Create an Agent
post
/20260430/aiDataPlatforms/{aiDataPlatformId}/workspaces/{workspaceKey}/agents
Create an Agent.
Request
Path Parameters
-
aiDataPlatformId(required): string
Minimum Length:
1Maximum Length:500The [OCID](/iaas/Content/General/Concepts/identifiers.htm) of the AI Data Platform (Data Lake) instance. -
workspaceKey(required): string
Minimum Length:
1Maximum Length:200The key of the Workspace
Header Parameters
-
opc-request-id: string
Maximum Length:
255Unique Oracle-assigned identifier for the request. If you need to contact Oracle about a particular request, please provide the request ID. The only valid characters for request IDs are letters, numbers, underscore, and dash. -
opc-retry-token: string
Maximum Length:
64A token that uniquely identifies a request so it can be retried in case of a timeout or server error without risk of running that same action again. Retry tokens expire after 24 hours, but can be invalidated before then due to conflicting operations. For example, if a resource has been deleted and removed from the system, then a retry of the original creation request might be rejected. -
should-update-recent: boolean
A flag to identify if the recent list should be updated.Default Value:
false
Details for the new Agent.
Root Schema : CreateAgentDetails
Type:
objectThe data to create a agent.
Show Source
-
agentCardConfig:
object AgentCardConfigDetail
Details of the A2A publish request payload.
-
computeKey:
string
Minimum Length:
1Maximum Length:255The key of the agent compute associated with this agent. -
dependenciesFilePath:
string
The path to dependencies file.
-
description:
string
Agent description.
-
diagram:
object AgentDiagram
Diagrammatic representation of the agent with all node and edge details.
-
displayName(required):
string
Agent name.
-
entryFilePath:
string
The path to project entry file.
-
guardrails:
object GuardrailsConfiguration
Safety guardrails configuration for an Agent.
-
pathInfo(required):
string
Path inside volume where the agent JSON is written.
-
sessionConfig:
object SessionConfiguration
Session configuration for an Agent.
-
type:
string
Allowed Values:
[ "CANVAS", "CODE" ]The type of agent (Canvas or Code).
Nested Schema : AgentCardConfigDetail
Type:
objectDetails of the A2A publish request payload.
Show Source
-
capabilities:
object AgentCardCapabilitiesDetail
Capabilities included in publish request.
-
description:
string
Maximum Length:
1048576Human-readable agent description. -
documentationUrl:
string
Documentation URL for the agent. Serialized as `documentation_url`.
-
name(required):
string
Minimum Length:
1Maximum Length:255Human-readable agent name. -
provider:
object AgentProvider
Service provider information for an agent.
-
skills:
array skills
List of skills supported by the agent.
-
version:
string
Agent version string.
Nested Schema : AgentDiagram
Type:
objectDiagrammatic representation of the agent with all node and edge details.
Show Source
-
description:
string
Agent description.
-
displayName:
string
Agent name.
-
edges:
object edges
Additional Properties Allowed: AgentEdgeMapping of edgeId to edge objects.
-
guardrailsMap:
object guardrailsMap
Additional Properties Allowed: GuardrailsConfigurationA hash map with key=guardrails config key, value=guardrails definition. It is used to find the guardrails definition for a guardrails reference by name.
-
key:
string
The unique identifier (UUID) of the agent.
-
modelVersion:
string
The model and upgrade compatibility version for this agent diagram.
-
nodes:
object nodes
Additional Properties Allowed: AgentDiagramNodeMapping of nodeId to node objects.
-
toolsMap:
object toolsMap
Additional Properties Allowed: ToolA hash map with key=tool key, value=tool definition. It is used to find the tool definition for a tool reference, where the tool is marked as a reference.
Nested Schema : GuardrailsConfiguration
Type:
objectSafety guardrails configuration for an Agent.
Show Source
-
description:
string
Maximum Length:
1000Description of the guardrails configuration -
key:
string
The unique identifier (UUID) of the guardrails configuration.
-
name:
string
Maximum Length:
255Name of the guardrails configuration -
policies:
array policies
List of safety policies configured in this guardrails
Nested Schema : SessionConfiguration
Type:
objectSession configuration for an Agent.
Show Source
-
sessionRetentionConfig:
object SessionRetentionConfiguration
Session data retention configuration for an Agent.
-
variables:
object variables
Additional Properties Allowed: SessionVariableDetailsMap of session variable name to variable definition.
Nested Schema : AgentCardCapabilitiesDetail
Type:
objectCapabilities included in publish request.
Show Source
-
isStreaming:
boolean
Indicates whether streaming responses are supported.
