The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several standard CLI utilities to perform static analysis on the local codebase. This includes tools like grep and find for discovery, as well as pydeps and madge for dependency graph visualization and circular dependency detection. These commands are scoped to the project directory and used for read-only analysis.
[EXTERNAL_DOWNLOADS]: The documentation provides links to external GitHub repositories (e.g., FPF methodology and Quint-Code) as references for the underlying frameworks. No automated downloads or execution of remote scripts from these sources are initiated by the skill.
[SAFE]: The skill incorporates specific 'AI Assistance Guardrails' that instruct the agent to provide evidence-based critiques using file paths and line numbers. This helps prevent hallucinations and ensures the agent's output is grounded in the actual codebase being reviewed.

architecture-review