The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional markdown text without any executable scripts, subprocess calls, or shell commands.
[SAFE]: No sensitive data exposure or exfiltration patterns were detected. The instructions do not involve accessing local configuration files, environment variables, or private credentials.
[SAFE]: No obfuscation techniques, prompt injection attempts, or persistence mechanisms were identified within the content.
[SAFE]: References to external libraries like LiteLLM and specific AI models from established organizations (OpenAI, Anthropic, Microsoft, BAAI) are relevant to the skill's purpose and do not represent security risks.

model-comparator