Interface IMicroAgenticaExecutor<Model>

Executor of the Micro Agentic AI.

IMicroAgenticaExecutor represents an executor of the MicroAgentica, composing its internal agents to accomplish the Agentic AI through the LLM (Large Language Model) function calling.

You can customize one of these intnernal agents by configuring properties of the IMicroAgenticaExecutor type, and assigning it to the IMicroAgenticaConfig.executor property.

Samchon

interface IMicroAgenticaExecutor<Model extends ILlmSchema.Model> {
    call: (
        ctx: MicroAgenticaContext<Model>,
    ) => Promise<AgenticaExecuteHistory<Model>[]>;
    describe:
        | null
        | boolean
        | (
            ctx: MicroAgenticaContext<Model>,
            executes: AgenticaExecuteHistory<Model>[],
        ) => Promise<void>;
}

Type Parameters

  • Model extends ILlmSchema.Model

Properties

Properties

call: (
    ctx: MicroAgenticaContext<Model>,
) => Promise<AgenticaExecuteHistory<Model>[]>

Function caller agent.

Call agent performs the LLM (Large Language Model) function calling from the candidate functions enrolled in the AgenticaContext.stack. And the scope of function calling is, not only just arguments filling, but also actual executing the function and returning the result.

By the way, conversation context with user can be not enough to filling the arguments of the candidate functions. In that case, the call agent will ask the user to fill the missing arguments.

Otherwise the cpnversation context is enough, so that succeeded to call some candidate functions, the call agent will step to the describe agent to explain the result of the function calling to the user as markdown content.

Type declaration

Recommend not to customize, due to its validation feedback strategy is working very well, and the call agent is the most general topic which can be universally applied to all domain fields.

describe:
    | null
    | boolean
    | (
        ctx: MicroAgenticaContext<Model>,
        executes: AgenticaExecuteHistory<Model>[],
    ) => Promise<void>

Describer agent of the function calling result.

Describe agent explains the results of the function callings to the user as markdown content.

If you configure this property as false or null, the describer agent never be used.

Context of the agent

List of function calling results

List of prompts generated by the describer