The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the mmdc command-line utility for diagram rendering and validation, which is the primary intended functionality. It uses standard shell patterns for input/output and file creation.
[PROMPT_INJECTION]: The skill is designed to process external Mermaid diagram definitions and Markdown files. This ingestion of untrusted data constitutes a surface for indirect prompt injection, as the content is rendered by a headless browser engine.
Ingestion points: Input files and Markdown blocks processed via the -i flag.
Boundary markers: Not explicitly implemented for the input data content.
Capability inventory: Execution of mmdc, which manages the Puppeteer rendering process.
Sanitization: Relies on the internal parsing logic and security configurations (e.g., 'strict' mode) of the Mermaid CLI.
[EXTERNAL_DOWNLOADS]: The documentation references manual installation from official sources and notes that icon packs are downloaded from unpkg.com (a well-known content delivery network).

mermaid-cli