The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a utility script render-graphs.js that executes system commands (dot and which) via the Node.js child_process module to generate SVG diagrams from DOT blocks in documentation. This is an intended local development tool for visualization.- [PROMPT_INJECTION]: The documentation provides guidelines on using imperative framing and 'authority' principles (e.g., 'YOU MUST', 'Delete means delete') within skills to enforce compliance with engineering standards. These patterns are explicitly documented as a reliability framework to prevent agent rationalization during task execution rather than as a means to bypass safety protocols.

writing-skills