The Agent Skills Directory

[COMMAND_EXECUTION]: The skill generates and executes bash scripts that interact with the local filesystem (reading/writing .env files) and external CLI tools (GitHub gh CLI). These operations are the primary function of the skill and are performed with explicit user confirmation and interaction.
[CREDENTIALS_SAFE]: The template.sh provides a ask_secret function that uses read -rs to hide terminal input when sensitive values are entered. It also correctly distinguishes between local .env storage and CI storage via gh secret set.
[EXTERNAL_DOWNLOADS]: The skill uses open_url to direct users to legitimate service dashboards (like Stripe) in their local browser. This is a safe interaction pattern as it relies on the user's existing authenticated browser session and does not download or execute remote scripts.

wizard