The Agent Skills Directory

[SAFE]: The skill performs legitimate configuration management for model roles and budgets.
[SAFE]: File access is restricted to tool-specific configuration paths in the user's home directory (~/.cursor/rules/ and ~/.agents/), which is standard practice for per-user settings.
[SAFE]: Input validation is performed to ensure only detected or available model slugs are written to configuration files, preventing the injection of invalid or unauthorized identifiers.
[SAFE]: No network operations, obfuscation, or unauthorized command execution patterns were identified.

setup-pstack