prompt-executor-bedrock-client/ai.koog.prompt.executor.clients.bedrock

Package-level declarations

Types

class BedrockClientSettings(region: String = BedrockRegions.US_WEST_2.regionCode, timeoutConfig: ConnectionTimeoutConfig = ConnectionTimeoutConfig(), endpointUrl: String? = null, maxRetries: Int = 3, enableLogging: Boolean = false, moderationGuardrailsSettings: BedrockGuardrailsSettings? = null)

Configuration settings for connecting to the AWS Bedrock API.

BedrockGuardrailsSettings

class BedrockGuardrailsSettings(guardrailIdentifier: String, guardrailVersion: String)

Represents the settings configuration for Bedrock guardrails.

BedrockInferencePrefixes

@Serializable

enum BedrockInferencePrefixes : Enum<BedrockInferencePrefixes>

AWS Bedrock inference ID prefixes.

BedrockLLMClient

class BedrockLLMClient(bedrockClient: BedrockRuntimeClient, moderationGuardrailsSettings: BedrockGuardrailsSettings? = null, clock: Clock = Clock.System) : LLMClient

Creates a new Bedrock LLM client configured with the specified AWS credentials and settings.

BedrockModel

@Serializable

data class BedrockModel(val model: LLModel, val modelId: String = model.id, val inferenceProfilePrefix: String = BedrockInferencePrefixes.US.prefix)

Represents a Bedrock model with an optional inference profile prefix.

BedrockModelFamilies

@Serializable

sealed class BedrockModelFamilies

Represents a sealed class for AWS Bedrock model families.

BedrockModels

object BedrockModels : LLModelDefinitions

Bedrock models Models available through the AWS Bedrock API

BedrockRegions

@Serializable

enum BedrockRegions : Enum<BedrockRegions>

AWS Bedrock regions.

Functions

withInferenceProfile

fun LLModel.withInferenceProfile(inferencePrefix: String): LLModel

Extension method to create a new LLModel with a different inference profile prefix. This allows you to easily switch inference profiles for predefined Bedrock models.