REST Resource: projects.answerRecords

Resource: AnswerRecord

Answer records are records to manage answer history and feedbacks for Dialogflow.

Currently, answer record includes:

  • human agent assistant article suggestion
  • human agent assistant faq article

It doesn't include:

  • DetectIntent intent matching
  • DetectIntent knowledge

Answer records are not related to the conversation history in the Dialogflow Console. A Record is generated even when the end-user disables conversation history in the console. Records are created when there's a human agent assistant suggestion generated.

A typical workflow for customers provide feedback to an answer is:

  1. For human agent assistant, customers get suggestion via suggestions.list API. Together with the answers, AnswerRecord.name are returned to the customers.
  2. The customer uses the AnswerRecord.name to call the [answerRecords.patch][] method to send feedback about a specific answer that they believe is wrong.
JSON representation 
{
  "name": string,
  "answerFeedback": {
    object (AnswerFeedback)
  },

  // The following is a list of mutually exclusive fields. At most one of the
  // fields will be set in a response:
  "agentAssistantRecord": {
    object (AgentAssistantRecord)
  }
  // End of mutually exclusive fields.
}
Fields
name

string

The unique identifier of this answer record. Required for AnswerRecords.UpdateAnswerRecord method. Format: projects/<Project ID>/locations/<Location ID>/answerRecords/<Answer Record ID>.
answerFeedback

object (AnswerFeedback)

Optional. The AnswerFeedback for this record. You can set this with AnswerRecords.UpdateAnswerRecord in order to give us feedback about this answer.
Output only. The record for this answer. The following is a list of mutually exclusive fields. At most one of the fields will be set in a response:
agentAssistantRecord

object (AgentAssistantRecord)

Output only. The record for human agent assistant.
End of mutually exclusive fields.

AnswerFeedback

Represents feedback the customer has about the quality & correctness of a certain answer in a conversation.

JSON representation 
{
  "correctnessLevel": enum (CorrectnessLevel),
  "clicked": boolean,
  "clickTime": string,
  "displayed": boolean,
  "displayTime": string,

  // The following is a list of mutually exclusive fields. At most one of the
  // fields will be set in a response:
  "agentAssistantDetailFeedback": {
    object (AgentAssistantFeedback)
  }
  // End of mutually exclusive fields.
}
Fields
correctnessLevel

enum (CorrectnessLevel)

The correctness level of the specific answer.
clicked

boolean

Indicates whether the answer/item was clicked by the human agent or not. Default to false. For knowledge search, the answer record is considered to be clicked if the answer was copied or any URI was clicked.
clickTime

string (Timestamp format)

Time when the answer/item was clicked.

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".
displayed

boolean

Indicates whether the answer/item was displayed to the human agent in the agent desktop UI. Default to false.
displayTime

string (Timestamp format)

Time when the answer/item was displayed.

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".
Normally, detail feedback is provided when answer is not fully correct. The following is a list of mutually exclusive fields. At most one of the fields will be set in a response:
agentAssistantDetailFeedback

object (AgentAssistantFeedback)

Optional. Detail feedback of agent assistant suggestions.
End of mutually exclusive fields.

CorrectnessLevel

The correctness level of an answer.

Enums
CORRECTNESS_LEVEL_UNSPECIFIED Correctness level unspecified.
NOT_CORRECT Answer is totally wrong.
PARTIALLY_CORRECT Answer is partially correct.
FULLY_CORRECT Answer is fully correct.

AgentAssistantFeedback

Detail feedback of Agent Assistant result.

JSON representation 
{
  "answerRelevance": enum (AnswerRelevance),
  "documentCorrectness": enum (DocumentCorrectness),
  "documentEfficiency": enum (DocumentEfficiency),
  "summarizationFeedback": {
    object (SummarizationFeedback)
  },
  "knowledgeSearchFeedback": {
    object (KnowledgeSearchFeedback)
  },
  "knowledgeAssistFeedback": {
    object (KnowledgeAssistFeedback)
  },
  "companionFeedback": {
    object (CompanionFeedback)
  }
}
Fields
answerRelevance

enum (AnswerRelevance)

Optional. Whether or not the suggested answer is relevant.

For example:
documentCorrectness

enum (DocumentCorrectness)

Optional. Whether or not the information in the document is correct.

