The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute curl commands for interacting with a LiteLLM proxy API. This behavior is standard for managing API-driven services and is restricted to the user-provided proxy URL.
[SAFE]: Sensitive information, such as the LITELLM_API_KEY and server credentials, is requested from the user at runtime. There are no hardcoded secrets or unauthorized file access patterns detected.
[SAFE]: Network activity is directed to the LiteLLM proxy defined by the user and references official LiteLLM documentation, which is expected for a skill authored by BerriAI to support their own platform.

add-mcp