The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a Node.js script render-graphs.js that executes the system's dot command via child_process.execSync. This tool is used for rendering SVG diagrams from Graphviz markdown blocks provided in the skill documentation. The script also invokes which dot to check for the command's presence on the host system.
[EXTERNAL_DOWNLOADS]: Documentation and templates within the skill suggest installing several well-known third-party libraries (e.g., pdfplumber, pypdf, pdf2image, pytesseract, docx-js) using standard package managers like pip and npm. These recommendations are contextual examples for implementing specific types of tasks and rely on official package registries.
[SAFE]: The 'Anti-Rationalization' and 'Persuasion Principles' sections provide educational content on how to write authoritative and resistant instructions for AI agents. These techniques are intended for developers to ensure that agents follow critical workflows (such as Test-Driven Development) and do not represent malicious prompt injection or attempts to bypass system safety filters.

writing-skills