prompt-engineering

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as an educational guide for prompt engineering techniques, including chain-of-thought, few-shot prompting, and output formatting.
  • [SAFE]: All command-line examples utilize the vendor's official CLI tool (infsh) and follow the restricted scope defined in the allowed-tools configuration.
  • [SAFE]: References to external skills and the installation of the CLI tool use the vendor's official namespace (inference-sh), aligning with expected platform behavior.
  • [SAFE]: Image assets and links point to the vendor's verified domain (inference.sh) with no evidence of typosquatting or malicious redirection.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 09:05 PM