For example:

  • Query: "Can I return the package in 2 days once received?"
  • Suggested document says: "Items must be returned/exchanged within 60 days of the purchase date."
  • Ground truth: "No return or exchange is allowed."
  • documentCorrectness: INCORRECT
documentEfficiency

enum (DocumentEfficiency)

Optional. Whether or not the suggested document is efficient. For example, if the document is poorly written, hard to understand, hard to use or too long to find useful information, documentEfficiency is DocumentEfficiency.INEFFICIENT.
summarizationFeedback

object (SummarizationFeedback)

Feedback for conversation summarization.
knowledgeSearchFeedback

object (KnowledgeSearchFeedback)

Optional. Feedback for knowledge search.
knowledgeAssistFeedback

object (KnowledgeAssistFeedback)

Optional. Feedback for knowledge assist.
companionFeedback

object (CompanionFeedback)

Optional. Feedback for companion agent.

AnswerRelevance

Relevance of an answer.

Enums
ANSWER_RELEVANCE_UNSPECIFIED Answer relevance unspecified.
IRRELEVANT Answer is irrelevant to query.
RELEVANT Answer is relevant to query.

DocumentCorrectness

Correctness of document.

Enums
DOCUMENT_CORRECTNESS_UNSPECIFIED Document correctness unspecified.
INCORRECT Information in document is incorrect.
CORRECT Information in document is correct.

DocumentEfficiency

Efficiency of document.

Enums
DOCUMENT_EFFICIENCY_UNSPECIFIED Document efficiency unspecified.
INEFFICIENT Document is inefficient.
EFFICIENT Document is efficient.

SummarizationFeedback

Feedback for conversation summarization.

JSON representation 
{
  "startTimestamp": string,
  "submitTimestamp": string,
  "summaryText": string,
  "textSections": {
    string: string,
    ...
  }
}
Fields
startTimestamp

string (Timestamp format)

Timestamp when composing of the summary starts.

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".
submitTimestamp

string (Timestamp format)

Timestamp when the summary was submitted.

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".
summaryText

string

Text of actual submitted summary.
textSections

map (key: string, value: string)

Optional. Actual text sections of submitted summary.

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

KnowledgeSearchFeedback

Feedback for knowledge search.

JSON representation 
{
  "answerCopied": boolean,
  "clickedUris": [
    string
  ]
}
Fields
answerCopied

boolean

Whether the answer was copied by the human agent or not. If the value is set to be true, AnswerFeedback.clicked will be updated to be true.
clickedUris[]

string

The URIs clicked by the human agent. The value is appended for each UpdateAnswerRecordRequest. If the value is not empty, AnswerFeedback.clicked will be updated to be true.

KnowledgeAssistFeedback

Feedback for knowledge assist.

JSON representation 
{
  "answerCopied": boolean,
  "clickedUris": [
    string
  ]
}
Fields
answerCopied

boolean

Whether the suggested answer was copied by the human agent. If the value is set to be true, AnswerFeedback.clicked will be updated to be true.
clickedUris[]

string

The URIs clicked by the human agent. The value is appended for each UpdateAnswerRecordRequest. If the value is not empty, AnswerFeedback.clicked will be updated to be true.

CompanionFeedback

This type has no fields.

Feedback for companion agent.

AgentAssistantRecord

Represents a record of a human agent assistant answer.

JSON representation 
{

  // The following is a list of mutually exclusive fields. At most one of the
  // fields will be set in a response:
  "articleSuggestionAnswer": {
    object (ArticleAnswer)
  },
  "faqAnswer": {
    object (FaqAnswer)
  },
  "dialogflowAssistAnswer": {
    object (DialogflowAssistAnswer)
  },
  "generatorSuggestion": {
    object (GeneratorSuggestion)
  },
  "companionSuggestion": {
    object (CompanionSuggestion)
  },
  "reactiveCompanionSuggestion": {
    object (ReactiveModeResponse)
  }
  // End of mutually exclusive fields.
}
Fields
Output only. The agent assistant answer. The following is a list of mutually exclusive fields. At most one of the fields will be set in a response:
articleSuggestionAnswer

object (ArticleAnswer)

Output only. The article suggestion answer.
faqAnswer

object (FaqAnswer)

Output only. The FAQ answer.
dialogflowAssistAnswer

object (DialogflowAssistAnswer)

Output only. The Dialogflow assist answer.
generatorSuggestion

object (GeneratorSuggestion)

Output only. The generator suggestion.
companionSuggestion

object (CompanionSuggestion)

Output only. The companion suggestion.
reactiveCompanionSuggestion

object (ReactiveModeResponse)

Output only. The reactive companion suggestion.
End of mutually exclusive fields.