Nested Schema : AgentProvider
Type:
objectService provider information for an agent.
Show Source
-
organization:
string
The name of the agent provider's organization.
-
url:
string
A URL for the agent provider's website or documentation.
Nested Schema : skills
Type:
arrayList of skills supported by the agent.
Show Source
-
Array of:
object AgentCardSkillDetail
A single skill in the publish request.
Nested Schema : AgentCardSkillDetail
Type:
objectA single skill in the publish request.
Show Source
-
description:
string
Maximum Length:
1048576Description of what the skill does. -
examples:
array examples
Optional example prompts for this skill.
-
id(required):
string
Minimum Length:
1Maximum Length:255Unique skill identifier. -
name(required):
string
Minimum Length:
1Maximum Length:255Display name of the skill. -
tags:
array tags
Optional tags for categorization/search.
Nested Schema : examples
Type:
arrayOptional example prompts for this skill.
Show Source
-
Array of:
string
Minimum Length:
1Maximum Length:1048576
Nested Schema : edges
Type:
objectAdditional Properties Allowed
Show Source
Mapping of edgeId to edge objects.
Default Value:
{
}Nested Schema : guardrailsMap
Type:
objectAdditional Properties Allowed
Show Source
-
object GuardrailsConfiguration
Safety guardrails configuration for an Agent.
A hash map with key=guardrails config key, value=guardrails definition. It is used to find the guardrails definition for a guardrails reference by name.
Nested Schema : nodes
Type:
objectAdditional Properties Allowed
Show Source
-
object AgentDiagramNode
Discriminator:
typeA node in a diagram, representing logic, data, or operation.
Mapping of nodeId to node objects.
Default Value:
{
}Nested Schema : toolsMap
Type:
objectAdditional Properties Allowed
Show Source
-
object Tool
Discriminator:
toolTypeA tool is a function that an agent can call. Tools can retrieve data, call external APIs, HTTP endpoints, execute a snippet of code, entire Python scripts, etc. | A tool is stateless, doesn't reason (it just executes), and can be reused across agents.
A hash map with key=tool key, value=tool definition. It is used to find the tool definition for a tool reference, where the tool is marked as a reference.
Nested Schema : AgentEdge
Type:
objectAn edge connecting two nodes in the diagram, with context and metadata.
Show Source
-
destNodeId(required):
string
Destination node for this edge.
-
destNodeInput(required):
string
Input port on the destination node.
-
edgeHandles:
array edgeHandles
Geometry handle coordinates.
-
edgeProperties:
object edgeProperties
Additional Properties Allowed: additionalPropertiesExtensible properties of the edge.
-
edgeType:
string
Style/type for edge.
-
key(required):
string
Unique edge identifier.
-
parentNodeId:
string
Parent node, if hierarchical.
-
srcNodeId(required):
string
Source node for this edge.
-
srcNodeOutput(required):
string
Output port on the source node.
-
timeCreated:
string(date-time)
Default Value:
2024-12-11T22:41:25.622463ZRFC3339 timestamp when edge was created. -
timeUpdated:
string(date-time)
Default Value:
2024-12-11T22:41:25.622470ZRFC3339 timestamp when edge was last updated. -
validationErrors:
array validationErrors
List of validation errors encountered in the diagram.
Nested Schema : edgeHandles
Type:
arrayGeometry handle coordinates.
Show Source
-
Array of:
object Point
A 2D point for edge handles on diagram elements.
Nested Schema : edgeProperties
Type:
objectAdditional Properties Allowed
Show Source
Extensible properties of the edge.
Nested Schema : validationErrors
Type:
arrayList of validation errors encountered in the diagram.
Show Source
-
Array of:
object ValidationError
Validation error for a component in agent diagram.
Nested Schema : Point
Type:
objectA 2D point for edge handles on diagram elements.
Show Source
-
x(required):
number(float)
X coordinate.
-
y(required):
number(float)
Y coordinate.
Nested Schema : additionalProperties
Type:
objectNested Schema : ValidationError
Type:
objectValidation error for a component in agent diagram.
Show Source
-
key(required):
string
Id of the component that caused the error
-
message(required):
string
Error message.
-
name(required):
string
Error name
-
type(required):
string
Type of error.
Nested Schema : AgentDiagramNode
Type:
objectDiscriminator:
typeA node in a diagram, representing logic, data, or operation.
Show Source
-
configuration:
object configuration
Additional Properties Allowed: additionalPropertiesConfiguration object for this node.
-
description:
string
Default Value:
New nodeDescription of this node. -
inputs:
array inputs
Array of NodeInput objects.
