AgentsforBedrock / Client / create_agent

create_agent

AgentsforBedrock.Client.create_agent(**kwargs)

Creates an Amazon Bedrock Agent

See also: AWS API Documentation

Request Syntax

response = client.create_agent(
    agentName='string',
    clientToken='string',
    instruction='string',
    foundationModel='string',
    description='string',
    idleSessionTTLInSeconds=123,
    agentResourceRoleArn='string',
    customerEncryptionKeyArn='string',
    tags={
        'string': 'string'
    },
    promptOverrideConfiguration={
        'promptConfigurations': [
            {
                'promptType': 'PRE_PROCESSING'|'ORCHESTRATION'|'POST_PROCESSING'|'KNOWLEDGE_BASE_RESPONSE_GENERATION',
                'promptCreationMode': 'DEFAULT'|'OVERRIDDEN',
                'promptState': 'ENABLED'|'DISABLED',
                'basePromptTemplate': 'string',
                'inferenceConfiguration': {
                    'temperature': ...,
                    'topP': ...,
                    'topK': 123,
                    'maximumLength': 123,
                    'stopSequences': [
                        'string',
                    ]
                },
                'parserMode': 'DEFAULT'|'OVERRIDDEN'
            },
        ],
        'overrideLambda': 'string'
    }
)

Parameters:

• agentName (string) –

  [REQUIRED]

  Name for a resource.

• clientToken (string) –

  Client specified token used for idempotency checks

  This field is autopopulated if not provided.

• instruction (string) – Instruction for the agent.

• foundationModel (string) – ARN or name of a Bedrock model.

• description (string) – Description of the Resource.

• idleSessionTTLInSeconds (integer) – Max Session Time.

• agentResourceRoleArn (string) –

  [REQUIRED]

  ARN of a IAM role.

• customerEncryptionKeyArn (string) – A KMS key ARN

• tags (dict) –

  A map of tag keys and values

  • (string) –

    Key of a tag

    • (string) –

      Value of a tag

• promptOverrideConfiguration (dict) –

  Configuration for prompt override.

  • promptConfigurations (list) – [REQUIRED]

    List of BasePromptConfiguration

    • (dict) –

      BasePromptConfiguration per Prompt Type.

      • promptType (string) –

        Prompt Type.

      • promptCreationMode (string) –

        Creation Mode for Prompt Configuration.

      • promptState (string) –

        Prompt State.

      • basePromptTemplate (string) –

        Base Prompt Template.

      • inferenceConfiguration (dict) –

        Configuration for inference in prompt configuration

        • temperature (float) –

          Controls randomness, higher values increase diversity

        • topP (float) –

          Cumulative probability cutoff for token selection

        • topK (integer) –

          Sample from the k most likely next tokens

        • maximumLength (integer) –

          Maximum length of output

        • stopSequences (list) –

          List of stop sequences

          • (string) –

      • parserMode (string) –

        Creation Mode for Prompt Configuration.

  • overrideLambda (string) –

    ARN of a Lambda.

Return type:

dict

Returns:

Response Syntax

{
    'agent': {
        'agentId': 'string',
        'agentName': 'string',
        'agentArn': 'string',
        'agentVersion': 'string',
        'clientToken': 'string',
        'instruction': 'string',
        'agentStatus': 'CREATING'|'PREPARING'|'PREPARED'|'NOT_PREPARED'|'DELETING'|'FAILED'|'VERSIONING'|'UPDATING',
        'foundationModel': 'string',
        'description': 'string',
        'idleSessionTTLInSeconds': 123,
        'agentResourceRoleArn': 'string',
        'customerEncryptionKeyArn': 'string',
        'createdAt': datetime(2015, 1, 1),
        'updatedAt': datetime(2015, 1, 1),
        'preparedAt': datetime(2015, 1, 1),
        'failureReasons': [
            'string',
        ],
        'recommendedActions': [
            'string',
        ],
        'promptOverrideConfiguration': {
            'promptConfigurations': [
                {
                    'promptType': 'PRE_PROCESSING'|'ORCHESTRATION'|'POST_PROCESSING'|'KNOWLEDGE_BASE_RESPONSE_GENERATION',
                    'promptCreationMode': 'DEFAULT'|'OVERRIDDEN',
                    'promptState': 'ENABLED'|'DISABLED',
                    'basePromptTemplate': 'string',
                    'inferenceConfiguration': {
                        'temperature': ...,
                        'topP': ...,
                        'topK': 123,
                        'maximumLength': 123,
                        'stopSequences': [
                            'string',
                        ]
                    },
                    'parserMode': 'DEFAULT'|'OVERRIDDEN'
                },
            ],
            'overrideLambda': 'string'
        }
    }
}

Response Structure

• (dict) –

  Create Agent Response

  • agent (dict) –

    Contains the information of an agent

    • agentId (string) –

      Identifier for a resource.

    • agentName (string) –

      Name for a resource.

    • agentArn (string) –

      Arn representation of the Agent.

    • agentVersion (string) –

      Draft Agent Version.

    • clientToken (string) –

      Client specified token used for idempotency checks

    • instruction (string) –

      Instruction for the agent.

    • agentStatus (string) –

      Schema Type for Action APIs.

    • foundationModel (string) –

      ARN or name of a Bedrock model.

    • description (string) –

      Description of the Resource.

    • idleSessionTTLInSeconds (integer) –

      Max Session Time.

    • agentResourceRoleArn (string) –

      ARN of a IAM role.

    • customerEncryptionKeyArn (string) –

      A KMS key ARN

    • createdAt (datetime) –

      Time Stamp.

    • updatedAt (datetime) –

      Time Stamp.

    • preparedAt (datetime) –

      Time Stamp.

    • failureReasons (list) –

      Failure Reasons for Error.

      • (string) –

        Failure Reason for Error.

    • recommendedActions (list) –

      The recommended actions users can take to resolve an error in failureReasons.

      • (string) –

        The recommended action users can take to resolve an error in failureReasons.

    • promptOverrideConfiguration (dict) –

      Configuration for prompt override.

      • promptConfigurations (list) –

        List of BasePromptConfiguration

        • (dict) –

          BasePromptConfiguration per Prompt Type.

          • promptType (string) –

            Prompt Type.

          • promptCreationMode (string) –

            Creation Mode for Prompt Configuration.

          • promptState (string) –

            Prompt State.

          • basePromptTemplate (string) –

            Base Prompt Template.

          • inferenceConfiguration (dict) –

            Configuration for inference in prompt configuration

            • temperature (float) –

              Controls randomness, higher values increase diversity

            • topP (float) –

              Cumulative probability cutoff for token selection

            • topK (integer) –

              Sample from the k most likely next tokens

            • maximumLength (integer) –

              Maximum length of output

            • stopSequences (list) –

              List of stop sequences

              • (string) –

          • parserMode (string) –

            Creation Mode for Prompt Configuration.

      • overrideLambda (string) –

        ARN of a Lambda.

Exceptions

• AgentsforBedrock.Client.exceptions.ThrottlingException

• AgentsforBedrock.Client.exceptions.AccessDeniedException

• AgentsforBedrock.Client.exceptions.ValidationException

• AgentsforBedrock.Client.exceptions.InternalServerException

• AgentsforBedrock.Client.exceptions.ConflictException

• AgentsforBedrock.Client.exceptions.ServiceQuotaExceededException