ArticleAnswer

Represents article answer.

JSON representation 
{
  "title": string,
  "uri": string,
  "snippets": [
    string
  ],
  "metadata": {
    string: string,
    ...
  },
  "answerRecord": string
}
Fields
title

string

The article title.
uri

string

The article URI.
snippets[]

string

Output only. Article snippets.
metadata

map (key: string, value: string)

A map that contains metadata about the answer and the document from which it originates.

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

string

The name of answer record, in the format of "projects//locations//answerRecords/"

FaqAnswer

Represents answer from "frequently asked questions".

JSON representation 
{
  "answer": string,
  "confidence": number,
  "question": string,
  "source": string,
  "metadata": {
    string: string,
    ...
  },
  "answerRecord": string
}
Fields
answer

string

The piece of text from the source knowledge base document.
confidence

number

The system's confidence score that this Knowledge answer is a good match for this conversational query, range from 0.0 (completely uncertain) to 1.0 (completely certain).
question

string

The corresponding FAQ question.
source

string

Indicates which Knowledge Document this answer was extracted from. Format: projects/<Project ID>/locations/<Location ID>/agent/knowledgeBases/<Knowledge Base ID>/documents/<Document ID>.
metadata

map (key: string, value: string)

A map that contains metadata about the answer and the document from which it originates.

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

string

The name of answer record, in the format of "projects//locations//answerRecords/"

DialogflowAssistAnswer

Represents a Dialogflow assist answer.

JSON representation 
{
  "answerRecord": string,

  // The following is a list of mutually exclusive fields. At most one of the
  // fields will be set in a response:
  "queryResult": {
    object (QueryResult)
  },
  "intentSuggestion": {
    object (IntentSuggestion)
  }
  // End of mutually exclusive fields.
}
Fields
answerRecord

string

The name of answer record, in the format of "projects//locations//answerRecords/"
Result from DetectIntent for one matched intent. The following is a list of mutually exclusive fields. At most one of the fields will be set in a response:
queryResult

object (QueryResult)

Result from v2 agent.
intentSuggestion

object (IntentSuggestion)

An intent suggestion generated from conversation.
End of mutually exclusive fields.

QueryResult

Represents the result of conversational query or event processing.

JSON representation 
{
  "queryText": string,
  "languageCode": string,
  "speechRecognitionConfidence": number,
  "action": string,
  "parameters": {
    object
  },
  "allRequiredParamsPresent": boolean,
  "cancelsSlotFilling": boolean,
  "fulfillmentText": string,
  "fulfillmentMessages": [
    {
      object (Message)
    }
  ],
  "webhookSource": string,
  "webhookPayload": {
    object
  },
  "outputContexts": [
    {
      object (Context)
    }
  ],
  "intent": {
    object (Intent)
  },
  "intentDetectionConfidence": number,
  "diagnosticInfo": {
    object
  },
  "sentimentAnalysisResult": {
    object (SentimentAnalysisResult)
  },
  "knowledgeAnswers": {
    object (KnowledgeAnswers)
  }
}
Fields
queryText

string

The original conversational query text:

  • If natural language text was provided as input, queryText contains a copy of the input.
  • If natural language speech audio was provided as input, queryText contains the speech recognition result. If speech recognizer produced multiple alternatives, a particular one is picked.
  • If automatic spell correction is enabled, queryText will contain the corrected user input.
languageCode

string

The language that was triggered during intent detection. See Language Support for a list of the currently supported language codes.
speechRecognitionConfidence

number

The Speech recognition confidence between 0.0 and 1.0. A higher number indicates an estimated greater likelihood that the recognized words are correct. The default of 0.0 is a sentinel value indicating that confidence was not set.

This field is not guaranteed to be accurate or set. In particular this field isn't set for StreamingDetectIntent since the streaming endpoint has separate confidence estimates per portion of the audio in StreamingRecognitionResult.
action

string

The action name from the matched intent.
parameters

object (Struct format)

The collection of extracted parameters.

