Sign in Dashboard Get started
  1. Packages
  2. Packages
  3. AWS
  4. API Docs
  5. bedrock
  6. AgentcoreEvaluator
Viewing docs for AWS v7.39.0
published on Thursday, Jul 23, 2026 by Pulumi
Schema (JSON)
pulumi/pulumi-aws

aws.bedrock.AgentcoreEvaluator

aws logo
Viewing docs for AWS v7.39.0
published on Thursday, Jul 23, 2026 by Pulumi
Schema (JSON)
pulumi/pulumi-aws

    Manages an AWS Bedrock AgentCore Evaluator. An evaluator scores how an agent performs. You can configure it in one of two ways: an LLM-as-a-Judge evaluator that uses a model to score agent behavior against your instructions and a rating scale, or a code-based evaluator that runs a Lambda function you provide.

    Example Usage

    LLM-as-a-Judge with Numerical Rating Scale

    import * as pulumi from "@pulumi/pulumi";
import * as aws from "@pulumi/aws";

const example = new aws.bedrock.AgentcoreEvaluator("example", {
    evaluatorName: "helpfulness_evaluator",
    description: "Rates assistant helpfulness from 1 to 5",
    level: "TRACE",
    evaluatorConfig: {
        llmAsAJudge: {
            instructions: "Given the {context} and the {assistant_turn}, compare against {expected_response} and rate from 1 to 5.",
            ratingScale: {
                numericals: [
                    {
                        definition: "Not helpful at all.",
                        value: 1,
                        label: "1",
                    },
                    {
                        definition: "Extremely helpful.",
                        value: 5,
                        label: "5",
                    },
                ],
            },
            modelConfig: {
                bedrockEvaluatorModelConfig: {
                    modelId: "us.amazon.nova-2-lite-v1:0",
                    inferenceConfig: {
                        maxTokens: 1024,
                        temperature: 0,
                        topP: 1,
                    },
                },
            },
        },
    },
});

    import pulumi
import pulumi_aws as aws

example = aws.bedrock.AgentcoreEvaluator("example",
    evaluator_name="helpfulness_evaluator",
    description="Rates assistant helpfulness from 1 to 5",
    level="TRACE",
    evaluator_config={
        "llm_as_a_judge": {
            "instructions": "Given the {context} and the {assistant_turn}, compare against {expected_response} and rate from 1 to 5.",
            "rating_scale": {
                "numericals": [
                    {
                        "definition": "Not helpful at all.",
                        "value": float(1),
                        "label": "1",
                    },
                    {
                        "definition": "Extremely helpful.",
                        "value": float(5),
                        "label": "5",
                    },
                ],
            },
            "model_config": {
                "bedrock_evaluator_model_config": {
                    "model_id": "us.amazon.nova-2-lite-v1:0",
                    "inference_config": {
                        "max_tokens": 1024,
                        "temperature": float(0),
                        "top_p": float(1),
                    },
                },
            },
        },
    })

    package main

import (
	"github.com/pulumi/pulumi-aws/sdk/v7/go/aws/bedrock"
	"github.com/pulumi/pulumi/sdk/v3/go/pulumi"
)

func main() {
	pulumi.Run(func(ctx *pulumi.Context) error {
		_, err := bedrock.NewAgentcoreEvaluator(ctx, "example", &bedrock.AgentcoreEvaluatorArgs{
			EvaluatorName: pulumi.String("helpfulness_evaluator"),
			Description:   pulumi.String("Rates assistant helpfulness from 1 to 5"),
			Level:         pulumi.String("TRACE"),
			EvaluatorConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigArgs{
				LlmAsAJudge: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs{
					Instructions: pulumi.String("Given the {context} and the {assistant_turn}, compare against {expected_response} and rate from 1 to 5."),
					RatingScale: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs{
						Numericals: bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArray{
							&bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArgs{
								Definition: pulumi.String("Not helpful at all."),
								Value:      pulumi.Float64(1),
								Label:      pulumi.String("1"),
							},
							&bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArgs{
								Definition: pulumi.String("Extremely helpful."),
								Value:      pulumi.Float64(5),
								Label:      pulumi.String("5"),
							},
						},
					},
					ModelConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs{
						BedrockEvaluatorModelConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs{
							ModelId: pulumi.String("us.amazon.nova-2-lite-v1:0"),
							InferenceConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfigArgs{
								MaxTokens:   pulumi.Int(1024),
								Temperature: pulumi.Float64(0),
								TopP:        pulumi.Float64(1),
							},
						},
					},
				},
			},
		})
		if err != nil {
			return err
		}
		return nil
	})
}

    using System.Collections.Generic;
using System.Linq;
using Pulumi;
using Aws = Pulumi.Aws;

return await Deployment.RunAsync(() => 
{
    var example = new Aws.Bedrock.AgentcoreEvaluator("example", new()
    {
        EvaluatorName = "helpfulness_evaluator",
        Description = "Rates assistant helpfulness from 1 to 5",
        Level = "TRACE",
        EvaluatorConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigArgs
        {
            LlmAsAJudge = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs
            {
                Instructions = "Given the {context} and the {assistant_turn}, compare against {expected_response} and rate from 1 to 5.",
                RatingScale = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs
                {
                    Numericals = new[]
                    {
                        new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArgs
                        {
                            Definition = "Not helpful at all.",
                            Value = 1,
                            Label = "1",
                        },
                        new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArgs
                        {
                            Definition = "Extremely helpful.",
                            Value = 5,
                            Label = "5",
                        },
                    },
                },
                ModelConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs
                {
                    BedrockEvaluatorModelConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs
                    {
                        ModelId = "us.amazon.nova-2-lite-v1:0",
                        InferenceConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfigArgs
                        {
                            MaxTokens = 1024,
                            Temperature = 0,
                            TopP = 1,
                        },
                    },
                },
            },
        },
    });

});

    package generated_program;

import com.pulumi.Context;
import com.pulumi.Pulumi;
import com.pulumi.core.Output;
import com.pulumi.aws.bedrock.AgentcoreEvaluator;
import com.pulumi.aws.bedrock.AgentcoreEvaluatorArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfigArgs;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Map;
import java.io.File;
import java.nio.file.Files;
import java.nio.file.Paths;

public class App {
    public static void main(String[] args) {
        Pulumi.run(App::stack);
    }

    public static void stack(Context ctx) {
        var example = new AgentcoreEvaluator("example", AgentcoreEvaluatorArgs.builder()
            .evaluatorName("helpfulness_evaluator")
            .description("Rates assistant helpfulness from 1 to 5")
            .level("TRACE")
            .evaluatorConfig(AgentcoreEvaluatorEvaluatorConfigArgs.builder()
                .llmAsAJudge(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs.builder()
                    .instructions("Given the {context} and the {assistant_turn}, compare against {expected_response} and rate from 1 to 5.")
                    .ratingScale(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs.builder()
                        .numericals(                        
                            AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArgs.builder()
                                .definition("Not helpful at all.")
                                .value(1.0)
                                .label("1")
                                .build(),
                            AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArgs.builder()
                                .definition("Extremely helpful.")
                                .value(5.0)
                                .label("5")
                                .build())
                        .build())
                    .modelConfig(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs.builder()
                        .bedrockEvaluatorModelConfig(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs.builder()
                            .modelId("us.amazon.nova-2-lite-v1:0")
                            .inferenceConfig(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfigArgs.builder()
                                .maxTokens(1024)
                                .temperature(0.0)
                                .topP(1.0)
                                .build())
                            .build())
                        .build())
                    .build())
                .build())
            .build());