-
isExpanded:
boolean
Default Value:
falseIs node expanded in UI. -
key(required):
string
Unique identifier for the node.
-
name:
string
Default Value:
New nodeName of this node. -
nodeType:
string
This field is deprecated. | It does not need to be set to any value for API calls.
-
nodeTypeId:
string
Definition ID for this node type.
-
outputs:
array outputs
Array of NodeOutput objects.
-
parentNodeId:
string
Parent node ID, for subgraphs.
-
positionX:
number(float)
Default Value:
0Canvas X coordinate. -
positionY:
number(float)
Default Value:
0Canvas Y coordinate. -
timeCreated:
string(date-time)
Default Value:
2024-12-11T22:41:25.967777ZRFC3339 timestamp when node was created. -
timeUpdated:
string(date-time)
Default Value:
2024-12-11T22:41:25.967783ZRFC3339 timestamp when node was last updated. -
type:
string
Allowed Values:
[ "START_NODE", "GUARDRAILS", "AGENT", "SUPERVISOR_AGENT", "SQL_TOOL", "PROMPT_TOOL", "RAG_TOOL", "MCP_TOOL", "HTTP_TOOL", "CUSTOM_TOOL" ]Node type discriminator. -
validationErrors:
array validationErrors
List of validation errors encountered in the diagram.
Nested Schema : configuration
Type:
objectAdditional Properties Allowed
Show Source
Configuration object for this node.
Default Value:
{
}Nested Schema : inputs
Type:
arrayArray of NodeInput objects.
Show Source
-
Array of:
object NodeInput
An input slot for a node, including type and requirements.
Nested Schema : outputs
Type:
arrayArray of NodeOutput objects.
Show Source
-
Array of:
object NodeOutput
An output slot for a node, including type and requirements.
Nested Schema : validationErrors
Type:
arrayList of validation errors encountered in the diagram.
Show Source
-
Array of:
object ValidationError
Validation error for a component in agent diagram.
Nested Schema : additionalProperties
Type:
objectDefault Value:
{
}Nested Schema : NodeInput
Type:
objectAn input slot for a node, including type and requirements.
Show Source
-
description:
string
Documentation or tooltip shown in the UI.
-
isRequired(required):
boolean
True if this input is mandatory.
-
isSingleConnection:
boolean
Default Value:
falseTrue if only one connection is allowed; false allows multiple. -
name(required):
string
Unique name for the input slot.
-
portPosition:
string
Allowed Values:
[ "TOP", "BOTTOM", "RIGHT", "LEFT" ]UI position of the input port on the node. -
type(required):
object NodeIo
Defines the data type and color for an input or output.
Nested Schema : NodeIo
Type:
objectDefines the data type and color for an input or output.
Show Source
-
color(required):
object NodeIoColor
Text and background color configuration for a node input/output.
-
dataType(required):
string
Allowed Values:
[ "FLOW", "STR", "INT", "BOOL", "FLOAT", "LIST", "DICT", "ANY", "TOOLS", "LLMS", "AGENTS", "PLANNER", "FLOWS" ]Possible types of node inputs and outputs.
Nested Schema : NodeIoColor
Type:
objectText and background color configuration for a node input/output.
Show Source
-
backgroundColor(required):
string
Input/output badge background color.
-
handleColor(required):
string
Input/output connector handle color.
-
textColor(required):
string
Allowed Values:
[ "DARK", "LIGHT" ]Whether the text on badges should be dark or light for visibility.
Nested Schema : NodeOutput
Type:
objectAn output slot for a node, including type and requirements.
Show Source
-
description:
string
Documentation or tooltip shown in the UI.
-
isRequired(required):
boolean
True if this output is mandatory.
-
name(required):
string
Unique name for the output slot.
-
portPosition:
string
Allowed Values:
[ "TOP", "BOTTOM", "RIGHT", "LEFT" ]UI position of the output port on the node. -
type(required):
object NodeIo
Defines the data type and color for an input or output.
Nested Schema : Tool
Type:
objectDiscriminator:
toolTypeA tool is a function that an agent can call. Tools can retrieve data, call external APIs, HTTP endpoints, execute a snippet of code, entire Python scripts, etc. | A tool is stateless, doesn't reason (it just executes), and can be reused across agents.
Show Source
-
createdBy:
string
Minimum Length:
1Maximum Length:255The OCID of the user/principal who created the tool. -
description:
string
Tool description.
-
displayName:
string
Tool name.
