The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands to run generated Python scripts (uv run <filename>.py) and render Mermaid diagrams (npx -y @mermaid-js/mermaid-cli). It also includes instructions for system-level package installation (e.g., sudo apt install graphviz).
[DYNAMIC_EXECUTION]: The skill constructs Python source code at runtime using user-provided information and executes it. Additionally, the scripts/find-node.py script employs importlib.import_module() to dynamically load and inspect modules within the installed diagrams library.
[EXTERNAL_DOWNLOADS]: The skill fetches the @mermaid-js/mermaid-cli package from the NPM registry at runtime and specifies diagrams as a dependency for its generated scripts.
[PROMPT_INJECTION]: The skill's workflow presents an indirect prompt injection surface (Category 8), as it processes untrusted user descriptions to generate and execute code.
Ingestion points: User requests and descriptions provided to the agent as input for diagram generation.
Boundary markers: No specific delimiters or safety instructions are defined to separate user input from the logic that generates the diagram source code.
Capability inventory: The agent has permissions to execute shell commands, perform network operations (via npx and npm), and read/write files.
Sanitization: No explicit sanitization or validation of the user-provided descriptions is performed before the description influences the content of the generated executable scripts.

create-diagram