    }
}

    resources:
  example:
    type: aws:bedrock:AgentcoreEvaluator
    properties:
      evaluatorName: helpfulness_evaluator
      description: Rates assistant helpfulness from 1 to 5
      level: TRACE
      evaluatorConfig:
        llmAsAJudge:
          instructions: Given the {context} and the {assistant_turn}, compare against {expected_response} and rate from 1 to 5.
          ratingScale:
            numericals:
              - definition: Not helpful at all.
                value: 1
                label: '1'
              - definition: Extremely helpful.
                value: 5
                label: '5'
          modelConfig:
            bedrockEvaluatorModelConfig:
              modelId: us.amazon.nova-2-lite-v1:0
              inferenceConfig:
                maxTokens: 1024
                temperature: 0
                topP: 1

    pulumi {
  required_providers {
    aws = {
      source = "pulumi/aws"
    }
  }
}

resource "aws_bedrock_agentcoreevaluator" "example" {
  evaluator_name = "helpfulness_evaluator"
  description    = "Rates assistant helpfulness from 1 to 5"
  level          = "TRACE"
  evaluator_config = {
    llm_as_a_judge = {
      instructions = "Given the {context} and the {assistant_turn}, compare against {expected_response} and rate from 1 to 5."
      rating_scale = {
        numericals = [{
          "definition" = "Not helpful at all."
          "value"      = 1
          "label"      = "1"
          }, {
          "definition" = "Extremely helpful."
          "value"      = 5
          "label"      = "5"
        }]
      }
      model_config = {
        bedrock_evaluator_model_config = {
          model_id = "us.amazon.nova-2-lite-v1:0"
          inference_config = {
            max_tokens  = 1024
            temperature = 0
            top_p       = 1
          }
        }
      }
    }
  }
}

    LLM-as-a-Judge with Categorical Rating Scale

    import * as pulumi from "@pulumi/pulumi";
import * as aws from "@pulumi/aws";

const example = new aws.bedrock.AgentcoreEvaluator("example", {
    evaluatorName: "tone_evaluator",
    level: "SESSION",
    evaluatorConfig: {
        llmAsAJudge: {
            instructions: "Classify the tone of the {assistant_turn} given the {context}.",
            ratingScale: {
                categoricals: [
                    {
                        definition: "Friendly, helpful tone.",
                        label: "POSITIVE",
                    },
                    {
                        definition: "Neutral or terse tone.",
                        label: "NEUTRAL",
                    },
                    {
                        definition: "Unhelpful or dismissive tone.",
                        label: "NEGATIVE",
                    },
                ],
            },
            modelConfig: {
                bedrockEvaluatorModelConfig: {
                    modelId: "us.amazon.nova-2-lite-v1:0",
                },
            },
        },
    },
});

    import pulumi
import pulumi_aws as aws

example = aws.bedrock.AgentcoreEvaluator("example",
    evaluator_name="tone_evaluator",
    level="SESSION",
    evaluator_config={
        "llm_as_a_judge": {
            "instructions": "Classify the tone of the {assistant_turn} given the {context}.",
            "rating_scale": {
                "categoricals": [
                    {
                        "definition": "Friendly, helpful tone.",
                        "label": "POSITIVE",
                    },
                    {
                        "definition": "Neutral or terse tone.",
                        "label": "NEUTRAL",
                    },
                    {
                        "definition": "Unhelpful or dismissive tone.",
                        "label": "NEGATIVE",
                    },
                ],
            },
            "model_config": {
                "bedrock_evaluator_model_config": {
                    "model_id": "us.amazon.nova-2-lite-v1:0",
                },
            },
        },
    })

    package main

import (
	"github.com/pulumi/pulumi-aws/sdk/v7/go/aws/bedrock"
	"github.com/pulumi/pulumi/sdk/v3/go/pulumi"
)

func main() {
	pulumi.Run(func(ctx *pulumi.Context) error {
		_, err := bedrock.NewAgentcoreEvaluator(ctx, "example", &bedrock.AgentcoreEvaluatorArgs{
			EvaluatorName: pulumi.String("tone_evaluator"),
			Level:         pulumi.String("SESSION"),
			EvaluatorConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigArgs{
				LlmAsAJudge: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs{
					Instructions: pulumi.String("Classify the tone of the {assistant_turn} given the {context}."),
					RatingScale: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs{
						Categoricals: bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArray{
							&bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs{
								Definition: pulumi.String("Friendly, helpful tone."),
								Label:      pulumi.String("POSITIVE"),
							},
							&bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs{
								Definition: pulumi.String("Neutral or terse tone."),
								Label:      pulumi.String("NEUTRAL"),
							},
							&bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs{
								Definition: pulumi.String("Unhelpful or dismissive tone."),
								Label:      pulumi.String("NEGATIVE"),
							},
						},
					},
					ModelConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs{
						BedrockEvaluatorModelConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs{
							ModelId: pulumi.String("us.amazon.nova-2-lite-v1:0"),
						},
					},
				},
			},
		})
		if err != nil {
			return err
		}
		return nil
	})
}

    using System.Collections.Generic;
using System.Linq;
using Pulumi;
using Aws = Pulumi.Aws;

return await Deployment.RunAsync(() => 
{
    var example = new Aws.Bedrock.AgentcoreEvaluator("example", new()
    {
        EvaluatorName = "tone_evaluator",
        Level = "SESSION",
        EvaluatorConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigArgs
        {
            LlmAsAJudge = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs
            {
                Instructions = "Classify the tone of the {assistant_turn} given the {context}.",
                RatingScale = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs
                {
                    Categoricals = new[]
                    {
                        new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs
                        {
                            Definition = "Friendly, helpful tone.",
                            Label = "POSITIVE",
                        },
                        new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs
                        {
                            Definition = "Neutral or terse tone.",
                            Label = "NEUTRAL",
                        },
                        new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs
                        {
                            Definition = "Unhelpful or dismissive tone.",
                            Label = "NEGATIVE",
                        },
                    },
                },
                ModelConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs
                {
                    BedrockEvaluatorModelConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs
                    {
                        ModelId = "us.amazon.nova-2-lite-v1:0",
                    },
                },
            },
        },
    });

});

    package generated_program;

import com.pulumi.Context;
import com.pulumi.Pulumi;
import com.pulumi.core.Output;
import com.pulumi.aws.bedrock.AgentcoreEvaluator;
import com.pulumi.aws.bedrock.AgentcoreEvaluatorArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Map;
import java.io.File;
import java.nio.file.Files;
import java.nio.file.Paths;

public class App {
    public static void main(String[] args) {
        Pulumi.run(App::stack);
    }

    public static void stack(Context ctx) {
        var example = new AgentcoreEvaluator("example", AgentcoreEvaluatorArgs.builder()
            .evaluatorName("tone_evaluator")
            .level("SESSION")
            .evaluatorConfig(AgentcoreEvaluatorEvaluatorConfigArgs.builder()
                .llmAsAJudge(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs.builder()
                    .instructions("Classify the tone of the {assistant_turn} given the {context}.")
                    .ratingScale(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs.builder()
                        .categoricals(                        
                            AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs.builder()
                                .definition("Friendly, helpful tone.")
                                .label("POSITIVE")
                                .build(),
                            AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs.builder()
                                .definition("Neutral or terse tone.")
                                .label("NEUTRAL")
                                .build(),
                            AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs.builder()
                                .definition("Unhelpful or dismissive tone.")
                                .label("NEGATIVE")
                                .build())
                        .build())
                    .modelConfig(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs.builder()
                        .bedrockEvaluatorModelConfig(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs.builder()
                            .modelId("us.amazon.nova-2-lite-v1:0")
                            .build())
                        .build())
                    .build())
                .build())
            .build());

