teach
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is comprised of instructional text and metadata without any accompanying scripts or executable files.
- [INDIRECT_PROMPT_INJECTION]: Assessment of attack surface: 1. Ingestion points: Code snippets and conversation history (SKILL.md). 2. Boundary markers: Not present. 3. Capability inventory: None; the skill is limited to providing explanations. 4. Sanitization: Not applicable for this descriptive task. The risk is non-existent as the skill lacks actionable permissions.
- [EXTERNAL_DOWNLOADS]: The documentation includes references to an upstream GitHub repository for attribution and maintenance, which is a standard and safe practice referencing a well-known platform.
Audit Metadata