REST Resource: projects.generators

Resource: Generator

LLM generator.

JSON representation 
{
  "name": string,
  "description": string,
  "inferenceParameter": {
    object (InferenceParameter)
  },
  "triggerEvent": enum (TriggerEvent),
  "createTime": string,
  "updateTime": string,
  "tools": [
    string
  ],
  "suggestionDedupingConfig": {
    object (SuggestionDedupingConfig)
  },

  // Union field context can be only one of the following:
  "freeFormContext": {
    object (FreeFormContext)
  },
  "agentCoachingContext": {
    object (AgentCoachingContext)
  },
  "summarizationContext": {
    object (SummarizationContext)
  }
  // End of list of possible types for union field context.

  // Union field foundation_model can be only one of the following:
  "publishedModel": string
  // End of list of possible types for union field foundation_model.
}
Fields
name

string

Output only. Identifier. The resource name of the generator. Format: projects/<Project ID>/locations/<Location ID>/generators/<Generator ID>
description

string

Optional. Human readable description of the generator.
inferenceParameter

object (InferenceParameter)

Optional. Inference parameters for this generator.
triggerEvent

enum (TriggerEvent)

Optional. The trigger event of the generator. It defines when the generator is triggered in a conversation.
createTime

string (Timestamp format)

Output only. Creation time of this generator.

Uses RFC 3339, where generated output will always be Z-normalized and use 0, 3, 6 or 9 fractional digits. Offsets other than "Z" are also accepted. Examples: "2014-10-02T15:01:23Z", "2014-10-02T15:01:23.045123456Z" or "2014-10-02T15:01:23+05:30".
updateTime

string (Timestamp format)

Output only. Update time of this generator.

Uses RFC 3339, where generated output will always be Z-normalized and use 0, 3, 6 or 9 fractional digits. Offsets other than "Z" are also accepted. Examples: "2014-10-02T15:01:23Z", "2014-10-02T15:01:23.045123456Z" or "2014-10-02T15:01:23+05:30".
tools[]

string

Optional. Resource names of the tools that the generator can choose from. Format: projects/<Project ID>/locations/<Location ID>/tools/<tool ID>.
suggestionDedupingConfig

object (SuggestionDedupingConfig)

Optional. Configuration for suggestion deduping. This is only applicable to AI Coach feature.
Union field context. Required. Input context of the generator. context can be only one of the following:
freeFormContext

object (FreeFormContext)

Input of free from generator to LLM.
agentCoachingContext

object (AgentCoachingContext)

Input of prebuilt Agent Coaching feature.
summarizationContext

object (SummarizationContext)

Input of prebuilt Summarization feature.
Union field foundation_model. The foundation model to use for generating suggestions. If a foundation model isn't specified here, a model specifically tuned for the feature type (and version when applicable) will be used. foundation_model can be only one of the following:
publishedModel

string

Optional. The published Large Language Model name. * To use the latest model version, specify the model name without version number. Example: text-bison * To use a stable model version, specify the version number as well. Example: text-bison@002.

FreeFormContext

Free form generator context that customer can configure.

JSON representation 
{
  "text": string
}
Fields
text

string

Optional. Free form text input to LLM.

AgentCoachingContext

Agent Coaching context that customer can configure.

JSON representation 
{
  "overarchingGuidance": string,
  "instructions": [
    {
      object (AgentCoachingInstruction)
    }
  ],
  "version": string,
  "outputLanguageCode": string
}
Fields
overarchingGuidance

string

Optional. The overarching guidance for the agent coaching. This should be set only for v1.5 and later versions.
instructions[]

object (AgentCoachingInstruction)

Optional. Customized instructions for agent coaching.
version

string

Optional. Version of the feature. If not set, default to latest version. Current candidates are ["1.2"].
outputLanguageCode

string

Optional. Output language code.

AgentCoachingInstruction

Agent Coaching instructions that customer can configure.

