BuiltinEvaluator

class aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator(value)

Bases: object

(experimental) Built-in evaluators provided by Amazon Bedrock AgentCore.

These evaluators assess different aspects of agent performance at various levels (session, trace, or tool call).

Stability:

experimental

ExampleMetadata:

infused

Example:

# Basic usage with built-in evaluators
evaluation = agentcore.OnlineEvaluationConfig(self, "MyEvaluation",
    online_evaluation_config_name="my_evaluation",
    evaluators=[
        agentcore.EvaluatorReference.builtin(agentcore.BuiltinEvaluator.HELPFULNESS),
        agentcore.EvaluatorReference.builtin(agentcore.BuiltinEvaluator.CORRECTNESS)
    ],
    data_source=agentcore.DataSourceConfig.from_cloud_watch_logs(
        log_group_names=["/aws/bedrock-agentcore/my-agent"],
        service_names=["my-agent.default"]
    )
)
Parameters:

value (str) –

  • The evaluator identifier string.

Stability:

experimental

Attributes

COHERENCE = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
CONCISENESS = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
CORRECTNESS = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
FAITHFULNESS = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
GOAL_SUCCESS_RATE = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
HARMFULNESS = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
HELPFULNESS = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
INSTRUCTION_FOLLOWING = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
REFUSAL = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
RESPONSE_RELEVANCE = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
STEREOTYPING = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
TOOL_PARAMETER_ACCURACY = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
TOOL_SELECTION_ACCURACY = <aws_cdk.aws_bedrock_agentcore_alpha.BuiltinEvaluator object>
value

(experimental) The string value of the built-in evaluator.

Stability:

experimental