security-first-2025
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides educational content and secure coding templates designed to help users write hardened Bash scripts. It does not execute any autonomous actions or exhibit malicious behavior.- [COMMAND_EXECUTION]: The skill provides clear guidance on how to avoid command injection by sanitizing user input and using safe execution patterns like array-based arguments instead of shell evaluation.- [CREDENTIALS_UNSAFE]: The skill explicitly warns against hardcoding credentials and provides safe patterns for using secrets from environment variables, files, or cloud secret managers.- [INDIRECT_PROMPT_INJECTION]: The provided templates focus on validating and sanitizing untrusted inputs from files and variables, which reduces the surface area for injection attacks in script automation.
Audit Metadata