prompt-engineering-agent
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional, providing templates and strategies for agent prompt engineering.
- [SAFE]: Includes explicit safety guidelines, such as stop conditions for when a user's request is unsafe or out of scope.
- [SAFE]: Promotes secure tool design rules, including schema-first validation, idempotent read operations, and confirmed writes.
- [SAFE]: No malicious obfuscation, data exfiltration, or unauthorized command execution patterns were found.
Audit Metadata