The Agent Skills Directory

[COMMAND_EXECUTION]: The Python utility scripts (extract_mermaid.py, mermaid_to_image.py, and resilient_diagram.py) execute the Mermaid CLI tool (mmdc) via subprocess.run. The implementation follows security best practices by passing arguments as a list rather than a single shell string, which prevents shell injection vulnerabilities.
[EXTERNAL_DOWNLOADS]: The skill's documentation guides users to install the Mermaid CLI dependency via NPM (@mermaid-js/mermaid-cli) and provides installation instructions via GitHub. These sources are established, well-known services within the developer ecosystem and are treated as safe for the purposes of this skill.
[INDIRECT_PROMPT_INJECTION]: The skill features a data ingestion surface where it processes user-provided markdown files and Mermaid code blocks. However, the risk is low as the skill's logic is restricted to extracting text and passing it to an external renderer. It uses clear boundary markers (code blocks) and does not interpret the content as instructions for the agent's core behavioral logic.

mermaid-creator