paniolo-naming-best-practices
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No evidence of instruction override, safety bypass, or role-play injection. The instructions are focused entirely on providing coding standards and naming conventions.
- [DATA_EXFILTRATION]: No sensitive file paths, hardcoded credentials, or network operations were detected. The skill specifically notes that no network access is needed.
- [OBFUSCATION]: The content is clear-text markdown and TypeScript examples. No hidden characters, homoglyphs, or encoded payloads were found.
- [REMOTE_CODE_EXECUTION]: No remote script downloads or piped shell executions. The skill mentions standard local tools like
npm run lintfor validation. - [INDIRECT_PROMPT_INJECTION]: The skill has a standard ingestion surface as it analyzes user-provided code and symbols to suggest naming improvements. However, it lacks the combined capabilities (like network exfiltration) to pose a significant risk via this vector.
- [DYNAMIC_CONTEXT_INJECTION]: No dynamic execution placeholders (e.g.,
!command) were found in the skill definitions.
Audit Metadata