jung-perspective
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed of markdown instructions and reference files. There is no executable code (Python, JavaScript, or shell scripts) included in the skill package.
- [DATA_EXPOSURE]: No hardcoded credentials, API keys, or sensitive file paths were detected. The skill does not attempt to access the local file system or environment variables.
- [REMOTE_CODE_EXECUTION]: No remote code execution patterns, such as piped shell commands (curl | bash) or dynamic code loading, were found.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes user-provided text (such as dreams or life challenges), it lacks any tools or system-level capabilities (e.g., file writing, network requests) that could be exploited via injection. The instructions are safely scoped to role-playing and psychological analysis.
- [OBFUSCATION]: All instructions and examples are provided in clear, readable text. No Base64, hex encoding, or homoglyph attacks were identified.
Audit Metadata