    }
}

    resources:
  example:
    type: aws:bedrock:AgentcoreEvaluator
    properties:
      evaluatorName: tone_evaluator
      level: SESSION
      evaluatorConfig:
        llmAsAJudge:
          instructions: Classify the tone of the {assistant_turn} given the {context}.
          ratingScale:
            categoricals:
              - definition: Friendly, helpful tone.
                label: POSITIVE
              - definition: Neutral or terse tone.
                label: NEUTRAL
              - definition: Unhelpful or dismissive tone.
                label: NEGATIVE
          modelConfig:
            bedrockEvaluatorModelConfig:
              modelId: us.amazon.nova-2-lite-v1:0

    pulumi {
  required_providers {
    aws = {
      source = "pulumi/aws"
    }
  }
}

resource "aws_bedrock_agentcoreevaluator" "example" {
  evaluator_name = "tone_evaluator"
  level          = "SESSION"
  evaluator_config = {
    llm_as_a_judge = {
      instructions = "Classify the tone of the {assistant_turn} given the {context}."
      rating_scale = {
        categoricals = [{
          "definition" = "Friendly, helpful tone."
          "label"      = "POSITIVE"
          }, {
          "definition" = "Neutral or terse tone."
          "label"      = "NEUTRAL"
          }, {
          "definition" = "Unhelpful or dismissive tone."
          "label"      = "NEGATIVE"
        }]
      }
      model_config = {
        bedrock_evaluator_model_config = {
          model_id = "us.amazon.nova-2-lite-v1:0"
        }
      }
    }
  }
}

    Code-based Evaluator (Lambda)

    import * as pulumi from "@pulumi/pulumi";
import * as aws from "@pulumi/aws";

const example = new aws.bedrock.AgentcoreEvaluator("example", {
    evaluatorName: "lambda_evaluator",
    level: "TOOL_CALL",
    evaluatorConfig: {
        codeBased: {
            lambdaConfig: {
                lambdaArn: exampleAwsLambdaFunction.arn,
                lambdaTimeoutInSeconds: 60,
            },
        },
    },
});

    import pulumi
import pulumi_aws as aws

example = aws.bedrock.AgentcoreEvaluator("example",
    evaluator_name="lambda_evaluator",
    level="TOOL_CALL",
    evaluator_config={
        "code_based": {
            "lambda_config": {
                "lambda_arn": example_aws_lambda_function["arn"],
                "lambda_timeout_in_seconds": 60,
            },
        },
    })

    package main

import (
	"github.com/pulumi/pulumi-aws/sdk/v7/go/aws/bedrock"
	"github.com/pulumi/pulumi/sdk/v3/go/pulumi"
)

func main() {
	pulumi.Run(func(ctx *pulumi.Context) error {
		_, err := bedrock.NewAgentcoreEvaluator(ctx, "example", &bedrock.AgentcoreEvaluatorArgs{
			EvaluatorName: pulumi.String("lambda_evaluator"),
			Level:         pulumi.String("TOOL_CALL"),
			EvaluatorConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigArgs{
				CodeBased: &bedrock.AgentcoreEvaluatorEvaluatorConfigCodeBasedArgs{
					LambdaConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfigArgs{
						LambdaArn:              pulumi.Any(exampleAwsLambdaFunction.Arn),
						LambdaTimeoutInSeconds: pulumi.Int(60),
					},
				},
			},
		})
		if err != nil {
			return err
		}
		return nil
	})
}

    using System.Collections.Generic;
using System.Linq;
using Pulumi;
using Aws = Pulumi.Aws;

return await Deployment.RunAsync(() => 
{
    var example = new Aws.Bedrock.AgentcoreEvaluator("example", new()
    {
        EvaluatorName = "lambda_evaluator",
        Level = "TOOL_CALL",
        EvaluatorConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigArgs
        {
            CodeBased = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigCodeBasedArgs
            {
                LambdaConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfigArgs
                {
                    LambdaArn = exampleAwsLambdaFunction.Arn,
                    LambdaTimeoutInSeconds = 60,
                },
            },
        },
    });

});

    package generated_program;

import com.pulumi.Context;
import com.pulumi.Pulumi;
import com.pulumi.core.Output;
import com.pulumi.aws.bedrock.AgentcoreEvaluator;
import com.pulumi.aws.bedrock.AgentcoreEvaluatorArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigCodeBasedArgs;
import com.pulumi.aws.bedrock.inputs.AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfigArgs;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Map;
import java.io.File;
import java.nio.file.Files;
import java.nio.file.Paths;

public class App {
    public static void main(String[] args) {
        Pulumi.run(App::stack);
    }

    public static void stack(Context ctx) {
        var example = new AgentcoreEvaluator("example", AgentcoreEvaluatorArgs.builder()
            .evaluatorName("lambda_evaluator")
            .level("TOOL_CALL")
            .evaluatorConfig(AgentcoreEvaluatorEvaluatorConfigArgs.builder()
                .codeBased(AgentcoreEvaluatorEvaluatorConfigCodeBasedArgs.builder()
                    .lambdaConfig(AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfigArgs.builder()
                        .lambdaArn(exampleAwsLambdaFunction.arn())
                        .lambdaTimeoutInSeconds(60)
                        .build())
                    .build())
                .build())
            .build());

    }
}

    resources:
  example:
    type: aws:bedrock:AgentcoreEvaluator
    properties:
      evaluatorName: lambda_evaluator
      level: TOOL_CALL
      evaluatorConfig:
        codeBased:
          lambdaConfig:
            lambdaArn: ${exampleAwsLambdaFunction.arn}
            lambdaTimeoutInSeconds: 60

    pulumi {
  required_providers {
    aws = {
      source = "pulumi/aws"
    }
  }
}

resource "aws_bedrock_agentcoreevaluator" "example" {
  evaluator_name = "lambda_evaluator"
  level          = "TOOL_CALL"
  evaluator_config = {
    code_based = {
      lambda_config = {
        lambda_arn                = exampleAwsLambdaFunction.arn
        lambda_timeout_in_seconds = 60
      }
    }
  }
}

    Create AgentcoreEvaluator Resource

    Resources are created with functions called constructors. To learn more about declaring and configuring resources, see Resources.

    Constructor syntax

    new AgentcoreEvaluator(name: string, args: AgentcoreEvaluatorArgs, opts?: CustomResourceOptions);
    @overload
def AgentcoreEvaluator(resource_name: str,
                       args: AgentcoreEvaluatorArgs,
                       opts: Optional[ResourceOptions] = None)

@overload
def AgentcoreEvaluator(resource_name: str,
                       opts: Optional[ResourceOptions] = None,
                       evaluator_config: Optional[AgentcoreEvaluatorEvaluatorConfigArgs] = None,
                       evaluator_name: Optional[str] = None,
                       level: Optional[str] = None,
                       description: Optional[str] = None,
                       kms_key_arn: Optional[str] = None,
                       region: Optional[str] = None,
                       tags: Optional[Mapping[str, str]] = None,
                       timeouts: Optional[AgentcoreEvaluatorTimeoutsArgs] = None)
    func NewAgentcoreEvaluator(ctx *Context, name string, args AgentcoreEvaluatorArgs, opts ...ResourceOption) (*AgentcoreEvaluator, error)
    public AgentcoreEvaluator(string name, AgentcoreEvaluatorArgs args, CustomResourceOptions? opts = null)
    public AgentcoreEvaluator(String name, AgentcoreEvaluatorArgs args)
public AgentcoreEvaluator(String name, AgentcoreEvaluatorArgs args, CustomResourceOptions options)

    type: aws:bedrock:AgentcoreEvaluator
properties: # The arguments to resource properties.
options: # Bag of options to control resource's behavior.


    resource "aws_bedrock_agentcore_evaluator" "name" {
    # resource properties
}

    Parameters

    name string
    The unique name of the resource.
    args AgentcoreEvaluatorArgs
    The arguments to resource properties.
    opts CustomResourceOptions
    Bag of options to control resource's behavior.
    resource_name str
    The unique name of the resource.
    args AgentcoreEvaluatorArgs
    The arguments to resource properties.
    opts ResourceOptions
    Bag of options to control resource's behavior.
    ctx Context
    Context object for the current deployment.
    name string
    The unique name of the resource.
    args AgentcoreEvaluatorArgs
    The arguments to resource properties.
    opts ResourceOption
    Bag of options to control resource's behavior.
    name string
    The unique name of the resource.
    args AgentcoreEvaluatorArgs
    The arguments to resource properties.
    opts CustomResourceOptions
    Bag of options to control resource's behavior.
    name String
    The unique name of the resource.
    args AgentcoreEvaluatorArgs
    The arguments to resource properties.
    options CustomResourceOptions
    Bag of options to control resource's behavior.