-
key:
string
The unique identifier of the tool
-
positionX:
number(float)
Default Value:
0Canvas X coordinate of the Tool node -
positionY:
number(float)
Default Value:
0Canvas Y coordinate of the Tool node -
properties:
object properties
Additional Properties Allowed: additionalPropertiesA list of key-value pairs to use for configuring the tool
-
timeCreated:
string(date-time)
The date and time the tool was created.
-
timeUpdated:
string(date-time)
The date and time the tool was updated.
-
toolType:
string
Allowed Values:
[ "RAG", "SQL", "PROMPT", "NL2SQL", "MCP", "CUSTOM", "HTTP" ]Type of tool. Managed, external or mount tool. -
updatedBy:
string
Minimum Length:
1Maximum Length:255The ID of the user who last updated the schema. -
workspaceKey:
string
The key of the Workspace to which this tool belongs.
Nested Schema : properties
Type:
objectAdditional Properties Allowed
Show Source
A list of key-value pairs to use for configuring the tool
Nested Schema : additionalProperties
Type:
objectNested Schema : policies
Type:
arrayList of safety policies configured in this guardrails
Show Source
-
Array of:
object SafetyPolicy
Discriminator:
policyTypeIndividual safety policy configuration.
Nested Schema : SafetyPolicy
Type:
objectDiscriminator:
policyTypeIndividual safety policy configuration.
Show Source
-
action(required):
string
Allowed Values:
[ "BLOCK", "INFORM", "MASK" ]Action to take when policy is violated. -
policyDescription:
string
Maximum Length:
1000Description of the policy. -
policyName:
string
Maximum Length:
255Custom name for the policy. -
policyType(required):
string
Allowed Values:
[ "CONTENT_MODERATION", "PROMPT_ATTACKS_PREVENTION", "PII_DETECTION", "DENIED_TOPICS", "WORD_FILTERS", "CONTEXTUAL_GROUNDING", "CUSTOM_POLICY" ]Type of safety policy. -
scope(required):
string
Allowed Values:
[ "USER_REQUEST", "AGENT_RESPONSE", "BOTH" ]Scope of policy application. -
threshold:
number(float)
Minimum Value:
0Maximum Value:1Default Value:0.5Threshold value for policy violation (0.0 to 1.0).
Nested Schema : SessionRetentionConfiguration
Type:
objectSession data retention configuration for an Agent.
Show Source
-
retentionPeriodInDays:
integer
Minimum Value:
1No. of days session data will be kept -
sessionSizeLimit:
integer
Minimum Value:
1Max storage allocated to session data (in MB). -
threadCountLimit:
integer
Minimum Value:
1Maximum no. of user prompt and agent response pairs per session
Nested Schema : variables
Type:
objectAdditional Properties Allowed
Show Source
-
object SessionVariableDetails
Session Variable Details.
Map of session variable name to variable definition.
Default Value:
{
}Nested Schema : SessionVariableDetails
Type:
objectSession Variable Details.
Show Source
-
description:
string
Description of the Session Variable
-
isRequired:
boolean
Default Value:
falseIf this Variable is required or not -
isSystem:
boolean
Default Value:
falseTrue if Session Variable is defined by System -
name(required):
string
Name of the Session Variable.
-
shouldLog:
boolean
Default Value:
trueIf we should log this Session Variable or not -
value:
string
Default Value of this Session Variable
Response
Supported Media Types
- application/json
200 Response
Request accepted. Agent is created.
Headers
-
content-location: string(uri-reference)
Same as location.
-
etag: string
For optimistic concurrency control. See `if-match`.
-
location: string(uri-reference)
URL for the created Agent. The Agent key is generated after this request is sent.
-
opc-request-id: string
Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.
Root Schema : Agent
Type:
objectAn agent is a sequence of nodes and edges defining an end-to-end agentic application. | A diagram can be composed of multiple nodes, each node being a single agent or a multi-agent system, working independently | or collaboratively to accomplish an overall objective.
Show Source
-
agentCardConfig:
object AgentCardConfigDetail
Details of the A2A publish request payload.
-
computeKey:
string
Minimum Length:
1Maximum Length:255The key of the agent compute associated with this agent. -
createdBy:
string
Minimum Length:
1Maximum Length:255The OCID of the user/principal who created the agent. -
dependenciesFilePath:
string
The path to dependencies file.
-
deploymentComputeKey:
string
The key of the compute where agent is deployed.
-
deploymentMode:
string
Agent deployment mode.
-
description(required):
string
Agent description.
-
diagram:
object AgentDiagram
Diagrammatic representation of the agent with all node and edge details.
-
displayName(required):
string
Agent name.