JSON representation 
{
  "displayName": string,
  "displayDetails": string,
  "condition": string,
  "agentAction": string,
  "systemAction": string,
  "duplicateCheckResult": {
    object (DuplicateCheckResult)
  }
}
Fields
displayName

string

Optional. Display name for the instruction.
displayDetails

string

Optional. The detailed description of this instruction.
condition

string

Optional. The condition of the instruction. For example, "the customer wants to cancel an order". If the users want the instruction to be triggered unconditionally, the condition can be empty.
agentAction

string

Optional. The action that human agent should take. For example, "apologize for the slow shipping". If the users only want to use agent coaching for intent detection, agentAction can be empty
systemAction

string

Optional. The action that system should take. For example, "call GetOrderTime with order_number={order number provided by the customer}". If the users don't have plugins or don't want to trigger plugins, the systemAction can be empty
duplicateCheckResult

object (DuplicateCheckResult)

Output only. Duplication check for the AgentCoachingInstruction.

DuplicateCheckResult

Duplication check for the suggestion.

JSON representation 
{
  "duplicateSuggestions": [
    {
      object (DuplicateSuggestion)
    }
  ]
}
Fields
duplicateSuggestions[]

object (DuplicateSuggestion)

Output only. The duplicate suggestions.

DuplicateSuggestion

The duplicate suggestion details.

JSON representation 
{
  "answerRecord": string,
  "suggestionIndex": integer,
  "similarityScore": number
}
Fields
answerRecord

string

Output only. The answer record id of the past duplicate suggestion.
suggestionIndex

integer

Output only. The index of the duplicate suggestion in the past suggestion list.
similarityScore

number

Output only. The similarity score of between the past and current suggestion.

SummarizationContext

Summarization context that customer can configure.

JSON representation 
{
  "summarizationSections": [
    {
      object (SummarizationSection)
    }
  ],
  "fewShotExamples": [
    {
      object (FewShotExample)
    }
  ],
  "version": string,
  "outputLanguageCode": string
}
Fields
summarizationSections[]

object (SummarizationSection)

Optional. List of sections. Note it contains both predefined section sand customer defined sections.
fewShotExamples[]

object (FewShotExample)

Optional. List of few shot examples.
version

string

Optional. Version of the feature. If not set, default to latest version. Current candidates are ["1.0"].
outputLanguageCode

string

Optional. The target language of the generated summary. The language code for conversation will be used if this field is empty. Supported 2.0 and later versions.

SummarizationSection

Represents the section of summarization.

JSON representation 
{
  "key": string,
  "definition": string,
  "type": enum (Type)
}
Fields
key

string

Optional. Name of the section, for example, "situation".
definition

string

Optional. Definition of the section, for example, "what the customer needs help with or has question about."
type

enum (Type)

Optional. Type of the summarization section.

Type

Type enum of the summarization sections.

Enums
TYPE_UNSPECIFIED Undefined section type, does not return anything.
SITUATION What the customer needs help with or has question about. Section name: "situation".
ACTION What the agent does to help the customer. Section name: "action".
RESOLUTION Result of the customer service. A single word describing the result of the conversation. Section name: "resolution".
REASON_FOR_CANCELLATION Reason for cancellation if the customer requests for a cancellation. "N/A" otherwise. Section name: "reason_for_cancellation".
CUSTOMER_SATISFACTION "Unsatisfied" or "Satisfied" depending on the customer's feelings at the end of the conversation. Section name: "customer_satisfaction".
ENTITIES Key entities extracted from the conversation, such as ticket number, order number, dollar amount, etc. Section names are prefixed by "entities/".
CUSTOMER_DEFINED Customer defined sections.
SITUATION_CONCISE Concise version of the situation section. This type is only available if type SITUATION is not selected.
ACTION_CONCISE Concise version of the action section. This type is only available if type ACTION is not selected.

