The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to transform user-provided descriptions into diagrams, which inherently creates a surface for indirect prompt injection.
Ingestion points: User requests to "visualize", "model", or "diagram" specific logic or architectures (SKILL.md).
Boundary markers: The instructions do not define specific delimiters or guidelines for the agent to distinguish between text meant for diagramming and potentially malicious instructions.
Capability inventory: The skill does not request access to any dangerous tools, scripts, or network capabilities, which mitigates the risk of an injection being used to perform unauthorized actions.
Sanitization: No input sanitization instructions are provided.
[EXTERNAL_DOWNLOADS]: The documentation contains references to official packages and tools for the Mermaid ecosystem.
The skill provides instructions for installing the official @mermaid-js/mermaid-cli via npm and using a Docker container for rendering (README.md).
References are made to official and well-known services including mermaid.live and mermaid.js.org for documentation and interactive editing (SKILL.md).
[NO_CODE]: The skill is composed entirely of Markdown documentation and does not ship with any executable Python, JavaScript, or shell scripts.

mermaid-diagrams