-
entryFilePath:
string
The path to project entry file.
-
guardrails:
object GuardrailsConfiguration
Safety guardrails configuration for an Agent.
-
key(required):
string
The unique identifier (UUID) of the agent.
-
lifecycleDetails:
string
A message that describes the current state of the agent in more detail. For example, can be used to provide actionable information for a resource in the Failed state.
-
lifecycleState(required):
string
Allowed Values:
[ "DRAFT", "DEPLOYED" ]The current state of the agent. -
pathInfo(required):
string
Path inside volume where the agent JSON is written.
-
sessionConfig:
object SessionConfiguration
Session configuration for an Agent.
-
timeCreated:
string(date-time)
The date and time the agent was created.
-
timeUpdated:
string(date-time)
The date and time the agent was updated.
-
type:
string
Allowed Values:
[ "CANVAS", "CODE" ]The type of agent (Canvas or Code). -
updatedBy:
string
Minimum Length:
1Maximum Length:255The ID of the user who last updated the schema. -
uri:
string
Agent URI.
-
uriState:
string
Agent URI state.
-
version:
integer
A number indicating the version of the record. Each time the record is updated, this version will be incremented. This will be used for optimistic locking.
-
workspaceKey:
string
The key of the workspace to which this agent belongs.
Nested Schema : AgentCardConfigDetail
Type:
objectDetails of the A2A publish request payload.
Show Source
-
capabilities:
object AgentCardCapabilitiesDetail
Capabilities included in publish request.
-
description:
string
Maximum Length:
1048576Human-readable agent description. -
documentationUrl:
string
Documentation URL for the agent. Serialized as `documentation_url`.
-
name(required):
string
Minimum Length:
1Maximum Length:255Human-readable agent name. -
provider:
object AgentProvider
Service provider information for an agent.
-
skills:
array skills
List of skills supported by the agent.
-
version:
string
Agent version string.
Nested Schema : AgentDiagram
Type:
objectDiagrammatic representation of the agent with all node and edge details.
Show Source
-
description:
string
Agent description.
-
displayName:
string
Agent name.
-
edges:
object edges
Additional Properties Allowed: AgentEdgeMapping of edgeId to edge objects.
-
guardrailsMap:
object guardrailsMap
Additional Properties Allowed: GuardrailsConfigurationA hash map with key=guardrails config key, value=guardrails definition. It is used to find the guardrails definition for a guardrails reference by name.
-
key:
string
The unique identifier (UUID) of the agent.
-
modelVersion:
string
The model and upgrade compatibility version for this agent diagram.
-
nodes:
object nodes
Additional Properties Allowed: AgentDiagramNodeMapping of nodeId to node objects.
-
toolsMap:
object toolsMap
Additional Properties Allowed: ToolA hash map with key=tool key, value=tool definition. It is used to find the tool definition for a tool reference, where the tool is marked as a reference.
Nested Schema : GuardrailsConfiguration
Type:
objectSafety guardrails configuration for an Agent.
Show Source
-
description:
string
Maximum Length:
1000Description of the guardrails configuration -
key:
string
The unique identifier (UUID) of the guardrails configuration.
-
name:
string
Maximum Length:
255Name of the guardrails configuration -
policies:
array policies
List of safety policies configured in this guardrails
Nested Schema : SessionConfiguration
Type:
objectSession configuration for an Agent.
Show Source
-
sessionRetentionConfig:
object SessionRetentionConfiguration
Session data retention configuration for an Agent.
-
variables:
object variables
Additional Properties Allowed: SessionVariableDetailsMap of session variable name to variable definition.
Nested Schema : AgentCardCapabilitiesDetail
Type:
objectCapabilities included in publish request.
Show Source
-
isStreaming:
boolean
Indicates whether streaming responses are supported.
Nested Schema : AgentProvider
Type:
objectService provider information for an agent.
Show Source
-
organization:
string
The name of the agent provider's organization.
-
url:
string
A URL for the agent provider's website or documentation.
Nested Schema : skills
Type:
arrayList of skills supported by the agent.
Show Source
-
Array of:
object AgentCardSkillDetail
A single skill in the publish request.
Nested Schema : AgentCardSkillDetail
Type:
objectA single skill in the publish request.
Show Source
-
description:
string
Maximum Length:
1048576Description of what the skill does. -
examples:
array examples
Optional example prompts for this skill.
-
id(required):
string
Minimum Length:
1Maximum Length:255Unique skill identifier. -
name(required):
string
Minimum Length:
1Maximum Length:255Display name of the skill. -
tags:
array tags
Optional tags for categorization/search.
