The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads the 'archify' engine from the author's own repository (Rheinmir/archify) using a skill installation command.
[COMMAND_EXECUTION]: The skill executes shell commands to perform its core functions:
Invokes node to run the archify.mjs script for diagram generation.
Runs python3 to execute internal validation tools such as fdk/tools/frontend-antipattern.py and fdk/tools/visual-receipt.py on the generated HTML files.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied descriptions and Mermaid syntax to generate diagrams, which represents a typical injection surface for agent skills. The skill attempts to mitigate risks through:
Ingestion points: User prompts and provided Mermaid code.
Boundary markers: None explicitly specified for prompt interpolation.
Capability inventory: The skill can execute shell commands (node, python3) and write to the local file system.
Sanitization: The skill relies on deterministic engines and post-generation validation scripts (frontend-antipattern.py) to detect and reject malformed or non-compliant output, including checks for remote references.

diagram