The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill's verification recipe utilizes npx to download and run the @mermaid-js/mermaid-cli package. This is a well-known and trusted tool provided by the official Mermaid project and is used here for its intended purpose of validating diagram syntax.
[COMMAND_EXECUTION]: The SKILL.md provides a bash-based verification recipe that uses standard shell utilities like awk and the Mermaid CLI. These commands are used to process and validate generated diagram code blocks.
[INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection as it ingests user-provided descriptions to generate code that is subsequently processed by a verification script.
Ingestion points: User-supplied diagram descriptions provided during conversation.
Boundary markers: The generated Mermaid code is correctly encapsulated within ```mermaid fenced blocks to separate it from other text.
Capability inventory: The skill instructs the agent to execute shell commands to verify the generated diagrams (SKILL.md).
Sanitization: The skill relies on the strict syntax parsing of the Mermaid CLI and does not implement additional sanitization for user-provided data.

mermaid