ai-vibe-detector
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a non-executable, pure-prompt implementation. It is composed entirely of Markdown instructions and reference material, lacking any scripts (Python, Node.js, or Shell) or binary files.
- [PROMPT_INJECTION]: The instructions in SKILL.md and README.md do not contain any patterns for bypassing AI safety guidelines. On the contrary, they include explicit directives to avoid promising the defeat of AI detectors or providing specific evasion tactics.
- [DATA_EXFILTRATION]: No network communication patterns (such as curl, wget, or API fetch) or sensitive file access paths (such as SSH keys or environment files) were identified.
- [EXTERNAL_DOWNLOADS]: The skill does not define any external dependencies or remote script execution patterns. All analysis logic is contained within the local text files.
Audit Metadata