present in the documentation or code snippets.
- [EXTERNAL_DOWNLOADS] (SAFE): While the skill mentions installing packages via npm or npx (e.g., '@pinia/testing', 'nuxi'), these are standard, well-known tools within the Vue ecosystem and are used as intended for library installation.
- [COMMAND_EXECUTION] (SAFE): Subprocess execution patterns are limited to standard package management commands (npm, npx). No dangerous or arbitrary command execution (e.g., 'sudo' or 'rm') is requested.
- [OBFUSCATION] (SAFE): All markdown files and code snippets are provided in clear, readable text. No hidden Unicode characters, Base64 encoding, or homoglyph attacks were detected.
- [REMOTE_CODE_EXECUTION] (SAFE): There are no patterns involving the download and immediate execution of remote scripts (e.g., 'curl | bash').
Audit Metadata