Depending on your protocol or client library language, this is a map, associative array, symbol table, dictionary, or JSON object composed of a collection of (MapKey, MapValue) pairs:

  • MapKey type: string
  • MapKey value: parameter name
  • MapValue type: If parameter's entity type is a composite entity then use map, otherwise, depending on the parameter value type, it could be one of string, number, boolean, null, list or map.
  • MapValue value: If parameter's entity type is a composite entity then use map from composite entity property names to property values, otherwise, use parameter value.
allRequiredParamsPresent

boolean

This field is set to:

  • false if the matched intent has required parameters and not all of the required parameter values have been collected.
  • true if all required parameter values have been collected, or if the matched intent doesn't contain any required parameters.
cancelsSlotFilling

boolean

Indicates whether the conversational query triggers a cancellation for slot filling. For more information, see the cancel slot filling documentation.
fulfillmentText

string

The text to be pronounced to the user or shown on the screen. Note: This is a legacy field, fulfillmentMessages should be preferred.
fulfillmentMessages[]

object (Message)

The collection of rich messages to present to the user.
webhookSource

string

If the query was fulfilled by a webhook call, this field is set to the value of the source field returned in the webhook response.
webhookPayload

object (Struct format)

If the query was fulfilled by a webhook call, this field is set to the value of the payload field returned in the webhook response.
outputContexts[]

object (Context)

The collection of output contexts. If applicable, outputContexts.parameters contains entries with name <parameter name>.original containing the original parameter values before the query.
intent

object (Intent)

The intent that matched the conversational query. Some, not all fields are filled in this message, including but not limited to: name, displayName, endInteraction and isFallback.
intentDetectionConfidence

number

The intent detection confidence. Values range from 0.0 (completely uncertain) to 1.0 (completely certain). This value is for informational purpose only and is only used to help match the best intent within the classification threshold. This value may change for the same end-user expression at any time due to a model retraining or change in implementation. If there are multiple knowledgeAnswers messages, this value is set to the greatest knowledgeAnswers.match_confidence value in the list.
diagnosticInfo

object (Struct format)

Free-form diagnostic information for the associated detect intent request. The fields of this data can change without notice, so you should not write code that depends on its structure. The data may contain:

  • webhook call latency
  • webhook errors
sentimentAnalysisResult

object (SentimentAnalysisResult)

The sentiment analysis result, which depends on the sentimentAnalysisRequestConfig specified in the request.
knowledgeAnswers

object (KnowledgeAnswers)

The result from Knowledge Connector (if any), ordered by decreasing KnowledgeAnswers.match_confidence.

SentimentAnalysisResult

The result of sentiment analysis. Sentiment analysis inspects user input and identifies the prevailing subjective opinion, especially to determine a user's attitude as positive, negative, or neutral. For [Participants.DetectIntent][], it needs to be configured in DetectIntentRequest.query_params. For [Participants.StreamingDetectIntent][], it needs to be configured in StreamingDetectIntentRequest.query_params. And for Participants.AnalyzeContent and Participants.StreamingAnalyzeContent, it needs to be configured in ConversationProfile.human_agent_assistant_config

JSON representation 
{
  "queryTextSentiment": {
    object (Sentiment)
  }
}
Fields
queryTextSentiment

object (Sentiment)

The sentiment analysis result for queryText.

Sentiment

The sentiment, such as positive/negative feeling or association, for a unit of analysis, such as the query text. See: https://cloud.google.com/natural-language/docs/basics#interpreting_sentiment_analysis_values for how to interpret the result.

JSON representation 
{
  "score": number,
  "magnitude": number
}
Fields
score

number

Sentiment score between -1.0 (negative sentiment) and 1.0 (positive sentiment).
magnitude

number

A non-negative number in the [0, +inf) range, which represents the absolute magnitude of sentiment, regardless of score (positive or negative).

KnowledgeAnswers

Represents the result of querying a Knowledge base.

JSON representation 
{
  "answers": [
    {
      object (Answer)
    }
  ]
}
Fields
answers[]

object (Answer)

A list of answers from Knowledge Connector.

Answer

An answer from Knowledge Connector.

JSON representation 
{
  "source": string,
  "faqQuestion": string,
  "answer": string,
  "matchConfidenceLevel": enum (MatchConfidenceLevel),
  "matchConfidence": number
}
Fields
source

string

Indicates which Knowledge Document this answer was extracted from. Format: projects/<Project ID>/knowledgeBases/<Knowledge Base ID>/documents/<Document ID>.
faqQuestion

string

The corresponding FAQ question if the answer was extracted from a FAQ Document, empty otherwise.
answer

string

The piece of text from the source knowledge base document that answers this conversational query.
matchConfidenceLevel

enum (MatchConfidenceLevel)

The system's confidence level that this knowledge answer is a good match for this conversational query. NOTE: The confidence level for a given <query, answer> pair may change without notice, as it depends on models that are constantly being improved. However, it will change less frequently than the confidence score below, and should be preferred for referencing the quality of an answer.
matchConfidence

number

The system's confidence score that this Knowledge answer is a good match for this conversational query. The range is from 0.0 (completely uncertain) to 1.0 (completely certain). Note: The confidence score is likely to vary somewhat (possibly even for identical requests), as the underlying model is under constant improvement. It may be deprecated in the future. We recommend using matchConfidenceLevel which should be generally more stable.

