ai-friendly-cli

Pass

Audited by Gen Agent Trust Hub on Mar 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, hidden commands, or security risks were detected. The skill is entirely instructional and focused on promoting secure software development practices for AI environments.
  • [PROMPT_INJECTION]: The skill includes explicit guidance on implementing response sanitization to defend against indirect prompt injection, specifically suggesting the removal of instructions like 'IGNORE ALL PREVIOUS INSTRUCTIONS'.
  • [DATA_EXFILTRATION]: Includes best practices for preventing the exposure of sensitive data, such as masking secrets in output and using environment variables instead of command-line flags for credential injection.
  • [COMMAND_EXECUTION]: Provides defensive patterns for input hardening, including robust validation logic to prevent path traversal and control character injection during CLI command construction.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 22, 2026, 11:55 PM