The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documentation includes instructions to run local Python scripts (validate_env.py, scan_exposed.py, sync_secrets.py) which are intended to automate the validation of environment variables and synchronization with external platforms like Vercel.
[CREDENTIALS_UNSAFE]: The guide contains examples of environment variable keys such as JWT_SECRET, DATABASE_URL, and STRIPE_SECRET_KEY. These are populated with educational placeholders (e.g., your-jwt-secret) rather than real credentials, and the content explicitly provides security warnings against committing secrets to version control or exposing them via the NEXT_PUBLIC_ prefix.

nextjs