MatchConfidenceLevel

Represents the system's confidence that this knowledge answer is a good match for this conversational query.

Enums
MATCH_CONFIDENCE_LEVEL_UNSPECIFIED Not specified.
LOW Indicates that the confidence is low.
MEDIUM Indicates our confidence is medium.
HIGH Indicates our confidence is high.

IntentSuggestion

Represents an intent suggestion.

JSON representation 
{
  "displayName": string,
  "description": string,

  // The following is a list of mutually exclusive fields. At most one of the
  // fields will be set in a response:
  "intentV2": string
  // End of mutually exclusive fields.
}
Fields
displayName

string

The display name of the intent.
description

string

Human readable description for better understanding an intent like its scope, content, result etc. Maximum character limit: 140 characters.
The name of the intent. The following is a list of mutually exclusive fields. At most one of the fields will be set in a response:
intentV2

string

The unique identifier of this intent. Format: projects/<Project ID>/locations/<Location ID>/agent/intents/<Intent ID>.
End of mutually exclusive fields.

CompanionSuggestion

Represents a companion suggestion answer.

JSON representation 
{
  "guidances": [
    {
      object (Guidance)
    }
  ]
}
Fields
guidances[]

object (Guidance)

List of guidances generated by the Companion Agent.

Guidance

Represents guidance for companion suggestion.

JSON representation 
{
  "suggestedReply": string,
  "suggestedAction": string,
  "instructionSource": {
    object (GuidanceInstruction)
  },
  "knowledgeSources": [
    {
      object (KnowledgeSource)
    }
  ],
  "explanation": string,
  "toolCalls": [
    {
      object (ToolCallSuggestion)
    }
  ],
  "triggeringToolCallAnswerRecords": [
    string
  ],
  "groundingMetadata": {
    object (GroundingMetadata)
  }
}
Fields
suggestedReply

string

Suggested reply for the agent to respond to the customer.
suggestedAction

string

Suggested action to human agent. For example, "Ask the customer for their shipping address" or "Create a new case for the customer".
instructionSource

object (GuidanceInstruction)

Guidance instruction source.
knowledgeSources[]

object (KnowledgeSource)

Knowledge sources.
explanation

string

Optional. A brief explanation of why the action is suggested.
toolCalls[]

object (ToolCallSuggestion)

Optional. Rich structured tool calls and their associated citations delivered directly as suggestions.
triggeringToolCallAnswerRecords[]

string

Optional. The AnswerRecords of the tool calls that triggered this guidance. Format: projects/<Project ID>/locations/<Location ID>/answerRecords/<Answer Record ID>.
groundingMetadata

object (GroundingMetadata)

Optional. Grounding metadata for the generated guidance.

KnowledgeSource

Represents a knowledge source for the guidance.

JSON representation 
{
  "knowledgeArticleUrl": string,
  "knowledgeArticleTitle": string,
  "knowledgeSnippet": string
}
Fields
knowledgeArticleUrl

string

URL of the knowledge article.
knowledgeArticleTitle

string

Knowledge article title.
knowledgeSnippet

string

Knowledge snippet.

GroundingMetadata

Grounding metadata contains sources, citations, and search entry points used to ground a generated answer or suggestion.

JSON representation 
{
  "webSearchQueries": [
    string
  ],
  "searchEntryPoint": {
    object (SearchEntryPoint)
  },
  "groundingChunks": [
    {
      object (GroundingChunk)
    }
  ],
  "groundingSupports": [
    {
      object (GroundingSupport)
    }
  ]
}
Fields
webSearchQueries[]

string

Optional. The web search queries that were used to generate the content.
searchEntryPoint

object (SearchEntryPoint)

Optional. A web search entry point that can be used to display search results.
groundingChunks[]

object (GroundingChunk)

Optional. A list of supporting references retrieved from the grounding source.
groundingSupports[]

object (GroundingSupport)

Optional. A list of grounding supports that connect the generated content to the grounding chunks.