Nested Schema : examples
Type:
arrayOptional example prompts for this skill.
Show Source
-
Array of:
string
Minimum Length:
1Maximum Length:1048576
Nested Schema : edges
Type:
objectAdditional Properties Allowed
Show Source
Mapping of edgeId to edge objects.
Default Value:
{
}Nested Schema : guardrailsMap
Type:
objectAdditional Properties Allowed
Show Source
-
object GuardrailsConfiguration
Safety guardrails configuration for an Agent.
A hash map with key=guardrails config key, value=guardrails definition. It is used to find the guardrails definition for a guardrails reference by name.
Nested Schema : nodes
Type:
objectAdditional Properties Allowed
Show Source
-
object AgentDiagramNode
Discriminator:
typeA node in a diagram, representing logic, data, or operation.
Mapping of nodeId to node objects.
Default Value:
{
}Nested Schema : toolsMap
Type:
objectAdditional Properties Allowed
Show Source
-
object Tool
Discriminator:
toolTypeA tool is a function that an agent can call. Tools can retrieve data, call external APIs, HTTP endpoints, execute a snippet of code, entire Python scripts, etc. | A tool is stateless, doesn't reason (it just executes), and can be reused across agents.
A hash map with key=tool key, value=tool definition. It is used to find the tool definition for a tool reference, where the tool is marked as a reference.
Nested Schema : AgentEdge
Type:
objectAn edge connecting two nodes in the diagram, with context and metadata.
Show Source
-
destNodeId(required):
string
Destination node for this edge.
-
destNodeInput(required):
string
Input port on the destination node.
-
edgeHandles:
array edgeHandles
Geometry handle coordinates.
-
edgeProperties:
object edgeProperties
Additional Properties Allowed: additionalPropertiesExtensible properties of the edge.
-
edgeType:
string
Style/type for edge.
-
key(required):
string
Unique edge identifier.
-
parentNodeId:
string
Parent node, if hierarchical.
-
srcNodeId(required):
string
Source node for this edge.
-
srcNodeOutput(required):
string
Output port on the source node.
-
timeCreated:
string(date-time)
Default Value:
2024-12-11T22:41:25.622463ZRFC3339 timestamp when edge was created. -
timeUpdated:
string(date-time)
Default Value:
2024-12-11T22:41:25.622470ZRFC3339 timestamp when edge was last updated. -
validationErrors:
array validationErrors
List of validation errors encountered in the diagram.
Nested Schema : edgeHandles
Type:
arrayGeometry handle coordinates.
Show Source
-
Array of:
object Point
A 2D point for edge handles on diagram elements.
Nested Schema : edgeProperties
Type:
objectAdditional Properties Allowed
Show Source
Extensible properties of the edge.
Nested Schema : validationErrors
Type:
arrayList of validation errors encountered in the diagram.
Show Source
-
Array of:
object ValidationError
Validation error for a component in agent diagram.
Nested Schema : Point
Type:
objectA 2D point for edge handles on diagram elements.
Show Source
-
x(required):
number(float)
X coordinate.
-
y(required):
number(float)
Y coordinate.
Nested Schema : additionalProperties
Type:
objectNested Schema : ValidationError
Type:
objectValidation error for a component in agent diagram.
Show Source
-
key(required):
string
Id of the component that caused the error
-
message(required):
string
Error message.
-
name(required):
string
Error name
-
type(required):
string
Type of error.
Nested Schema : AgentDiagramNode
Type:
objectDiscriminator:
typeA node in a diagram, representing logic, data, or operation.
Show Source
-
configuration:
object configuration
Additional Properties Allowed: additionalPropertiesConfiguration object for this node.
-
description:
string
Default Value:
New nodeDescription of this node. -
inputs:
array inputs
Array of NodeInput objects.
-
isExpanded:
boolean
Default Value:
falseIs node expanded in UI. -
key(required):
string
Unique identifier for the node.
-
name:
string
Default Value:
New nodeName of this node. -
nodeType:
string
This field is deprecated. | It does not need to be set to any value for API calls.
-
nodeTypeId:
string
Definition ID for this node type.
-
outputs:
array outputs
Array of NodeOutput objects.
-
parentNodeId:
string
Parent node ID, for subgraphs.
