The Agent Skills Directory

[SAFE]: The skill's primary function is to gather context from local LLM configuration files (e.g., vLLM, TensorRT-LLM) and provide optimization advice. This behavior is consistent with its stated purpose and does not involve any dangerous system operations.
[SAFE]: No patterns associated with data exfiltration, credential theft, or remote code execution were found. The skill does not use network-capable tools or shell execution commands.
[SAFE]: The use of the llm-optimization-advisor agent is a standard multi-agent orchestration pattern for specialized analysis and does not introduce security risks in this context.
[SAFE]: The skill includes an inherent attack surface for indirect prompt injection because it reads external content (configuration files and inference scripts). However, because the skill lacks the ability to execute code or transmit data externally, this risk is categorized as safe and represents standard operational behavior for a static analysis tool.

optimize-llm