FewShotExample

Providing examples in the generator (i.e. building a few-shot generator) helps convey the desired format of the LLM response.

JSON representation 
{
  "conversationContext": {
    object (ConversationContext)
  },
  "extraInfo": {
    string: string,
    ...
  },
  "output": {
    object (GeneratorSuggestion)
  },

  // Union field instruction_list can be only one of the following:
  "summarizationSectionList": {
    object (SummarizationSectionList)
  }
  // End of list of possible types for union field instruction_list.
}
Fields
conversationContext

object (ConversationContext)

Optional. Conversation transcripts.
extraInfo

map (key: string, value: string)

Optional. Key is the placeholder field name in input, value is the value of the placeholder. E.g. instruction contains "@price", and ingested data has <"price", "10">

An object containing a list of "key": value pairs. Example: { "name": "wrench", "mass": "1.3kg", "count": "3" }.
output

object (GeneratorSuggestion)

Required. Example output of the model.
Union field instruction_list. Instruction list of this few_shot example. instruction_list can be only one of the following:
summarizationSectionList

object (SummarizationSectionList)

Summarization sections.

ConversationContext

Context of the conversation, including transcripts.

JSON representation 
{
  "messageEntries": [
    {
      object (MessageEntry)
    }
  ]
}
Fields
messageEntries[]

object (MessageEntry)

Optional. List of message transcripts in the conversation.

SummarizationSectionList

List of summarization sections.

JSON representation 
{
  "summarizationSections": [
    {
      object (SummarizationSection)
    }
  ]
}
Fields
summarizationSections[]

object (SummarizationSection)

Optional. Summarization sections.

GeneratorSuggestion

Suggestion generated using a Generator.

JSON representation 
{
  "toolCallInfo": [
    {
      object (ToolCallInfo)
    }
  ],

  // Union field suggestion can be only one of the following:
  "freeFormSuggestion": {
    object (FreeFormSuggestion)
  },
  "summarySuggestion": {
    object (SummarySuggestion)
  },
  "agentCoachingSuggestion": {
    object (AgentCoachingSuggestion)
  }
  // End of list of possible types for union field suggestion.
}
Fields
toolCallInfo[]

object (ToolCallInfo)

Optional. List of request and response for tool calls executed.
Union field suggestion. The suggestion could be one of the many types suggestion can be only one of the following:
freeFormSuggestion

object (FreeFormSuggestion)

Optional. Free form suggestion.
summarySuggestion

object (SummarySuggestion)

Optional. Suggested summary.
agentCoachingSuggestion

object (AgentCoachingSuggestion)

Optional. Suggestion to coach the agent.

FreeFormSuggestion

Suggestion generated using free form generator.

JSON representation 
{
  "response": string
}
Fields
response

string

Required. Free form suggestion.

SummarySuggestion

Suggested summary of the conversation.

JSON representation 
{
  "summarySections": [
    {
      object (SummarySection)
    }
  ]
}
Fields
summarySections[]

object (SummarySection)

Required. All the parts of generated summary.

AgentCoachingSuggestion

Suggestion for coaching agents.

JSON representation 
{
  "applicableInstructions": [
    {
      object (AgentCoachingInstruction)
    }
  ],
  "agentActionSuggestions": [
    {
      object (AgentActionSuggestion)
    }
  ],
  "sampleResponses": [
    {
      object (SampleResponse)
    }
  ]
}
Fields
applicableInstructions[]

object (AgentCoachingInstruction)

Optional. Instructions applicable based on the current context.
agentActionSuggestions[]

object (AgentActionSuggestion)

Optional. Suggested actions for the agent to take.
sampleResponses[]

object (SampleResponse)

Optional. Sample response for the Agent.

AgentActionSuggestion

Actions suggested for the agent. This is based on applicable instructions.

