The Agent Skills Directory

[SAFE]: The skill consists entirely of documentation and templates designed to guide developers and AI agents in building secure and maintainable software. It does not contain any executable scripts or binary files.- [COMMAND_EXECUTION]: The skill includes a settings.json.template that defines a secure execution environment for AI agents, specifically listing dangerous commands like sudo and piping remote scripts to bash in a deny block to prevent unauthorized actions.- [EXTERNAL_DOWNLOADS]: The documentation recommends the use of well-known and trusted technology packages (such as Next.js, React, Supabase, and OpenAI SDKs) from official package registries like npm.- [CREDENTIALS_UNSAFE]: The skill provides explicit guidance on managing secrets, instructing users to keep API keys and other credentials in environment variables and specifically forbidding their inclusion in version control.- [PROMPT_INJECTION]: The skill does not contain any instructions intended to override agent safety filters or bypass system prompts. It focuses on providing technical patterns and project structure guidance.

secret-sauce