The Agent Skills Directory

[SAFE]: The skill utilizes platform-level configuration (allowed-tools) to restrict its execution environment to specific, pre-defined npm commands (npm run depcruise and npm run depgraph). This practice significantly reduces the attack surface for command injection.
[SAFE]: Analysis of the instructions confirms no attempts at prompt injection, obfuscation, or persistence. The skill strictly operates on local repository metadata to generate reports without external network dependencies.

architecture-guard