The Agent Skills Directory

[SAFE]: The skill defines a set of internal coding guidelines and standards for a monitoring agent (LoongCollector). All provided code snippets are benign and focus on metric naming, alarm level definitions, and proper initialization patterns in C++ and Go. There are no instances of prompt injection, data exfiltration, or unauthorized command execution. The instructions promote safe coding practices, such as using existing framework interfaces and implementing alarm rate-limiting to prevent 'alarm storms'.

selfmonitor