JSON representation 
{
  "agentAction": string,
  "sources": {
    object (Sources)
  },
  "duplicateCheckResult": {
    object (DuplicateCheckResult)
  }
}
Fields
agentAction

string

Optional. The suggested action for the agent.
sources

object (Sources)

Output only. Sources for the agent action suggestion.
duplicateCheckResult

object (DuplicateCheckResult)

Output only. Duplicate check result for the agent action suggestion.

Sources

Sources for the suggestion.

JSON representation 
{
  "instructionIndexes": [
    integer
  ]
}
Fields
instructionIndexes[]

integer

Output only. Source instruction indexes for the suggestion. This is the index of the applicableInstructions field.

DuplicateCheckResult

Duplication check for the suggestion.

JSON representation 
{
  "duplicateSuggestions": [
    {
      object (DuplicateSuggestion)
    }
  ]
}
Fields
duplicateSuggestions[]

object (DuplicateSuggestion)

Output only. The duplicate suggestions.

DuplicateSuggestion

The duplicate suggestion details. Keeping answerRecord and sources together as they are identifiers for duplicate suggestions.

JSON representation 
{
  "answerRecord": string,
  "sources": {
    object (Sources)
  },
  "suggestionIndex": integer,
  "similarityScore": number
}
Fields
answerRecord

string

Output only. The answer record id of the past duplicate suggestion.
sources

object (Sources)

Output only. Sources for the suggestion.
suggestionIndex

integer

Output only. The index of the duplicate suggestion in the past suggestion list.
similarityScore

number

Output only. The similarity score of between the past and current suggestion.

SampleResponse

Sample response that the agent can use. This could be based on applicable instructions and ingested data from other systems.

JSON representation 
{
  "responseText": string,
  "sources": {
    object (Sources)
  },
  "duplicateCheckResult": {
    object (DuplicateCheckResult)
  }
}
Fields
responseText

string

Optional. Sample response for Agent in text.
sources

object (Sources)

Output only. Sources for the Sample Response.
duplicateCheckResult

object (DuplicateCheckResult)

Output only. Duplicate check result for the sample response.

ToolCallInfo

Request and response for a tool call.

JSON representation 
{
  "toolCall": {
    object (ToolCall)
  },
  "toolCallResult": {
    object (ToolCallResult)
  }
}
Fields
toolCall

object (ToolCall)

Required. Request for a tool call.
toolCallResult

object (ToolCallResult)

Required. Response for a tool call.

ToolCall

Represents a call of a specific tool's action with the specified inputs.

JSON representation 
{
  "toolDisplayName": string,
  "toolDisplayDetails": string,
  "action": string,
  "inputParameters": {
    object
  },
  "createTime": string,
  "answerRecord": string,
  "state": enum (State),

  // Union field source can be only one of the following:
  "tool": string
  // End of list of possible types for union field source.
}
Fields
toolDisplayName

string

Optional. A human readable short name of the tool, to be shown on the UI.
toolDisplayDetails

string

Optional. A human readable description of the tool.
action

string

Optional. The name of the tool's action associated with this call.
inputParameters

object (Struct format)

Optional. The action's input parameters.
createTime

string (Timestamp format)

Output only. Create time of the tool call.

Uses RFC 3339, where generated output will always be Z-normalized and use 0, 3, 6 or 9 fractional digits. Offsets other than "Z" are also accepted. Examples: "2014-10-02T15:01:23Z", "2014-10-02T15:01:23.045123456Z" or "2014-10-02T15:01:23+05:30".
answerRecord

string

Optional. The answer record associated with this tool call.
state

enum (State)

Output only. State of the tool call.
Union field source. Specifies the source of this tool call. source can be only one of the following:
tool

string

Optional. The tool associated with this call. Format: projects/<ProjectID>/locations/<LocationID>/tools/<ToolID>.

State

Tool call states.

Enums
STATE_UNSPECIFIED Default value.
TRIGGERED The tool call has been triggered.
NEEDS_CONFIRMATION The tool call requires confirmation from a human.

ToolCallResult

The result of calling a tool's action.

