The Agent Skills Directory

[SAFE]: The skill serves as an advisory framework for security auditing and does not perform any active or dangerous operations.
[SAFE]: All instructions and code examples provided in audit-reference.md are aligned with industry-standard secure coding practices, specifically focusing on Micronaut and Kotlin.
[SAFE]: The allowed_tools configuration is restricted to passive file system operations (Read, Glob, Grep), which are appropriate for the skill's stated purpose of auditing code.
[SAFE]: No external network requests, remote script downloads, or dynamic code execution patterns are present in the provided files.
[SAFE]: There is no evidence of obfuscation, hardcoded credentials, or attempts to bypass agent safety guidelines.

security-checklist