    Constructor example

    The following reference example uses placeholder values for all input properties.

    var agentcoreEvaluatorResource = new Aws.Bedrock.AgentcoreEvaluator("agentcoreEvaluatorResource", new()
{
    EvaluatorConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigArgs
    {
        CodeBased = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigCodeBasedArgs
        {
            LambdaConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfigArgs
            {
                LambdaArn = "string",
                LambdaTimeoutInSeconds = 0,
            },
        },
        LlmAsAJudge = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs
        {
            Instructions = "string",
            ModelConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs
            {
                BedrockEvaluatorModelConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs
                {
                    ModelId = "string",
                    AdditionalModelRequestFields = "string",
                    InferenceConfig = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfigArgs
                    {
                        MaxTokens = 0,
                        StopSequences = new[]
                        {
                            "string",
                        },
                        Temperature = 0,
                        TopP = 0,
                    },
                },
            },
            RatingScale = new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs
            {
                Categoricals = new[]
                {
                    new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs
                    {
                        Definition = "string",
                        Label = "string",
                    },
                },
                Numericals = new[]
                {
                    new Aws.Bedrock.Inputs.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArgs
                    {
                        Definition = "string",
                        Label = "string",
                        Value = 0,
                    },
                },
            },
        },
    },
    EvaluatorName = "string",
    Level = "string",
    Description = "string",
    KmsKeyArn = "string",
    Region = "string",
    Tags = 
    {
        { "string", "string" },
    },
    Timeouts = new Aws.Bedrock.Inputs.AgentcoreEvaluatorTimeoutsArgs
    {
        Create = "string",
        Delete = "string",
        Update = "string",
    },
});

    example, err := bedrock.NewAgentcoreEvaluator(ctx, "agentcoreEvaluatorResource", &bedrock.AgentcoreEvaluatorArgs{
	EvaluatorConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigArgs{
		CodeBased: &bedrock.AgentcoreEvaluatorEvaluatorConfigCodeBasedArgs{
			LambdaConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfigArgs{
				LambdaArn:              pulumi.String("string"),
				LambdaTimeoutInSeconds: pulumi.Int(0),
			},
		},
		LlmAsAJudge: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs{
			Instructions: pulumi.String("string"),
			ModelConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs{
				BedrockEvaluatorModelConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs{
					ModelId:                      pulumi.String("string"),
					AdditionalModelRequestFields: pulumi.String("string"),
					InferenceConfig: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfigArgs{
						MaxTokens: pulumi.Int(0),
						StopSequences: pulumi.StringArray{
							pulumi.String("string"),
						},
						Temperature: pulumi.Float64(0),
						TopP:        pulumi.Float64(0),
					},
				},
			},
			RatingScale: &bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs{
				Categoricals: bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArray{
					&bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs{
						Definition: pulumi.String("string"),
						Label:      pulumi.String("string"),
					},
				},
				Numericals: bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArray{
					&bedrock.AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArgs{
						Definition: pulumi.String("string"),
						Label:      pulumi.String("string"),
						Value:      pulumi.Float64(0),
					},
				},
			},
		},
	},
	EvaluatorName: pulumi.String("string"),
	Level:         pulumi.String("string"),
	Description:   pulumi.String("string"),
	KmsKeyArn:     pulumi.String("string"),
	Region:        pulumi.String("string"),
	Tags: pulumi.StringMap{
		"string": pulumi.String("string"),
	},
	Timeouts: &bedrock.AgentcoreEvaluatorTimeoutsArgs{
		Create: pulumi.String("string"),
		Delete: pulumi.String("string"),
		Update: pulumi.String("string"),
	},
})

    resource "aws_bedrock_agentcore_evaluator" "agentcoreEvaluatorResource" {
  lifecycle {
    create_before_destroy = true
  }
  evaluator_config = {
    code_based = {
      lambda_config = {
        lambda_arn                = "string"
        lambda_timeout_in_seconds = 0
      }
    }
    llm_as_a_judge = {
      instructions = "string"
      model_config = {
        bedrock_evaluator_model_config = {
          model_id                        = "string"
          additional_model_request_fields = "string"
          inference_config = {
            max_tokens     = 0
            stop_sequences = ["string"]
            temperature    = 0
            top_p          = 0
          }
        }
      }
      rating_scale = {
        categoricals = [{
          definition = "string"
          label      = "string"
        }]
        numericals = [{
          definition = "string"
          label      = "string"
          value      = 0
        }]
      }
    }
  }
  evaluator_name = "string"
  level          = "string"
  description    = "string"
  kms_key_arn    = "string"
  region         = "string"
  tags = {
    "string" = "string"
  }
  timeouts = {
    create = "string"
    delete = "string"
    update = "string"
  }
}

    var agentcoreEvaluatorResource = new AgentcoreEvaluator("agentcoreEvaluatorResource", AgentcoreEvaluatorArgs.builder()
    .evaluatorConfig(AgentcoreEvaluatorEvaluatorConfigArgs.builder()
        .codeBased(AgentcoreEvaluatorEvaluatorConfigCodeBasedArgs.builder()
            .lambdaConfig(AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfigArgs.builder()
                .lambdaArn("string")
                .lambdaTimeoutInSeconds(0)
                .build())
            .build())
        .llmAsAJudge(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs.builder()
            .instructions("string")
            .modelConfig(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs.builder()
                .bedrockEvaluatorModelConfig(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs.builder()
                    .modelId("string")
                    .additionalModelRequestFields("string")
                    .inferenceConfig(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfigArgs.builder()
                        .maxTokens(0)
                        .stopSequences("string")
                        .temperature(0.0)
                        .topP(0.0)
                        .build())
                    .build())
                .build())
            .ratingScale(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs.builder()
                .categoricals(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs.builder()
                    .definition("string")
                    .label("string")
                    .build())
                .numericals(AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArgs.builder()
                    .definition("string")
                    .label("string")
                    .value(0.0)
                    .build())
                .build())
            .build())
        .build())
    .evaluatorName("string")
    .level("string")
    .description("string")
    .kmsKeyArn("string")
    .region("string")
    .tags(Map.of("string", "string"))
    .timeouts(AgentcoreEvaluatorTimeoutsArgs.builder()
        .create("string")
        .delete("string")
        .update("string")
        .build())
    .build());

    agentcore_evaluator_resource = aws.bedrock.AgentcoreEvaluator("agentcoreEvaluatorResource",
    evaluator_config={
        "code_based": {
            "lambda_config": {
                "lambda_arn": "string",
                "lambda_timeout_in_seconds": 0,
            },
        },
        "llm_as_a_judge": {
            "instructions": "string",
            "model_config": {
                "bedrock_evaluator_model_config": {
                    "model_id": "string",
                    "additional_model_request_fields": "string",
                    "inference_config": {
                        "max_tokens": 0,
                        "stop_sequences": ["string"],
                        "temperature": float(0),
                        "top_p": float(0),
                    },
                },
            },
            "rating_scale": {
                "categoricals": [{
                    "definition": "string",
                    "label": "string",
                }],
                "numericals": [{
                    "definition": "string",
                    "label": "string",
                    "value": float(0),
                }],
            },
        },
    },
    evaluator_name="string",
    level="string",
    description="string",
    kms_key_arn="string",
    region="string",
    tags={
        "string": "string",
    },
    timeouts={
        "create": "string",
        "delete": "string",
        "update": "string",
    })