JSON representation 
{
  "action": string,
  "createTime": string,
  "answerRecord": string,

  // Union field source can be only one of the following:
  "tool": string
  // End of list of possible types for union field source.

  // Union field result can be only one of the following:
  "error": {
    object (Error)
  },
  "rawContent": string,
  "content": string
  // End of list of possible types for union field result.
}
Fields
action

string

Optional. The name of the tool's action associated with this call.
createTime

string (Timestamp format)

Output only. Create time of the tool call result.

Uses RFC 3339, where generated output will always be Z-normalized and use 0, 3, 6 or 9 fractional digits. Offsets other than "Z" are also accepted. Examples: "2014-10-02T15:01:23Z", "2014-10-02T15:01:23.045123456Z" or "2014-10-02T15:01:23+05:30".
answerRecord

string

Optional. The answer record associated with this tool call result.
Union field source. Specifies the source of this tool call. source can be only one of the following:
tool

string

Optional. The tool associated with this call. Format: projects/<ProjectID>/locations/<LocationID>/tools/<ToolID>.
Union field result. The tool call's result. result can be only one of the following:
error

object (Error)

The tool call's error.
rawContent

string (bytes format)

Only populated if the response content is not utf-8 encoded. (by definition byte fields are base64 encoded).

A base64-encoded string.
content

string

Only populated if the response content is utf-8 encoded.

Error

An error produced by the tool call.

JSON representation 
{
  "message": string
}
Fields
message

string

Optional. The error message of the function.

InferenceParameter

The parameters of inference.

JSON representation 
{
  "maxOutputTokens": integer,
  "temperature": number,
  "topK": integer,
  "topP": number
}
Fields
maxOutputTokens

integer

Optional. Maximum number of the output tokens for the generator.
temperature

number

Optional. Controls the randomness of LLM predictions. Low temperature = less random. High temperature = more random. If unset (or 0), uses a default value of 0.
topK

integer

Optional. Top-k changes how the model selects tokens for output. A top-k of 1 means the selected token is the most probable among all tokens in the model's vocabulary (also called greedy decoding), while a top-k of 3 means that the next token is selected from among the 3 most probable tokens (using temperature). For each token selection step, the top K tokens with the highest probabilities are sampled. Then tokens are further filtered based on topP with the final token selected using temperature sampling. Specify a lower value for less random responses and a higher value for more random responses. Acceptable value is [1, 40], default to 40.
topP

number

Optional. Top-p changes how the model selects tokens for output. Tokens are selected from most K (see topK parameter) probable to least until the sum of their probabilities equals the top-p value. For example, if tokens A, B, and C have a probability of 0.3, 0.2, and 0.1 and the top-p value is 0.5, then the model will select either A or B as the next token (using temperature) and doesn't consider C. The default top-p value is 0.95. Specify a lower value for less random responses and a higher value for more random responses. Acceptable value is [0.0, 1.0], default to 0.95.

TriggerEvent

The event that triggers the generator and LLM execution.

Enums
TRIGGER_EVENT_UNSPECIFIED Default value for TriggerEvent.
END_OF_UTTERANCE Triggers when each chat message or voice utterance ends.
MANUAL_CALL Triggers on the conversation manually by API calls, such as Conversations.GenerateStatelessSuggestion and Conversations.GenerateSuggestions.
CUSTOMER_MESSAGE Triggers after each customer message only.
AGENT_MESSAGE Triggers after each agent message only.

SuggestionDedupingConfig

Config for suggestion deduping. NEXT_ID: 3

JSON representation 
{
  "enableDeduping": boolean,
  "similarityThreshold": number
}
Fields
enableDeduping

boolean

Optional. Whether to enable suggestion deduping.
similarityThreshold

number

Optional. The threshold for similarity between two suggestions. Acceptable value is [0.0, 1.0], default to 0.8

Methods

create

 Creates a generator.

list

 Lists generators.