The Agent Skills Directory

[SAFE]: The skill instructions and reference files promote secure coding practices, specifically emphasizing Resource Acquisition Is Initialization (RAII), the use of smart pointers over raw memory management, and the integration of memory sanitizers (AddressSanitizer, UndefinedBehaviorSanitizer) to prevent vulnerabilities.
[SAFE]: Remote resources referenced in the build system configurations (e.g., fmtlib/fmt via GitHub) are well-known, standard libraries used in the C++ ecosystem. The use of CMake's FetchContent for these libraries is a standard and safe development practice.
[SAFE]: The skill uses standard package managers (Conan) and CI/CD workflows (GitHub Actions) as examples, following professional development standards with no evidence of malicious command injection or persistence mechanisms.
[SAFE]: No signs of prompt injection, obfuscation, data exfiltration, or unauthorized privilege escalation were detected across the skill body or reference documentation.

cpp-pro