The LLMResponse object is a core component of the response returned by an agent through the .run method. It encapsulates the text generated by the LLM and potentially additional information about the agent’s processing.

​
Props

type
"text" | "image" | "object"

This field indicates the type of response generated by the LLM. Currently, Only "text" is being used, signifying a text-based response. However, there’s a plan for supporting the other response types in the future depending on the evolving of Scoopika and the capabilities of LLMs.

content
string

This field contains the actual text content generated by the LLM in response to the provided prompt and session context.

tools_history
ToolHistory array

This array holds information about any external tools the agent used during its processing. Each element within the array represents a tool call and its corresponding result. The array can be empty if the agent didn’t use any external tools for the specific run.

Reference

Was this page helpful?