The Agent Skills Directory

[SAFE]: The skill serves as a comprehensive style guide and security framework for designing system prompts within a specific technical ecosystem. It does not contain any executable scripts or malicious command execution patterns.- [SAFE]: The skill explicitly promotes secure handling of credentials by instructing the use of a dedicated secrets management tool (joelclaw secrets) and prohibiting the storage of secrets in logs or version-controlled files.- [SAFE]: It establishes clear boundaries for AI behavior, such as preventing the fabrication of a specific person's voice or opinions and requiring factual sourcing for metrics and anecdotes.- [SAFE]: It mandates the use of internal inference abstractions (pi) and discourages direct access to authentication files or third-party APIs, reducing the risk of credential leakage.

system-prompt