The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: Fetches and executes the mermint CLI directly from the author's GitHub repository using npx. This allows the skill to render Mermaid diagrams from local markdown files.
[COMMAND_EXECUTION]: Executes shell commands including mkdir, rsvg-convert, magick, rg, and grep to manage files, generate image previews, and verify processing results.
[EXTERNAL_DOWNLOADS]: Downloads the mermint package and potentially Playwright browser binaries from GitHub and standard registries during execution.
[PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests and processes untrusted markdown files. * Ingestion points: Reads Mermaid code blocks from markdown files (e.g., README.md) which could contain malicious instructions. * Boundary markers: No specific delimiters or instructions are used to prevent the agent from following instructions embedded within the processed diagrams. * Capability inventory: The skill can execute remote code via npx, run shell commands for image processing, and perform file system operations. * Sanitization: No explicit sanitization or validation of the input markdown content is performed before processing.

mermint-markdown-workflow