The Agent Skills Directory

[SAFE]: The skill instructions define legitimate diagramming capabilities using the platform's show_widget and read_me tools without any unauthorized access to files or network resources.- [SAFE]: Safety measures are included in the instructions, such as prohibiting the inclusion of root HTML tags (, , etc.) in rendered widgets, which mitigates potential visualization-based attacks.- [SAFE]: The skill uses static reference files for templates and syntax guides, showing no signs of obfuscation, remote code execution, or credential theft patterns.- [SAFE]: The processing of user input for diagram generation is handled with specific structural templates and requirements, adhering to the primary purpose of the skill without introducing dangerous autonomy or privilege escalation.

diagram-builder