The Agent Skills Directory

[Command Execution] (SAFE): The skill provides documentation on the 'Bash' tool and the '!' inline execution syntax. It promotes security by recommending the use of granular command filters (e.g., 'Bash(git:*)') to minimize the potential attack surface.
[Indirect Prompt Injection] (SAFE): The documentation explains how to interpolate file contents and user arguments into prompts. While this describes an ingestion surface for untrusted data, the skill is educational and does not implement any vulnerable logic.
[Credentials Unsafe] (SAFE): No hardcoded secrets, API keys, or private configuration files were detected. Examples correctly use placeholders for sensitive parameters.

Command Development