The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input to generate diagram source code and rendered assets, creating a surface for potential injection.
Ingestion points: User prompts requesting visualizations, technical architecture descriptions, or process flows (SKILL.md).
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within user-provided content.
Capability inventory: The skill possesses file modification capabilities (Write, Edit) and integrates with artifact-publish to host rendered diagrams as Claude artifacts.
Sanitization: No security-specific sanitization instructions are provided for user input; guidance focus is limited to escaping special characters for syntax rendering.
[EXTERNAL_DOWNLOADS]: The documentation references the installation and use of the @mermaid-js/mermaid-cli utility.
Details: Instructions include the use of npm install and npx to fetch the official Mermaid CLI from the public npm registry for diagram rendering (references/advanced-patterns.md, references/complexity-and-output.md).
[REMOTE_CODE_EXECUTION]: The skill provides patterns for executing the Mermaid CLI tool at runtime.
Details: The documentation describes executing mmdc (Mermaid CLI) and npx commands to transform diagram definitions into images or PDF files.

mermaid-charts