The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads and synchronizes itself from the official Anthropic GitHub repository as documented in the README.md.
[COMMAND_EXECUTION]: Several Python scripts (scripts/run_eval.py, eval-viewer/generate_review.py) use subprocess.Popen and subprocess.run to execute system commands such as claude -p for running evaluations or lsof for managing server ports. These are standard operational requirements for the skill's purpose.
[SAFE]: The code follows best practices for secret management by instructing users to use .env files for keys rather than hardcoding them. It also uses safe YAML loading via yaml.safe_load in scripts/quick_validate.py.

skill-creator