    const agentcoreEvaluatorResource = new aws.bedrock.AgentcoreEvaluator("agentcoreEvaluatorResource", {
    evaluatorConfig: {
        codeBased: {
            lambdaConfig: {
                lambdaArn: "string",
                lambdaTimeoutInSeconds: 0,
            },
        },
        llmAsAJudge: {
            instructions: "string",
            modelConfig: {
                bedrockEvaluatorModelConfig: {
                    modelId: "string",
                    additionalModelRequestFields: "string",
                    inferenceConfig: {
                        maxTokens: 0,
                        stopSequences: ["string"],
                        temperature: 0,
                        topP: 0,
                    },
                },
            },
            ratingScale: {
                categoricals: [{
                    definition: "string",
                    label: "string",
                }],
                numericals: [{
                    definition: "string",
                    label: "string",
                    value: 0,
                }],
            },
        },
    },
    evaluatorName: "string",
    level: "string",
    description: "string",
    kmsKeyArn: "string",
    region: "string",
    tags: {
        string: "string",
    },
    timeouts: {
        create: "string",
        "delete": "string",
        update: "string",
    },
});

    type: aws:bedrock:AgentcoreEvaluator
properties:
    description: string
    evaluatorConfig:
        codeBased:
            lambdaConfig:
                lambdaArn: string
                lambdaTimeoutInSeconds: 0
        llmAsAJudge:
            instructions: string
            modelConfig:
                bedrockEvaluatorModelConfig:
                    additionalModelRequestFields: string
                    inferenceConfig:
                        maxTokens: 0
                        stopSequences:
                            - string
                        temperature: 0
                        topP: 0
                    modelId: string
            ratingScale:
                categoricals:
                    - definition: string
                      label: string
                numericals:
                    - definition: string
                      label: string
                      value: 0
    evaluatorName: string
    kmsKeyArn: string
    level: string
    region: string
    tags:
        string: string
    timeouts:
        create: string
        delete: string
        update: string

    AgentcoreEvaluator Resource Properties

    To learn more about resource properties and how to use them, see Inputs and Outputs in the Architecture and Concepts docs.

    Inputs

    In Python, inputs that are objects can be passed either as argument classes or as dictionary literals.

    The AgentcoreEvaluator resource accepts the following input properties:

    EvaluatorConfig AgentcoreEvaluatorEvaluatorConfig
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    EvaluatorName string
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    Level string

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    Description string
    Description of the evaluator. Length 1–200.
    KmsKeyArn string
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    Region string
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    Tags Dictionary<string, string>
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    Timeouts AgentcoreEvaluatorTimeouts
    EvaluatorConfig AgentcoreEvaluatorEvaluatorConfigArgs
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    EvaluatorName string
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    Level string

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    Description string
    Description of the evaluator. Length 1–200.
    KmsKeyArn string
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    Region string
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    Tags map[string]string
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    Timeouts AgentcoreEvaluatorTimeoutsArgs
    evaluator_config object
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    evaluator_name string
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    level string

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    description string
    Description of the evaluator. Length 1–200.
    kms_key_arn string
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    region string
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    tags map(string)
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    timeouts object
    evaluatorConfig AgentcoreEvaluatorEvaluatorConfig
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    evaluatorName String
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    level String

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    description String
    Description of the evaluator. Length 1–200.
    kmsKeyArn String
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    region String
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    tags Map<String,String>
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    timeouts AgentcoreEvaluatorTimeouts
    evaluatorConfig AgentcoreEvaluatorEvaluatorConfig
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    evaluatorName string
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    level string

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    description string
    Description of the evaluator. Length 1–200.
    kmsKeyArn string
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    region string
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    tags {[key: string]: string}
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    timeouts AgentcoreEvaluatorTimeouts
    evaluator_config AgentcoreEvaluatorEvaluatorConfigArgs
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    evaluator_name str
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    level str

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    description str
    Description of the evaluator. Length 1–200.
    kms_key_arn str
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    region str
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    tags Mapping[str, str]
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    timeouts AgentcoreEvaluatorTimeoutsArgs
    evaluatorConfig Property Map
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    evaluatorName String
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    level String

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    description String
    Description of the evaluator. Length 1–200.
    kmsKeyArn String
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    region String
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    tags Map<String>
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    timeouts Property Map

    Outputs

    All input properties are implicitly available as output properties. Additionally, the AgentcoreEvaluator resource produces the following output properties:

    CreatedAt string
    Timestamp when the evaluator was created.
    EvaluatorArn string
    ARN of the evaluator.
    EvaluatorId string
    Unique identifier of the evaluator.
    Id string
    The provider-assigned unique ID for this managed resource.
    LockedForModification bool
    Whether the evaluator is locked because it is in use by an active online evaluation.
    Status string
    Current status of the evaluator.
    TagsAll Dictionary<string, string>
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    CreatedAt string
    Timestamp when the evaluator was created.
    EvaluatorArn string
    ARN of the evaluator.
    EvaluatorId string
    Unique identifier of the evaluator.
    Id string
    The provider-assigned unique ID for this managed resource.
    LockedForModification bool
    Whether the evaluator is locked because it is in use by an active online evaluation.
    Status string
    Current status of the evaluator.
    TagsAll map[string]string
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    created_at string
    Timestamp when the evaluator was created.
    evaluator_arn string
    ARN of the evaluator.
    evaluator_id string
    Unique identifier of the evaluator.
    id string
    The provider-assigned unique ID for this managed resource.
    locked_for_modification bool
    Whether the evaluator is locked because it is in use by an active online evaluation.
    status string
    Current status of the evaluator.
    tags_all map(string)
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    createdAt String
    Timestamp when the evaluator was created.
    evaluatorArn String
    ARN of the evaluator.
    evaluatorId String
    Unique identifier of the evaluator.
    id String
    The provider-assigned unique ID for this managed resource.
    lockedForModification Boolean
    Whether the evaluator is locked because it is in use by an active online evaluation.
    status String
    Current status of the evaluator.
    tagsAll Map<String,String>
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    createdAt string
    Timestamp when the evaluator was created.
    evaluatorArn string
    ARN of the evaluator.
    evaluatorId string
    Unique identifier of the evaluator.
    id string
    The provider-assigned unique ID for this managed resource.
    lockedForModification boolean
    Whether the evaluator is locked because it is in use by an active online evaluation.
    status string
    Current status of the evaluator.
    tagsAll {[key: string]: string}
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    created_at str
    Timestamp when the evaluator was created.
    evaluator_arn str
    ARN of the evaluator.
    evaluator_id str
    Unique identifier of the evaluator.
    id str
    The provider-assigned unique ID for this managed resource.
    locked_for_modification bool
    Whether the evaluator is locked because it is in use by an active online evaluation.
    status str
    Current status of the evaluator.
    tags_all Mapping[str, str]
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    createdAt String
    Timestamp when the evaluator was created.
    evaluatorArn String
    ARN of the evaluator.
    evaluatorId String
    Unique identifier of the evaluator.
    id String
    The provider-assigned unique ID for this managed resource.
    lockedForModification Boolean
    Whether the evaluator is locked because it is in use by an active online evaluation.
    status String
    Current status of the evaluator.
    tagsAll Map<String>
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.

