Namespace Oci.GenerativeaiagentService.Models
Classes
Agent
An agent is an LLM-based autonomous system that understands and generates human-like text, enabling natural-language processing interactions. OCI Generative AI Agents supports retrieval-augmented generation (RAG) agents. A RAG agent connects to a data source, retrieves data, and augments model responses with the information from the data sources to generate more relevant responses.
To use any of the API operations, you must be authorized in an IAM policy. If you're not authorized, talk to an administrator. If you're an administrator who needs to write policies to give users access, see Getting Started with Policies.
AgentCollection
Results of an agent search. Contains both AgentSummary items and other information, such as metadata.
AgentEndpoint
The endpoint to access a deployed agent.
To use any of the API operations, you must be authorized in an IAM policy. If you're not authorized, talk to an administrator. If you're an administrator who needs to write policies to give users access, see Getting Started with Policies.
AgentEndpointCollection
Results of an agentEndpoint search. Contains both AgentEndpointSummary items and other information, such as metadata.
AgentEndpointSummary
Summary information about an endpoint.
AgentSummary
Summary information about an agent.
ApiSchemaInlineInputLocation
The details of inline input of Api schema.
ApiSchemaInputLocation
The input location definition for Api schema.
ApiSchemaInputLocationModelConverter
ApiSchemaObjectStorageInputLocation
The details of OCI Object Storage object location for Api schema.
BasicAuthSecret
The details of Basic authentication configured as in OpenSearch.
ChangeAgentCompartmentDetails
The configuration details for the move operation.
ChangeAgentEndpointCompartmentDetails
The configuration details for the move operation.
ChangeKnowledgeBaseCompartmentDetails
The configuration details for the move operation.
ContentModerationConfig
The configuration details about whether to apply the content moderation feature to input and output. Content moderation removes toxic and biased content from responses. It is recommended to use content moderation.
ContentModerationGuardrailConfig
The configuration details about whether to apply the content moderation feature to input and output. Content moderation removes toxic and biased content from responses. It is recommended to use content moderation.
CreateAgentDetails
The data to create an agent.
CreateAgentEndpointDetails
The data to create an endpoint.
CreateDataIngestionJobDetails
The data to create a data ingestion job.
CreateDataSourceDetails
The data to create a data source.
CreateKnowledgeBaseDetails
The data to create a knowledge base.
CreateToolDetails
The data to create a Tool.
DataIngestionJob
When you create a data source, you specify the location of the data files. To make those files usable by an agent, you must download them into the agent's associated knowledge base, a process known as data ingestion. Data ingestion is a process that extracts data from data source documents, converts it into a structured format suitable for analysis, and then stores it in a knowledge base.
To use any of the API operations, you must be authorized in an IAM policy. If you're not authorized, talk to an administrator. If you're an administrator who needs to write policies to give users access, see Getting Started with Policies.
DataIngestionJobCollection
Results of a data ingestion job search. Contains both DataIngestionJobSummary items and other information, such as metadata.
DataIngestionJobStatistics
The statistics of data ingestion job.
DataIngestionJobSummary
Summary information about a data ingestion job.
DataSource
A data source points to the source of your data. After you add a data source to a knowledge base, you must ingest the data source's data, so that agents using the knowledge base can refer to the data.
To use any of the API operations, you must be authorized in an IAM policy. If you're not authorized, talk to an administrator. If you're an administrator who needs to write policies to give users access, see Getting Started with Policies.
DataSourceCollection
Results of a dataSource search. Contains both DataSourceSummary items and other information, such as metadata.
DataSourceConfig
The details of data source.
DataSourceConfigModelConverter
DataSourceSummary
Summary information about a data source.
DatabaseConnection
The connection type for Databases.
DatabaseConnectionModelConverter
DatabaseFunction
The details of Database function.
DatabaseToolConnection
The details of the customer Database Tools Connection.
DefaultIndexConfig
The default index is service managed vector store on behalf of the customer.
Function
Details of Function for Function calling tool.
FunctionCallingToolConfig
The configuration for Function calling Tool.
GuardrailConfig
The configuration details about whether to apply the guardrail checks to input and output.
HttpEndpointAuthConfig
Auth related information to be used when invoking external endpoint
HttpEndpointAuthConfigModelConverter
HttpEndpointDelegatedBearerAuthConfig
Specifies Bearer Token Authentication, where the same Bearer token received as part of the Agent Chat API request is used to invoke the external endpoint.
HttpEndpointIdcsAuthConfig
Specifies authentication using Oracle Identity Cloud Service (IDCS), leveraging OAuth 2.0 for token-based authentication.
HttpEndpointNoAuthConfig
Indicates that no authentication is required for invoking the endpoint.
HttpEndpointOciResourcePrincipalAuthConfig
Specifies authentication using Oracle Cloud Infrastructure (OCI) Resource Principal, leveraging OCI's identity-based authentication mechanism.
HttpEndpointToolConfig
The configuration for HTTP endpoint Tool.