-
positionX:
number(float)
Default Value:
0Canvas X coordinate. -
positionY:
number(float)
Default Value:
0Canvas Y coordinate. -
timeCreated:
string(date-time)
Default Value:
2024-12-11T22:41:25.967777ZRFC3339 timestamp when node was created. -
timeUpdated:
string(date-time)
Default Value:
2024-12-11T22:41:25.967783ZRFC3339 timestamp when node was last updated. -
type:
string
Allowed Values:
[ "START_NODE", "GUARDRAILS", "AGENT", "SUPERVISOR_AGENT", "SQL_TOOL", "PROMPT_TOOL", "RAG_TOOL", "MCP_TOOL", "HTTP_TOOL", "CUSTOM_TOOL" ]Node type discriminator. -
validationErrors:
array validationErrors
List of validation errors encountered in the diagram.
Nested Schema : configuration
Type:
objectAdditional Properties Allowed
Show Source
Configuration object for this node.
Default Value:
{
}Nested Schema : inputs
Type:
arrayArray of NodeInput objects.
Show Source
-
Array of:
object NodeInput
An input slot for a node, including type and requirements.
Nested Schema : outputs
Type:
arrayArray of NodeOutput objects.
Show Source
-
Array of:
object NodeOutput
An output slot for a node, including type and requirements.
Nested Schema : validationErrors
Type:
arrayList of validation errors encountered in the diagram.
Show Source
-
Array of:
object ValidationError
Validation error for a component in agent diagram.
Nested Schema : additionalProperties
Type:
objectDefault Value:
{
}Nested Schema : NodeInput
Type:
objectAn input slot for a node, including type and requirements.
Show Source
-
description:
string
Documentation or tooltip shown in the UI.
-
isRequired(required):
boolean
True if this input is mandatory.
-
isSingleConnection:
boolean
Default Value:
falseTrue if only one connection is allowed; false allows multiple. -
name(required):
string
Unique name for the input slot.
-
portPosition:
string
Allowed Values:
[ "TOP", "BOTTOM", "RIGHT", "LEFT" ]UI position of the input port on the node. -
type(required):
object NodeIo
Defines the data type and color for an input or output.
Nested Schema : NodeIo
Type:
objectDefines the data type and color for an input or output.
Show Source
-
color(required):
object NodeIoColor
Text and background color configuration for a node input/output.
-
dataType(required):
string
Allowed Values:
[ "FLOW", "STR", "INT", "BOOL", "FLOAT", "LIST", "DICT", "ANY", "TOOLS", "LLMS", "AGENTS", "PLANNER", "FLOWS" ]Possible types of node inputs and outputs.
Nested Schema : NodeIoColor
Type:
objectText and background color configuration for a node input/output.
Show Source
-
backgroundColor(required):
string
Input/output badge background color.
-
handleColor(required):
string
Input/output connector handle color.
-
textColor(required):
string
Allowed Values:
[ "DARK", "LIGHT" ]Whether the text on badges should be dark or light for visibility.
Nested Schema : NodeOutput
Type:
objectAn output slot for a node, including type and requirements.
Show Source
-
description:
string
Documentation or tooltip shown in the UI.
-
isRequired(required):
boolean
True if this output is mandatory.
-
name(required):
string
Unique name for the output slot.
-
portPosition:
string
Allowed Values:
[ "TOP", "BOTTOM", "RIGHT", "LEFT" ]UI position of the output port on the node. -
type(required):
object NodeIo
Defines the data type and color for an input or output.
Nested Schema : Tool
Type:
objectDiscriminator:
toolTypeA tool is a function that an agent can call. Tools can retrieve data, call external APIs, HTTP endpoints, execute a snippet of code, entire Python scripts, etc. | A tool is stateless, doesn't reason (it just executes), and can be reused across agents.
Show Source
-
createdBy:
string
Minimum Length:
1Maximum Length:255The OCID of the user/principal who created the tool. -
description:
string
Tool description.
-
displayName:
string
Tool name.
-
key:
string
The unique identifier of the tool
-
positionX:
number(float)
Default Value:
0Canvas X coordinate of the Tool node -
positionY:
number(float)
Default Value:
0Canvas Y coordinate of the Tool node -
properties:
object properties
Additional Properties Allowed: additionalPropertiesA list of key-value pairs to use for configuring the tool
-
timeCreated:
string(date-time)
The date and time the tool was created.
-
timeUpdated:
string(date-time)
The date and time the tool was updated.
-
toolType:
string
Allowed Values:
[ "RAG", "SQL", "PROMPT", "NL2SQL", "MCP", "CUSTOM", "HTTP" ]Type of tool. Managed, external or mount tool. -
updatedBy:
string
Minimum Length:
1Maximum Length:255The ID of the user who last updated the schema. -
workspaceKey:
string
The key of the Workspace to which this tool belongs.
