socratic
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE] (SAFE): The skill package contains no executable scripts (Python, JavaScript, etc.) or binary files. It is comprised entirely of Markdown-based prompt templates and reference documentation.\n- [SAFE] (SAFE): No malicious behaviors, obfuscation techniques, hardcoded credentials, or persistence mechanisms were detected. The skill's stated purpose (Socratic dialogue) aligns perfectly with its content.\n- [Indirect Prompt Injection] (SAFE): The skill processes user-supplied topics or beliefs to facilitate dialogue. While it lacks explicit boundary markers, it is considered safe because it has no underlying capabilities (e.g., shell access, network requests, file modification) that could be abused through malicious input.\n
- Ingestion points: User input areas in
SKILL.md,commands/question.md, andcommands/socratic.md.\n - Boundary markers: Absent (markdown headers used for structure only).\n
- Capability inventory: None. The skill does not call any external tools or execute commands.\n
- Sanitization: Absent.
Audit Metadata