prompt-executor-model/ai.koog.prompt.executor.model/DynamicPromptExecutor

DynamicPromptExecutor

abstract class DynamicPromptExecutor : PromptExecutor(source)

A PromptExecutor that introduces an explicit resolveModel step before every LLM operation.

Prefer this base class for new custom prompt executors that need fallback, routing, model substitution, or any other model-resolution policy.

Subclasses use resolveModel to translate a caller-requested LLModel into a ResolvedModel — for example, to swap in a fallback model when the requested one is not directly available. Each PromptExecutor entry point taking a plain LLModel is finalized here and delegates to the matching ResolvedModel-based overload, so model resolution always happens exactly once per call and subclasses cannot bypass it.

Subclasses must implement resolveModel together with the ResolvedModel-based execute, executeStreaming, executeMultipleChoices, and moderate overloads.

Inside those overrides, do not delegate to the LLModel-based overloads. They are finalized to re-enter model resolution and will recurse.

Constructors

DynamicPromptExecutor

constructor()

Functions

expect abstract fun close()

execute

abstract suspend override fun execute(prompt: Prompt, resolvedModel: ResolvedModel, tools: List<ToolDescriptor> = emptyList()): Message.Assistant

Executes prompt against an already-resolved resolvedModel.

suspend override fun execute(prompt: Prompt, model: LLModel, tools: List<ToolDescriptor> = emptyList()): Message.Assistant

Finalized to enforce that every call resolves through resolveModel before dispatch. Subclasses customize behavior by overriding the ResolvedModel-based execute overload.

executeMultipleChoices

abstract suspend override fun executeMultipleChoices(prompt: Prompt, resolvedModel: ResolvedModel, tools: List<ToolDescriptor>): LLMChoice

Returns multiple independent choices from an already-resolved resolvedModel.

suspend override fun executeMultipleChoices(prompt: Prompt, model: LLModel, tools: List<ToolDescriptor>): LLMChoice

Finalized to enforce that every call resolves through resolveModel before dispatch. Subclasses customize behavior by overriding the ResolvedModel-based executeMultipleChoices overload.

executeStreaming

abstract override fun executeStreaming(prompt: Prompt, resolvedModel: ResolvedModel, tools: List<ToolDescriptor> = emptyList()): Flow<StreamFrame>

Streams the response for prompt against an already-resolved resolvedModel.

override fun executeStreaming(prompt: Prompt, model: LLModel, tools: List<ToolDescriptor> = emptyList()): Flow<StreamFrame>

Finalized to enforce that every call resolves through resolveModel before dispatch. Subclasses customize behavior by overriding the ResolvedModel-based executeStreaming overload.

executeStructured

suspend fun <T> PromptExecutor.executeStructured(prompt: Prompt, model: LLModel, config: StructuredRequestConfig<T>, fixingParser: StructureFixingParser? = null): Result<StructuredResponse<T>>

inline suspend fun <T> PromptExecutor.executeStructured(prompt: Prompt, model: LLModel, examples: List<T> = emptyList(), fixingParser: StructureFixingParser? = null): Result<StructuredResponse<T>>

suspend fun <T> PromptExecutor.executeStructured(prompt: Prompt, model: LLModel, serializer: KSerializer<T>, examples: List<T> = emptyList(), fixingParser: StructureFixingParser? = null): Result<StructuredResponse<T>>

Executes a prompt with structured output, enhancing it with schema instructions or native structured output parameter, and parses the response into the defined structure.

getBasicJsonSchemaGenerator

open fun getBasicJsonSchemaGenerator(model: LLModel): BasicJsonSchemaGenerator

Basic JSON schema generator required for the given model. Return BasicJsonSchemaGenerator by default.

getStandardJsonSchemaGenerator

open fun getStandardJsonSchemaGenerator(model: LLModel): StandardJsonSchemaGenerator

Standard JSON schema generator required for the given model. Return StandardJsonSchemaGenerator by default.

models

open suspend fun models(): List<LLModel>

Retrieves a list of available models from all LLM clients managed by this executor.

moderate

abstract suspend override fun moderate(prompt: Prompt, model: ResolvedModel): ModerationResult

Moderates prompt against an already-resolved model.

suspend override fun moderate(prompt: Prompt, model: LLModel): ModerationResult

Finalized to enforce that every call resolves through resolveModel before dispatch. Subclasses customize behavior by overriding the ResolvedModel-based moderate overload.

parseResponseToStructuredResponse

suspend fun <T> PromptExecutor.parseResponseToStructuredResponse(response: Message.Assistant, config: StructuredRequestConfig<T>, model: LLModel, fixingParser: StructureFixingParser? = null): StructuredResponse<T>

Parses a structured response from the assistant message using the provided structured output configuration and language model. If a fixing parser is specified in the configuration, it will be used; otherwise, the structure will be parsed directly.

resolveModel

abstract suspend override fun resolveModel(model: LLModel, promptExecutorOperation: PromptExecutorOperation): ResolvedModel

Resolves model into the ResolvedModel to actually use for the given promptExecutorOperation.