The Agent Skills Directory

Local Command Execution Utility: The skill includes a Node.js utility, render-graphs.js, which utilizes child_process.execSync to invoke the system's Graphviz dot command.
This script is designed to extract DOT language blocks from SKILL.md and render them into SVG files for visualization.
While the execution of shell commands is a sensitive operation, this utility is provided as an optional tool for documentation rendering and operates on local files.
Filesystem Interaction: The provided utility script performs read and write operations within the local skill directory.
It reads the content of markdown files and writes generated SVG output to a diagrams/ subdirectory.
This behavior is localized and consistent with the skill's purpose of managing and visualizing documentation.
Educational Content on Agent Compliance: Several files, such as persuasion-principles.md and testing-skills-with-subagents.md, provide guidance on using persuasive language to ensure AI agents adhere to development standards.
These techniques are used to improve the reliability of agent behavior during complex tasks like TDD and are not intended to bypass system safety guidelines.

writing-skills