The Agent Skills Directory

[SAFE]: The skill instructions are focused exclusively on providing configuration guidance and testing steps for the ZenMux platform.
[COMMAND_EXECUTION]: Provides example curl commands and shell export instructions. These are standard configuration steps for API integration and target the vendor's own verified infrastructure (zenmux.ai).
[CREDENTIALS_UNSAFE]: References API keys and tokens (e.g., ANTHROPIC_AUTH_TOKEN, sk-ai-v1-) exclusively through placeholders or format descriptions for educational purposes. It follows industry-standard safe practices by advising users to store these in shell profiles or tool-specific settings rather than hardcoding them.
[PERSISTENCE]: Recommends adding configuration to shell profiles (~/.zshrc, ~/.bashrc), which is standard procedure for setting persistent environment variables in development environments and does not constitute a malicious persistence mechanism.
[DATA_EXFILTRATION]: No evidence of unauthorized data transfer. Network operations (via curl) are directed at the vendor's official API endpoints for verification purposes and are intended to be initiated by the user.

zenmux-setup