Nested Schema : properties
Type:
objectAdditional Properties Allowed
Show Source
A list of key-value pairs to use for configuring the tool
Nested Schema : additionalProperties
Type:
objectNested Schema : policies
Type:
arrayList of safety policies configured in this guardrails
Show Source
-
Array of:
object SafetyPolicy
Discriminator:
policyTypeIndividual safety policy configuration.
Nested Schema : SafetyPolicy
Type:
objectDiscriminator:
policyTypeIndividual safety policy configuration.
Show Source
-
action(required):
string
Allowed Values:
[ "BLOCK", "INFORM", "MASK" ]Action to take when policy is violated. -
policyDescription:
string
Maximum Length:
1000Description of the policy. -
policyName:
string
Maximum Length:
255Custom name for the policy. -
policyType(required):
string
Allowed Values:
[ "CONTENT_MODERATION", "PROMPT_ATTACKS_PREVENTION", "PII_DETECTION", "DENIED_TOPICS", "WORD_FILTERS", "CONTEXTUAL_GROUNDING", "CUSTOM_POLICY" ]Type of safety policy. -
scope(required):
string
Allowed Values:
[ "USER_REQUEST", "AGENT_RESPONSE", "BOTH" ]Scope of policy application. -
threshold:
number(float)
Minimum Value:
0Maximum Value:1Default Value:0.5Threshold value for policy violation (0.0 to 1.0).
Nested Schema : SessionRetentionConfiguration
Type:
objectSession data retention configuration for an Agent.
Show Source
-
retentionPeriodInDays:
integer
Minimum Value:
1No. of days session data will be kept -
sessionSizeLimit:
integer
Minimum Value:
1Max storage allocated to session data (in MB). -
threadCountLimit:
integer
Minimum Value:
1Maximum no. of user prompt and agent response pairs per session
Nested Schema : variables
Type:
objectAdditional Properties Allowed
Show Source
-
object SessionVariableDetails
Session Variable Details.
Map of session variable name to variable definition.
Default Value:
{
}Nested Schema : SessionVariableDetails
Type:
objectSession Variable Details.
Show Source
-
description:
string
Description of the Session Variable
-
isRequired:
boolean
Default Value:
falseIf this Variable is required or not -
isSystem:
boolean
Default Value:
falseTrue if Session Variable is defined by System -
name(required):
string
Name of the Session Variable.
-
shouldLog:
boolean
Default Value:
trueIf we should log this Session Variable or not -
value:
string
Default Value of this Session Variable
400 Response
Bad Request (invalid query parameters, malformed headers, and so on).
Headers
-
opc-request-id: string
Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.
Root Schema : Error
Type:
objectError information.
Show Source
-
code(required):
string
A short error code that defines the error, meant for programmatic parsing.
-
message(required):
string
A human-readable error message.
401 Response
Unauthorized (missing or expired credentials, and so on).
Headers
-
opc-request-id: string
Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.
Root Schema : Error
Type:
objectError information.
Show Source
-
code(required):
string
A short error code that defines the error, meant for programmatic parsing.
-
message(required):
string
A human-readable error message.
404 Response
Not Found. The requested resource was not found.
Headers
-
opc-request-id: string
Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.
Root Schema : Error
Type:
objectError information.
Show Source
-
code(required):
string
A short error code that defines the error, meant for programmatic parsing.
-
message(required):
string
A human-readable error message.
409 Response
Conflict. Request conflicts with the current state of the resource.
Headers
-
opc-request-id: string
Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.
Root Schema : Error
Type:
objectError information.
Show Source
-
code(required):
string
A short error code that defines the error, meant for programmatic parsing.
-
message(required):
string
A human-readable error message.
429 Response
Too Many Requests. Too many requests sent to the server in a short period.
Headers
-
opc-request-id: string
Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.
Root Schema : Error
Type:
objectError information.
Show Source
-
code(required):
string
A short error code that defines the error, meant for programmatic parsing.
-
message(required):
string
A human-readable error message.
500 Response
Internal Server Error. The server encountered an unexpected condition preventing fulfilment of the request.
Headers
-
opc-request-id: string
Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.
Root Schema : Error
Type:
objectError information.
Show Source
-
code(required):
string
A short error code that defines the error, meant for programmatic parsing.
-
message(required):
string
A human-readable error message.
Default Response
Unknown Error. Error is not recognized by the system.
Headers
-
opc-request-id: string
Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.
Root Schema : Error
Type:
objectError information.
Show Source
-
code(required):
string
A short error code that defines the error, meant for programmatic parsing.
-
message(required):
string
A human-readable error message.