    Look up Existing AgentcoreEvaluator Resource

    Get an existing AgentcoreEvaluator resource’s state with the given name, ID, and optional extra properties used to qualify the lookup.

    public static get(name: string, id: Input<ID>, state?: AgentcoreEvaluatorState, opts?: CustomResourceOptions): AgentcoreEvaluator
    @staticmethod
def get(resource_name: str,
        id: str,
        opts: Optional[ResourceOptions] = None,
        created_at: Optional[str] = None,
        description: Optional[str] = None,
        evaluator_arn: Optional[str] = None,
        evaluator_config: Optional[AgentcoreEvaluatorEvaluatorConfigArgs] = None,
        evaluator_id: Optional[str] = None,
        evaluator_name: Optional[str] = None,
        kms_key_arn: Optional[str] = None,
        level: Optional[str] = None,
        locked_for_modification: Optional[bool] = None,
        region: Optional[str] = None,
        status: Optional[str] = None,
        tags: Optional[Mapping[str, str]] = None,
        tags_all: Optional[Mapping[str, str]] = None,
        timeouts: Optional[AgentcoreEvaluatorTimeoutsArgs] = None) -> AgentcoreEvaluator
    func GetAgentcoreEvaluator(ctx *Context, name string, id IDInput, state *AgentcoreEvaluatorState, opts ...ResourceOption) (*AgentcoreEvaluator, error)
    public static AgentcoreEvaluator Get(string name, Input<string> id, AgentcoreEvaluatorState? state, CustomResourceOptions? opts = null)
    public static AgentcoreEvaluator get(String name, Output<String> id, AgentcoreEvaluatorState state, CustomResourceOptions options)
    resources:  _:    type: aws:bedrock:AgentcoreEvaluator    get:      id: ${id}
    import {
  to = aws_bedrock_agentcore_evaluator.example
  id = "${id}"
}
    name
    The unique name of the resulting resource.
    id
    The unique provider ID of the resource to lookup.
    state
    Any extra arguments used during the lookup.
    opts
    A bag of options that control this resource's behavior.
    resource_name
    The unique name of the resulting resource.
    id
    The unique provider ID of the resource to lookup.
    name
    The unique name of the resulting resource.
    id
    The unique provider ID of the resource to lookup.
    state
    Any extra arguments used during the lookup.
    opts
    A bag of options that control this resource's behavior.
    name
    The unique name of the resulting resource.
    id
    The unique provider ID of the resource to lookup.
    state
    Any extra arguments used during the lookup.
    opts
    A bag of options that control this resource's behavior.
    name
    The unique name of the resulting resource.
    id
    The unique provider ID of the resource to lookup.
    state
    Any extra arguments used during the lookup.
    opts
    A bag of options that control this resource's behavior.
    The following state arguments are supported:
    CreatedAt string
    Timestamp when the evaluator was created.
    Description string
    Description of the evaluator. Length 1–200.
    EvaluatorArn string
    ARN of the evaluator.
    EvaluatorConfig AgentcoreEvaluatorEvaluatorConfig
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    EvaluatorId string
    Unique identifier of the evaluator.
    EvaluatorName string
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    KmsKeyArn string
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    Level string

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    LockedForModification bool
    Whether the evaluator is locked because it is in use by an active online evaluation.
    Region string
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    Status string
    Current status of the evaluator.
    Tags Dictionary<string, string>
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    TagsAll Dictionary<string, string>
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    Timeouts AgentcoreEvaluatorTimeouts
    CreatedAt string
    Timestamp when the evaluator was created.
    Description string
    Description of the evaluator. Length 1–200.
    EvaluatorArn string
    ARN of the evaluator.
    EvaluatorConfig AgentcoreEvaluatorEvaluatorConfigArgs
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    EvaluatorId string
    Unique identifier of the evaluator.
    EvaluatorName string
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    KmsKeyArn string
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    Level string

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    LockedForModification bool
    Whether the evaluator is locked because it is in use by an active online evaluation.
    Region string
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    Status string
    Current status of the evaluator.
    Tags map[string]string
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    TagsAll map[string]string
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    Timeouts AgentcoreEvaluatorTimeoutsArgs
    created_at string
    Timestamp when the evaluator was created.
    description string
    Description of the evaluator. Length 1–200.
    evaluator_arn string
    ARN of the evaluator.
    evaluator_config object
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    evaluator_id string
    Unique identifier of the evaluator.
    evaluator_name string
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    kms_key_arn string
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    level string

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    locked_for_modification bool
    Whether the evaluator is locked because it is in use by an active online evaluation.
    region string
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    status string
    Current status of the evaluator.
    tags map(string)
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    tags_all map(string)
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    timeouts object
    createdAt String
    Timestamp when the evaluator was created.
    description String
    Description of the evaluator. Length 1–200.
    evaluatorArn String
    ARN of the evaluator.
    evaluatorConfig AgentcoreEvaluatorEvaluatorConfig
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    evaluatorId String
    Unique identifier of the evaluator.
    evaluatorName String
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    kmsKeyArn String
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    level String

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    lockedForModification Boolean
    Whether the evaluator is locked because it is in use by an active online evaluation.
    region String
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    status String
    Current status of the evaluator.
    tags Map<String,String>
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    tagsAll Map<String,String>
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    timeouts AgentcoreEvaluatorTimeouts
    createdAt string
    Timestamp when the evaluator was created.
    description string
    Description of the evaluator. Length 1–200.
    evaluatorArn string
    ARN of the evaluator.
    evaluatorConfig AgentcoreEvaluatorEvaluatorConfig
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    evaluatorId string
    Unique identifier of the evaluator.
    evaluatorName string
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    kmsKeyArn string
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    level string

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    lockedForModification boolean
    Whether the evaluator is locked because it is in use by an active online evaluation.
    region string
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    status string
    Current status of the evaluator.
    tags {[key: string]: string}
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    tagsAll {[key: string]: string}
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    timeouts AgentcoreEvaluatorTimeouts
    created_at str
    Timestamp when the evaluator was created.
    description str
    Description of the evaluator. Length 1–200.
    evaluator_arn str
    ARN of the evaluator.
    evaluator_config AgentcoreEvaluatorEvaluatorConfigArgs
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    evaluator_id str
    Unique identifier of the evaluator.
    evaluator_name str
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    kms_key_arn str
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    level str

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    locked_for_modification bool
    Whether the evaluator is locked because it is in use by an active online evaluation.
    region str
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    status str
    Current status of the evaluator.
    tags Mapping[str, str]
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    tags_all Mapping[str, str]
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    timeouts AgentcoreEvaluatorTimeoutsArgs
    createdAt String
    Timestamp when the evaluator was created.
    description String
    Description of the evaluator. Length 1–200.
    evaluatorArn String
    ARN of the evaluator.
    evaluatorConfig Property Map
    Configuration that defines how the evaluator assesses agent performance. See evaluatorConfig below.
    evaluatorId String
    Unique identifier of the evaluator.
    evaluatorName String
    Name of the evaluator. Must match the pattern ^[a-zA-Z][a-zA-Z0-9_]{0,47}$.
    kmsKeyArn String
    ARN of a customer-managed KMS key used to encrypt the evaluator's sensitive data. Only symmetric encryption keys are supported.
    level String

    Evaluation level that determines the scope of evaluation. Valid values: TOOL_CALL, TRACE, SESSION.

    The following arguments are optional:

    lockedForModification Boolean
    Whether the evaluator is locked because it is in use by an active online evaluation.
    region String
    Region where this resource will be managed. Defaults to the Region set in the provider configuration.
    status String
    Current status of the evaluator.
    tags Map<String>
    Map of tags assigned to the resource. If configured with a provider defaultTags configuration block present, tags with matching keys will overwrite those defined at the provider-level.
    tagsAll Map<String>
    Map of tags assigned to the resource, including those inherited from the provider defaultTags configuration block.
    timeouts Property Map

    Supporting Types

    AgentcoreEvaluatorEvaluatorConfig, AgentcoreEvaluatorEvaluatorConfigArgs

    CodeBased AgentcoreEvaluatorEvaluatorConfigCodeBased
    Configuration that runs a Lambda function you provide to score the agent. See codeBased below.
    LlmAsAJudge AgentcoreEvaluatorEvaluatorConfigLlmAsAJudge
    Configuration that uses a Bedrock model to score the agent. See llmAsAJudge below.
    CodeBased AgentcoreEvaluatorEvaluatorConfigCodeBased
    Configuration that runs a Lambda function you provide to score the agent. See codeBased below.
    LlmAsAJudge AgentcoreEvaluatorEvaluatorConfigLlmAsAJudge
    Configuration that uses a Bedrock model to score the agent. See llmAsAJudge below.
    code_based object
    Configuration that runs a Lambda function you provide to score the agent. See codeBased below.
    llm_as_a_judge object
    Configuration that uses a Bedrock model to score the agent. See llmAsAJudge below.
    codeBased AgentcoreEvaluatorEvaluatorConfigCodeBased
    Configuration that runs a Lambda function you provide to score the agent. See codeBased below.
    llmAsAJudge AgentcoreEvaluatorEvaluatorConfigLlmAsAJudge
    Configuration that uses a Bedrock model to score the agent. See llmAsAJudge below.
    codeBased AgentcoreEvaluatorEvaluatorConfigCodeBased
    Configuration that runs a Lambda function you provide to score the agent. See codeBased below.
    llmAsAJudge AgentcoreEvaluatorEvaluatorConfigLlmAsAJudge
    Configuration that uses a Bedrock model to score the agent. See llmAsAJudge below.
    code_based AgentcoreEvaluatorEvaluatorConfigCodeBased
    Configuration that runs a Lambda function you provide to score the agent. See codeBased below.
    llm_as_a_judge AgentcoreEvaluatorEvaluatorConfigLlmAsAJudge
    Configuration that uses a Bedrock model to score the agent. See llmAsAJudge below.
    codeBased Property Map
    Configuration that runs a Lambda function you provide to score the agent. See codeBased below.
    llmAsAJudge Property Map
    Configuration that uses a Bedrock model to score the agent. See llmAsAJudge below.