HumanInputConfig
Human Input Configuration for an AgentEndpoint.
IdcsSecret
The details of IDCS configured as OpenID setting in OpenSearch.
Index
OCI OpenSearch index details.
IndexConfig
The index configuration of Knowledge bases.
IndexConfigModelConverter
IndexSchema
The index schema details.
InlineInputLocation
The details of inline input location.
InputLocation
The input location definition.
InputLocationModelConverter
KnowledgeBase
A knowledge base is the base for all the data sources that an agent can use to retrieve information for its responses.
To use any of the API operations, you must be authorized in an IAM policy. If you're not authorized, talk to an administrator. If you're an administrator who needs to write policies to give users access, see Getting Started with Policies.
KnowledgeBaseCollection
Results of a knowledge base search. Contains both KnowledgeBaseSummary items and other information, such as metadata.
KnowledgeBaseConfig
The KnowledgeBase configuration this RAG Tool uses.
KnowledgeBaseStatistics
Statistics for Default Knowledge Base.
KnowledgeBaseSummary
Summary information about a knowledge base.
LlmConfig
Configuration to Agent LLM.
LlmCustomization
Configuration to customize LLM.
ObjectStorageInputLocation
The details of OCI Object Storage object.
ObjectStoragePrefix
The details of OCI Object Storage object.
ObjectStoragePrefixOutputLocation
The config of object storage.
OciDatabaseConfig
The details of the customer Database Connection.
OciObjectStorageDataSourceConfig
The details of OCI Search with OpenSearch data source information.
OciOpenSearchIndexConfig
The details of customer managed OCI OpenSearch.
OutputConfig
Configuration to store results generated by agent.
OutputLocation
Location of the output.
OutputLocationModelConverter
PersonallyIdentifiableInformationGuardrailConfig
The configuration details for Personally Identifiable Information.
PromptInjectionGuardrailConfig
The configuration details for Prompt Injection.
RagToolConfig
The configuration for RAG Tool.
SecretDetail
The details of configured security configuration on OpenSearch.
SecretDetailModelConverter
SessionConfig
Session Configuration on AgentEndpoint.
SqlToolConfig
The configuration for SQL Tool.
Tool
The description of Tool.
To use any of the API operations, you must be authorized in an IAM policy. If you're not authorized, talk to an administrator. If you're an administrator who needs to write policies to give users access, see Getting Started with Policies.
ToolCollection
Results of a Tool search. Contains both ToolSummary items and other information, such as metadata.
ToolConfig
The configuration and type of Tool.
ToolConfigModelConverter
ToolSummary
Summary information of a Tool.
UpdateAgentDetails
The data to update an agent.
UpdateAgentEndpointDetails
The data to update an endpoint.
UpdateDataSourceDetails
The data to update a data source.
UpdateKnowledgeBaseDetails
The data to update a knowledge base.
UpdateToolDetails
The data to update a Tool.
WorkRequest
An asynchronous work request. Work requests help you monitor long-running operations. When you start a long-running operation, the service creates a work request. A work request is an activity log that lets you track each step in the operation's progress. Each work request has an OCID that lets you interact with it programmatically and use it for automation.
WorkRequestError
An error encountered while performing an operation that is tracked by a work request.
WorkRequestErrorCollection
A list of work request errors. Can contain both errors and other information, such as metadata.
WorkRequestLogEntry
A log message from performing an operation that is tracked by a work request.
WorkRequestLogEntryCollection
A list of work request logs. Can contain both logs and other information, such as metadata.
WorkRequestResource
A resource created or operated on by a work request.
WorkRequestSummary
Summary information about an asynchronous work request.
WorkRequestSummaryCollection
A list of work requests. Can contain both work requests and other information, such as metadata.
Enums
ActionType
Possible types of actions.
Agent.LifecycleStateEnum
AgentEndpoint.LifecycleStateEnum
ApiSchemaInputLocation.ApiSchemaInputLocationTypeEnum
DataIngestionJob.LifecycleStateEnum
DataSource.LifecycleStateEnum
DataSourceConfig.DataSourceConfigTypeEnum
DatabaseConnection.ConnectionTypeEnum
GuardrailMode
Configuration setting applied to determine the guardrail mode for input or output.
HttpEndpointAuthConfig.HttpEndpointAuthConfigTypeEnum
IndexConfig.IndexConfigTypeEnum
InputLocation.InputLocationTypeEnum
KnowledgeBase.LifecycleStateEnum
OperationStatus
The status of the work request.
OperationType
The asynchronous operation tracked by the work request.
OutputLocation.OutputLocationTypeEnum
SecretDetail.TypeEnum
SortOrder
The sort order to use, either ascending (ASC
) or descending (DESC
). The displayName
sort order is case sensitive.
SqlToolConfig.DialectEnum
SqlToolConfig.ModelSizeEnum
Tool.LifecycleStateEnum
ToolConfig.ToolConfigTypeEnum
WorkRequestResourceMetadataKey
Possible metadata keys for work request resource metadata.