agents-core/ai.koog.agents.core.agent.config/AIAgentConfig/Companion/AIAgentConfigBuilder

AIAgentConfigBuilder

class AIAgentConfigBuilder(val model: LLModel, var prompt: Prompt? = null, var maxAgentIterations: Int? = null, var missingToolsConversionStrategy: MissingToolsConversionStrategy? = null, var responseProcessor: ResponseProcessor? = null, strategyExecutorService: ERROR CLASS: Symbol not found for ExecutorService?? = null, llmRequestExecutorService: ERROR CLASS: Symbol not found for ExecutorService?? = null, serializer: JSONSerializer = JacksonSerializer())(source)

A builder class for constructing an instance of AIAgentConfig with customizable configuration options.

Constructors

constructor(model: LLModel, prompt: Prompt? = null, maxAgentIterations: Int? = null, missingToolsConversionStrategy: MissingToolsConversionStrategy? = null, responseProcessor: ResponseProcessor? = null, strategyExecutorService: ERROR CLASS: Symbol not found for ExecutorService?? = null, llmRequestExecutorService: ERROR CLASS: Symbol not found for ExecutorService?? = null, serializer: JSONSerializer = JacksonSerializer())

Properties

maxAgentIterations

var maxAgentIterations: Int?

missingToolsConversionStrategy

var missingToolsConversionStrategy: MissingToolsConversionStrategy?

model

val model: LLModel

prompt

var prompt: Prompt?

responseProcessor

var responseProcessor: ResponseProcessor?

Functions

build

fun build(): AIAgentConfig

Constructs and returns an instance of AIAgentConfig using the values configured in the builder. The method validates that all required fields are provided and assigns default values to optional fields if they are not explicitly set.

llmRequestExecutorService

fun llmRequestExecutorService(executor: ERROR CLASS: Symbol not found for ExecutorService??): AIAgentConfig.Companion.AIAgentConfigBuilder

Sets the executor service for handling LLM (Language Learning Model) requests.

maxAgentIterations

fun maxAgentIterations(maxAgentIterations: Int): AIAgentConfig.Companion.AIAgentConfigBuilder

Sets the maximum number of iterations allowed for the AI agent during its execution.

missingToolsConversionStrategy

fun missingToolsConversionStrategy(strategy: MissingToolsConversionStrategy): AIAgentConfig.Companion.AIAgentConfigBuilder

Configures the strategy to handle missing tool definitions in prompts.

prompt

fun prompt(prompt: Prompt): AIAgentConfig.Companion.AIAgentConfigBuilder

Sets the prompt configuration for the AI agent.

responseProcessor

fun responseProcessor(processor: ResponseProcessor?): AIAgentConfig.Companion.AIAgentConfigBuilder

Assigns a custom response processor to the configuration builder. The response processor is responsible for processing and transforming the responses generated by the language model.

serializer

fun serializer(serializer: JSONSerializer): AIAgentConfig.Companion.AIAgentConfigBuilder

Sets serializr for underlying tool calls and LLM requests

strategyExecutorService

fun strategyExecutorService(executor: ERROR CLASS: Symbol not found for ExecutorService??): AIAgentConfig.Companion.AIAgentConfigBuilder

Sets the executor service to be used for executing strategies within the agent configuration.