    AgentcoreEvaluatorEvaluatorConfigCodeBased, AgentcoreEvaluatorEvaluatorConfigCodeBasedArgs

    LambdaConfig AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfig
    Lambda function configuration. See lambdaConfig below.
    LambdaConfig AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfig
    Lambda function configuration. See lambdaConfig below.
    lambda_config object
    Lambda function configuration. See lambdaConfig below.
    lambdaConfig AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfig
    Lambda function configuration. See lambdaConfig below.
    lambdaConfig AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfig
    Lambda function configuration. See lambdaConfig below.
    lambda_config AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfig
    Lambda function configuration. See lambdaConfig below.
    lambdaConfig Property Map
    Lambda function configuration. See lambdaConfig below.

    AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfig, AgentcoreEvaluatorEvaluatorConfigCodeBasedLambdaConfigArgs

    LambdaArn string
    ARN of the Lambda function that runs the evaluation.
    LambdaTimeoutInSeconds int
    Time in seconds to wait for the Lambda function before timing out. Defaults to 60. Range 1–300.
    LambdaArn string
    ARN of the Lambda function that runs the evaluation.
    LambdaTimeoutInSeconds int
    Time in seconds to wait for the Lambda function before timing out. Defaults to 60. Range 1–300.
    lambda_arn string
    ARN of the Lambda function that runs the evaluation.
    lambda_timeout_in_seconds number
    Time in seconds to wait for the Lambda function before timing out. Defaults to 60. Range 1–300.
    lambdaArn String
    ARN of the Lambda function that runs the evaluation.
    lambdaTimeoutInSeconds Integer
    Time in seconds to wait for the Lambda function before timing out. Defaults to 60. Range 1–300.
    lambdaArn string
    ARN of the Lambda function that runs the evaluation.
    lambdaTimeoutInSeconds number
    Time in seconds to wait for the Lambda function before timing out. Defaults to 60. Range 1–300.
    lambda_arn str
    ARN of the Lambda function that runs the evaluation.
    lambda_timeout_in_seconds int
    Time in seconds to wait for the Lambda function before timing out. Defaults to 60. Range 1–300.
    lambdaArn String
    ARN of the Lambda function that runs the evaluation.
    lambdaTimeoutInSeconds Number
    Time in seconds to wait for the Lambda function before timing out. Defaults to 60. Range 1–300.

    AgentcoreEvaluatorEvaluatorConfigLlmAsAJudge, AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeArgs

    Instructions string
    Instructions that tell the model how to score the agent.
    ModelConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfig
    Which Bedrock model to use. See modelConfig below.
    RatingScale AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScale
    Scale used to score the agent. See ratingScale below.
    Instructions string
    Instructions that tell the model how to score the agent.
    ModelConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfig
    Which Bedrock model to use. See modelConfig below.
    RatingScale AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScale
    Scale used to score the agent. See ratingScale below.
    instructions string
    Instructions that tell the model how to score the agent.
    model_config object
    Which Bedrock model to use. See modelConfig below.
    rating_scale object
    Scale used to score the agent. See ratingScale below.
    instructions String
    Instructions that tell the model how to score the agent.
    modelConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfig
    Which Bedrock model to use. See modelConfig below.
    ratingScale AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScale
    Scale used to score the agent. See ratingScale below.
    instructions string
    Instructions that tell the model how to score the agent.
    modelConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfig
    Which Bedrock model to use. See modelConfig below.
    ratingScale AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScale
    Scale used to score the agent. See ratingScale below.
    instructions str
    Instructions that tell the model how to score the agent.
    model_config AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfig
    Which Bedrock model to use. See modelConfig below.
    rating_scale AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScale
    Scale used to score the agent. See ratingScale below.
    instructions String
    Instructions that tell the model how to score the agent.
    modelConfig Property Map
    Which Bedrock model to use. See modelConfig below.
    ratingScale Property Map
    Scale used to score the agent. See ratingScale below.

    AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfig, AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigArgs

    BedrockEvaluatorModelConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfig
    Amazon Bedrock model configuration. See bedrockEvaluatorModelConfig below.
    BedrockEvaluatorModelConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfig
    Amazon Bedrock model configuration. See bedrockEvaluatorModelConfig below.
    bedrock_evaluator_model_config object
    Amazon Bedrock model configuration. See bedrockEvaluatorModelConfig below.
    bedrockEvaluatorModelConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfig
    Amazon Bedrock model configuration. See bedrockEvaluatorModelConfig below.
    bedrockEvaluatorModelConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfig
    Amazon Bedrock model configuration. See bedrockEvaluatorModelConfig below.
    bedrock_evaluator_model_config AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfig
    Amazon Bedrock model configuration. See bedrockEvaluatorModelConfig below.
    bedrockEvaluatorModelConfig Property Map
    Amazon Bedrock model configuration. See bedrockEvaluatorModelConfig below.

    AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfig, AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigArgs

    ModelId string
    Identifier of the Amazon Bedrock model to use for evaluation.
    AdditionalModelRequestFields string
    JSON-encoded model-specific request fields, for settings not covered by inferenceConfig.
    InferenceConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfig
    Settings that control how the model generates its response. See inferenceConfig below.
    ModelId string
    Identifier of the Amazon Bedrock model to use for evaluation.
    AdditionalModelRequestFields string
    JSON-encoded model-specific request fields, for settings not covered by inferenceConfig.
    InferenceConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfig
    Settings that control how the model generates its response. See inferenceConfig below.
    model_id string
    Identifier of the Amazon Bedrock model to use for evaluation.
    additional_model_request_fields string
    JSON-encoded model-specific request fields, for settings not covered by inferenceConfig.
    inference_config object
    Settings that control how the model generates its response. See inferenceConfig below.
    modelId String
    Identifier of the Amazon Bedrock model to use for evaluation.
    additionalModelRequestFields String
    JSON-encoded model-specific request fields, for settings not covered by inferenceConfig.
    inferenceConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfig
    Settings that control how the model generates its response. See inferenceConfig below.
    modelId string
    Identifier of the Amazon Bedrock model to use for evaluation.
    additionalModelRequestFields string
    JSON-encoded model-specific request fields, for settings not covered by inferenceConfig.
    inferenceConfig AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfig
    Settings that control how the model generates its response. See inferenceConfig below.
    model_id str
    Identifier of the Amazon Bedrock model to use for evaluation.
    additional_model_request_fields str
    JSON-encoded model-specific request fields, for settings not covered by inferenceConfig.
    inference_config AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfig
    Settings that control how the model generates its response. See inferenceConfig below.
    modelId String
    Identifier of the Amazon Bedrock model to use for evaluation.
    additionalModelRequestFields String
    JSON-encoded model-specific request fields, for settings not covered by inferenceConfig.
    inferenceConfig Property Map
    Settings that control how the model generates its response. See inferenceConfig below.

    AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfig, AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeModelConfigBedrockEvaluatorModelConfigInferenceConfigArgs

    MaxTokens int
    Maximum number of tokens to generate in the model response. Must be at least 1.
    StopSequences List<string>
    List of sequences that cause the model to stop generating tokens.
    Temperature double
    Temperature value that controls randomness. Range 0–1.
    TopP double
    Top-p sampling parameter. Range 0–1.
    MaxTokens int
    Maximum number of tokens to generate in the model response. Must be at least 1.
    StopSequences []string
    List of sequences that cause the model to stop generating tokens.
    Temperature float64
    Temperature value that controls randomness. Range 0–1.
    TopP float64
    Top-p sampling parameter. Range 0–1.
    max_tokens number
    Maximum number of tokens to generate in the model response. Must be at least 1.
    stop_sequences list(string)
    List of sequences that cause the model to stop generating tokens.
    temperature number
    Temperature value that controls randomness. Range 0–1.
    top_p number
    Top-p sampling parameter. Range 0–1.
    maxTokens Integer
    Maximum number of tokens to generate in the model response. Must be at least 1.
    stopSequences List<String>
    List of sequences that cause the model to stop generating tokens.
    temperature Double
    Temperature value that controls randomness. Range 0–1.
    topP Double
    Top-p sampling parameter. Range 0–1.
    maxTokens number
    Maximum number of tokens to generate in the model response. Must be at least 1.
    stopSequences string[]
    List of sequences that cause the model to stop generating tokens.
    temperature number
    Temperature value that controls randomness. Range 0–1.
    topP number
    Top-p sampling parameter. Range 0–1.
    max_tokens int
    Maximum number of tokens to generate in the model response. Must be at least 1.
    stop_sequences Sequence[str]
    List of sequences that cause the model to stop generating tokens.
    temperature float
    Temperature value that controls randomness. Range 0–1.
    top_p float
    Top-p sampling parameter. Range 0–1.
    maxTokens Number
    Maximum number of tokens to generate in the model response. Must be at least 1.
    stopSequences List<String>
    List of sequences that cause the model to stop generating tokens.
    temperature Number
    Temperature value that controls randomness. Range 0–1.
    topP Number
    Top-p sampling parameter. Range 0–1.

    AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScale, AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleArgs

    Categoricals List<AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategorical>
    One or more categorical rating scale definitions. See categorical below.
    Numericals List<AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumerical>
    One or more numerical rating scale definitions. See numerical below.
    Categoricals []AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategorical
    One or more categorical rating scale definitions. See categorical below.
    Numericals []AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumerical
    One or more numerical rating scale definitions. See numerical below.
    categoricals list(object)
    One or more categorical rating scale definitions. See categorical below.
    numericals list(object)
    One or more numerical rating scale definitions. See numerical below.
    categoricals List<AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategorical>
    One or more categorical rating scale definitions. See categorical below.
    numericals List<AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumerical>
    One or more numerical rating scale definitions. See numerical below.
    categoricals AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategorical[]
    One or more categorical rating scale definitions. See categorical below.
    numericals AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumerical[]
    One or more numerical rating scale definitions. See numerical below.
    categoricals Sequence[AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategorical]
    One or more categorical rating scale definitions. See categorical below.
    numericals Sequence[AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumerical]
    One or more numerical rating scale definitions. See numerical below.
    categoricals List<Property Map>
    One or more categorical rating scale definitions. See categorical below.
    numericals List<Property Map>
    One or more numerical rating scale definitions. See numerical below.

    AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategorical, AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleCategoricalArgs

    Definition string
    Description that explains what this categorical rating represents.
    Label string
    Label for this categorical rating option. Length 1–100.
    Definition string
    Description that explains what this categorical rating represents.
    Label string
    Label for this categorical rating option. Length 1–100.
    definition string
    Description that explains what this categorical rating represents.
    label string
    Label for this categorical rating option. Length 1–100.
    definition String
    Description that explains what this categorical rating represents.
    label String
    Label for this categorical rating option. Length 1–100.
    definition string
    Description that explains what this categorical rating represents.
    label string
    Label for this categorical rating option. Length 1–100.
    definition str
    Description that explains what this categorical rating represents.
    label str
    Label for this categorical rating option. Length 1–100.
    definition String
    Description that explains what this categorical rating represents.
    label String
    Label for this categorical rating option. Length 1–100.

    AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumerical, AgentcoreEvaluatorEvaluatorConfigLlmAsAJudgeRatingScaleNumericalArgs

    Definition string
    Description that explains what this numerical rating represents.
    Label string
    Label for this numerical rating option. Length 1–100.
    Value double
    Numerical value for this rating option. Must be at least 0.
    Definition string
    Description that explains what this numerical rating represents.
    Label string
    Label for this numerical rating option. Length 1–100.
    Value float64
    Numerical value for this rating option. Must be at least 0.
    definition string
    Description that explains what this numerical rating represents.
    label string
    Label for this numerical rating option. Length 1–100.
    value number
    Numerical value for this rating option. Must be at least 0.
    definition String
    Description that explains what this numerical rating represents.
    label String
    Label for this numerical rating option. Length 1–100.
    value Double
    Numerical value for this rating option. Must be at least 0.
    definition string
    Description that explains what this numerical rating represents.
    label string
    Label for this numerical rating option. Length 1–100.
    value number
    Numerical value for this rating option. Must be at least 0.
    definition str
    Description that explains what this numerical rating represents.
    label str
    Label for this numerical rating option. Length 1–100.
    value float
    Numerical value for this rating option. Must be at least 0.
    definition String
    Description that explains what this numerical rating represents.
    label String
    Label for this numerical rating option. Length 1–100.
    value Number
    Numerical value for this rating option. Must be at least 0.

    AgentcoreEvaluatorTimeouts, AgentcoreEvaluatorTimeoutsArgs

    Create string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    Delete string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours). Setting a timeout for a Delete operation is only applicable if changes are saved into state before the destroy operation occurs.
    Update string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    Create string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    Delete string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours). Setting a timeout for a Delete operation is only applicable if changes are saved into state before the destroy operation occurs.
    Update string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    create string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    delete string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours). Setting a timeout for a Delete operation is only applicable if changes are saved into state before the destroy operation occurs.
    update string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    create String
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    delete String
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours). Setting a timeout for a Delete operation is only applicable if changes are saved into state before the destroy operation occurs.
    update String
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    create string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    delete string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours). Setting a timeout for a Delete operation is only applicable if changes are saved into state before the destroy operation occurs.
    update string
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    create str
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    delete str
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours). Setting a timeout for a Delete operation is only applicable if changes are saved into state before the destroy operation occurs.
    update str
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    create String
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).
    delete String
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours). Setting a timeout for a Delete operation is only applicable if changes are saved into state before the destroy operation occurs.
    update String
    A string that can be parsed as a duration consisting of numbers and unit suffixes, such as "30s" or "2h45m". Valid time units are "s" (seconds), "m" (minutes), "h" (hours).

    Import

    Identity Schema

    Required

    • evaluatorId (String) Unique identifier of the evaluator.

    Optional

    • accountId (String) AWS Account where this resource is managed.
    • region (String) Region where this resource is managed.

    Using pulumi import, import Bedrock AgentCore Evaluator using the evaluator ID. For example:

    $ pulumi import aws:bedrock/agentcoreEvaluator:AgentcoreEvaluator example helpfulness_evaluator-abc1234567

    To learn more about importing existing cloud resources, see Importing resources.

    Package Details

    Repository
    AWS Classic pulumi/pulumi-aws
    License
    Apache-2.0
    Notes
    This Pulumi package is based on the aws Terraform Provider.
    aws logo
    Viewing docs for AWS v7.39.0
    published on Thursday, Jul 23, 2026 by Pulumi
    Schema (JSON)
    pulumi/pulumi-aws

      Try Pulumi Cloud free.
      Your team will thank you.

      Start free trial