The Agent Skills Directory

[SAFE]: The skill promotes secure development practices for LLM pipelines, such as using environment variables for credentials and implementing robust error handling and fallbacks.
[SAFE]: A static analysis script (verify.sh) is provided for local verification of code and configuration; the script is read-only and performs benign checks for identifying hardcoded keys and ensuring timeouts are set.
[SAFE]: External references are directed toward well-known libraries such as LiteLLM and services like Bifrost or Portkey, with a focus on reliability and cost management.

llm-pipeline