The Agent Skills Directory

[SAFE]: The skill implements a secure human-in-the-loop workflow. Instructions in SKILL.md explicitly mandate that the agent must propose changes and wait for user confirmation before using the Write or Edit tools.
[SAFE]: File system interactions are limited to standard utility tools like rg (ripgrep) and fd, as well as uv for Python environment management. These are well-known, industry-standard tools for the described tasks.
[SAFE]: No network activity, data exfiltration, or obfuscation techniques were detected in the skill instructions or templates.
[SAFE]: The skill uses localized templates within the templates/ directory to define its logic, avoiding external dependencies or remote script execution.

cmd-olshanskify