The Agent Skills Directory

External Tooling Dependency: The skill references the @mermaid-js/mermaid-cli package for diagram rendering. It explicitly instructs the agent to request user approval before installing the tool, which is a standard safety practice for managing environment dependencies.
Command-Line Execution: The skill describes using the mmdc command to render or validate diagrams. This is an expected functional component for a skill designed to produce visual architecture documentation.
Data Processing Considerations: The skill analyzes external code and documentation to construct diagrams (Category 8). Ingestion points: System code and documentation files identified in SKILL.md. Boundary markers: The skill requires the agent to ask the user for clarification when documentation and implementation disagree or when classifications are plausible, as noted in c4-modelling-instructions.md. Capability inventory: The skill can execute the mmdc CLI tool for rendering. Sanitization: The skill relies on manual user review and 'modelling uncertainty gates' rather than automated code sanitization, which is consistent with its intended use as a documentation assistant.

c4-architecture