prompt-guard

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill contains phrases like "Ignore all previous instructions" and "You are now in developer mode". These are clearly labeled as examples for testing the detection tool's accuracy and do not pose a threat to the agent's operating environment.
  • [EXTERNAL_DOWNLOADS]: The Python examples demonstrate downloading model assets from Hugging Face using the transformers library. These resources are hosted by a well-known service and are necessary for the skill's primary function of providing safety filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 07:07 PM
Security Audit — agent-trust-hub — prompt-guard