prompt-engineering

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides educational content and examples for prompt engineering. It demonstrates how to use the belt CLI tool (inference.sh) for various AI tasks such as code review, creative writing, and image generation.
  • [COMMAND_EXECUTION]: The skill includes numerous examples of belt app run commands. These are standard usage examples for the documented tool and do not involve unauthorized command execution or injection of untrusted input.
  • [EXTERNAL_DOWNLOADS]: The skill references a CLI installation script from a well-known service (GitHub/inference-sh). This is documented neutrally as a requirement for using the shown examples and does not escalate the verdict.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides templates for processing data (like code reviews or text extraction). While this defines an attack surface, the context is educational, and the risk is inherent to the primary purpose of prompt engineering instructions. Typical LLM guardrails apply at runtime.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 11:00 PM
Security Audit — agent-trust-hub — prompt-engineering