prompt-evaluator

Pass

Audited by Gen Agent Trust Hub on Jul 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions include clear boundaries to prevent prompt injection by mandating that user input be treated as inert text for evaluation rather than instructions.
  • [COMMAND_EXECUTION]: The skill includes hard rules that explicitly forbid the use of shell commands, tool calls, and file system access.
  • [DATA_EXFILTRATION]: No network or file-system capabilities are requested or used in the skill configuration or instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 4, 2026, 07:24 PM
Security Audit — agent-trust-hub — prompt-evaluator