Config for agent engine methods.
Package
@google-cloud/agentplatformProperties
abortSignal
abortSignal?: AbortSignal;Abort signal which can be used to cancel the request.
NOTE: AbortSignal is a client-only operation. Using it to cancel an operation will not cancel the request in the service. You will still be charged usage for any applicable operations.
agentConfigSource
agentConfigSource?: ReasoningEngineSpecSourceCodeSpecAgentConfigSource;The agent config source for the Agent Engine.
agentFramework
agentFramework?: 'google-adk' | 'langchain' | 'langgraph' | 'ag2' | 'llama-index' | 'custom';The agent framework to be used for the Agent Engine. The OSS agent framework used to develop the agent. Currently supported values: "google-adk", "langchain", "langgraph", "ag2", "llama-index", "custom". If not specified: - If agent is specified, the agent framework will be auto-detected. - If source_packages is specified, the agent framework will default to "custom".
agentGatewayConfig
agentGatewayConfig?: ReasoningEngineSpecDeploymentSpecAgentGatewayConfig;Agent Gateway configuration for a Reasoning Engine deployment.
buildOptions
buildOptions?: Record<string, string[]>;The build options for the Agent Engine. The following keys are supported: - installation_scripts: Optional. The paths to the installation scripts to be executed in the Docker image. The scripts must be located in the installation_scripts subdirectory and the path must be added to extra_packages.
classMethods
classMethods?: Record<string, unknown>[];The class methods to be used for the Agent Engine. If specified, they'll override the class methods that are autogenerated by default. By default, methods are generated by inspecting the agent object and generating a corresponding method for each method defined on the agent class.
containerConcurrency
containerConcurrency?: number;The container concurrency to be used for the Agent Engine. Recommended value: 2 * cpu + 1. Defaults to 9.
containerSpec
containerSpec?: ReasoningEngineSpecContainerSpec;The container spec for the Agent Engine.
contextSpec
contextSpec?: ReasoningEngineContextSpec;The context spec to be used for the Agent Engine.
description
description?: string;The description of the Agent Engine.
developerConnectSource
developerConnectSource?: ReasoningEngineSpecSourceCodeSpecDeveloperConnectConfig;Specifies the configuration for fetching source code from a Git repository that is managed by Developer Connect. This includes the repository, revision, and directory to use.
displayName
displayName?: string;The user-defined name of the Agent Engine.
The name can be up to 128 characters long and can comprise any UTF-8 character.
encryptionSpec
encryptionSpec?: genaiTypes.EncryptionSpec;The encryption spec to be used for the Agent Engine.
entrypointModule
entrypointModule?: string;The entrypoint module to be used for the Agent Engine This field only used when source_packages is specified.
entrypointObject
entrypointObject?: string;The entrypoint object to be used for the Agent Engine. This field only used when source_packages is specified.
envVars
envVars?: unknown;The environment variables to be set when running the Agent Engine.
If it is a dictionary, the keys are the environment variable names, and the values are the corresponding values.
extraPackages
extraPackages?: string[];The set of extra user-provided packages (if any).
gcsDirName
gcsDirName?: string;The GCS bucket directory under staging_bucket to use for staging the artifacts needed.
httpOptions
httpOptions?: genaiTypes.HttpOptions;Used to override HTTP request options.
identityType
identityType?: IdentityType;The identity type to use for the Agent Engine.
imageSpec
imageSpec?: ReasoningEngineSpecSourceCodeSpecImageSpec;The image spec for the Agent Engine.
keepAliveProbe
keepAliveProbe?: KeepAliveProbe;Optional. Specifies the configuration for keep-alive probe. Contains configuration on a specified endpoint that a deployment host should use to keep the container alive based on the probe settings.
labels
labels?: Record<string, string>;The labels to be used for the Agent Engine.
maxInstances
maxInstances?: number;The maximum number of instances to run for the Agent Engine. Defaults to 100. Range: [1, 1000]. If VPC-SC or PSC-I is enabled, the acceptable range is [1, 100].
minInstances
minInstances?: number;The minimum number of instances to run for the Agent Engine. Defaults to 1. Range: [0, 10].
pscInterfaceConfig
pscInterfaceConfig?: PscInterfaceConfig;The PSC interface config for PSC-I to be used for the Agent Engine.
pythonVersion
pythonVersion?: '3.10' | '3.11' | '3.12' | '3.13' | '3.14';The Python version to be used for the Agent Engine. If not specified, it will use the current Python version of the environment. Supported versions: "3.10", "3.11", "3.12", "3.13", "3.14".
requirements
requirements?: unknown;The set of PyPI dependencies needed.
It can either be the path to a single file (requirements.txt), or an ordered list of strings corresponding to each line of the requirements file.
requirementsFile
requirementsFile?: string;The user-provided path to the requirements file (if any). This field is only used when source_packages is specified. If not specified, agent engine will find and use the requirements.txt in the source package.
resourceLimits
resourceLimits?: Record<string, string>;The resource limits to be applied to the Agent Engine. Required keys: 'cpu' and 'memory'. Supported values for 'cpu': '1', '2', '4', '6', '8'. Supported values for 'memory': '1Gi', '2Gi', ..., '32Gi'.
serviceAccount
serviceAccount?: string;The service account to be used for the Agent Engine.
If not specified, the default Reasoning Engine P6SA service agent will be used.
sourcePackages
sourcePackages?: string[];The user-provided paths to the source packages (if any). If specified, the files in the source packages will be packed into a a tarball file, uploaded to Agent Engine's API, and deployed to the Agent Engine. The following fields will be ignored: - agent - extra_packages - staging_bucket - requirements The following fields will be used to install and use the agent from the source packages: - entrypoint_module (required) - entrypoint_object (required) - requirements_file (optional) - class_methods (required)
stagingBucket
stagingBucket?: string;The GCS bucket to use for staging the artifacts needed.
It must be a valid GCS bucket name, e.g. "gs://bucket-name". It is required if agent_engine is specified.