The Agent Skills Directory

[SAFE]: The skill acts as a meta-framework for skill development. It does not contain any malicious patterns, credentials, or unauthorized data access logic.
[COMMAND_EXECUTION]: The utility script render-graphs.js uses Node.js execSync to call the system dot command (Graphviz). This is a standard utility function for developers to visualize process flows described in the documentation. The script implements safe input handling by passing Graphviz content via stdin and validating filename characters via regex.
[PROMPT_INJECTION]: The documentation describes psychological persuasion principles (Authority, Commitment, Scarcity, etc.) intended to ensure agents follow strict engineering disciplines such as TDD. These instructions are designed to prevent agents from rationalizing away quality checks and do not target safety filters or system-level overrides.

writing-skills