prompt-engineering
Pass
Audited by Gen Agent Trust Hub on Mar 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely documentation-based, providing prompt engineering guides and patterns. It does not contain any executable scripts, shell commands, or network operations.
- [SAFE]: No obfuscation, data exfiltration, or credential exposure was detected. All code snippets are provided as static examples for the user's reference.
- [SAFE]: Although the skill lists tools like 'Bash' and 'WebFetch' in its frontmatter configuration, the body of the skill does not utilize these tools for any background operations or remote data fetching.
- [SAFE]: No prompt injection or bypass instructions were identified. The content focuses on legitimate prompt engineering techniques like Few-Shot Learning and Chain of Thought.
Audit Metadata