SearchEntryPoint

Google search entry point.

JSON representation 
{
  "renderedContent": string
}
Fields
renderedContent

string

Optional. An HTML snippet that can be embedded in a web page or an application's webview. This snippet displays a search result, including the title, URL, and a brief description of the search result.

GroundingChunk

A piece of evidence that supports a claim made by the model.

This is used to show a citation for a claim made by the model. It contains a reference to the source of the information.

JSON representation 
{

  // The following is a list of mutually exclusive fields. At most one of the
  // fields will be set in a response:
  "web": {
    object (Web)
  },
  "retrievedContext": {
    object (RetrievedContext)
  }
  // End of mutually exclusive fields.
}
Fields
The source of the grounding chunk. Currently we only support web search. The following is a list of mutually exclusive fields. At most one of the fields will be set in a response:
web

object (Web)

Optional. A grounding chunk from a web page, typically from web search. See the Web message for details.
retrievedContext

object (RetrievedContext)

Optional. A grounding chunk from a data source retrieved by a data store tool.
End of mutually exclusive fields.

Web

A Web chunk is a piece of evidence that comes from a web page. It contains the URI of the web page, the title of the page, and the domain of the page. This is used to provide the user with a link to the source of the information.

JSON representation 
{
  "uri": string,
  "title": string,
  "domain": string
}
Fields
uri

string

Output only. The URI of the web page that contains the evidence.
title

string

Output only. The title of the web page that contains the evidence.
domain

string

Output only. The domain of the web page that contains the evidence. This can be used to filter out low-quality sources.

RetrievedContext

Context retrieved from a data source to ground the model's response. This is used when a retrieval tool fetches information from a user-provided corpus or a public dataset.

JSON representation 
{
  "uri": string,
  "title": string,
  "text": string
}
Fields
uri

string

Output only. The URI of the retrieved data source.
title

string

Output only. The title of the retrieved data source.
text

string

Output only. The content of the retrieved data source.

GroundingSupport

A collection of supporting references for a segment or part of the model's response.

JSON representation 
{
  "segment": {
    object (Segment)
  },
  "groundingChunkIndices": [
    integer
  ]
}
Fields
segment

object (Segment)

Optional. Segment of the content this support belongs to.
groundingChunkIndices[]

integer

Optional. A list of indices into groundingChunks field specifying the citations associated with the claim. For instance [1, 3] means that groundingChunks[1] and groundingChunks[3] are the retrieved contents attributed to the claim.

Segment

A segment of the content.

JSON representation 
{
  "startIndex": integer,
  "endIndex": integer,
  "text": string
}
Fields
startIndex

integer

Output only. The start index of the segment, measured in bytes. This marks the beginning of the segment and is inclusive, meaning the byte at this index is the first byte of the segment.
endIndex

integer

Output only. The end index of the segment, measured in bytes. This marks the end of the segment and is exclusive, meaning the segment includes content up to, but not including, the byte at this index.
text

string

Output only. The text of the segment.

ToolCallSuggestion

Structured wrapper that pairs tool execution details with strongly-typed citations.

JSON representation 
{
  "toolCallInfo": {
    object (ToolCallInfo)
  },
  "textUpdate": string
}
Fields
toolCallInfo

object (ToolCallInfo)

Optional. Tool execution request/response details.
textUpdate

string

Optional. The conversational text update generated by the agent accompanying this tool call (e.g. "The status of your order 12345 is currently being retrieved..."). This is a status update emitted alongside in-flight tool execution and is not accumulated into the final response.

ReactiveModeResponse

The complete response generated by the reactive mode capturing the entire accumulated streaming response, grounding metadata, and tool calls.

JSON representation 
{
  "response": string,
  "groundingMetadata": {
    object (GroundingMetadata)
  },
  "toolCalls": [
    {
      object (ToolCallSuggestion)
    }
  ]
}
Fields
response

string

Optional. The full reactive suggestion generated by the Companion Agent. After the last streaming chunk is sent, this is the final response concatenating all the streaming response chunks together to represent the full reactive suggestion.
groundingMetadata

object (GroundingMetadata)

Optional. Grounding metadata for the generated response.
toolCalls[]

object (ToolCallSuggestion)

Optional. Captures all the rich structured tool calls and their associated citations used by the reactive agent to generate the reactive suggestion.

Methods

get
(deprecated)

 Deprecated.

list

 Returns the list of all answer records in the specified project in reverse chronological order.

patch

 Updates the specified answer record.