OpenGradient

Use when the user wants to write code using the OpenGradient SDK, including LLM inference, chat completions, streaming, tool calling, on-chain model inference, LangChain agents, model hub operations, or digital twins. Also use when the user asks how to use a specific OpenGradient feature.

Use when the user wants to create, deploy, run, monitor, update, delete, or manage personal AI agents on the OpenGradient agent executor. Also use when the user asks about agent status, run history, scheduling agents, configuring MCP tool servers, or writing agent definition files.