Chat Data

App · 24 actions

Create and train Chat Data chatbots, send messages, and manage conversations, leads, actions and analytics.

Actions

ActionWhat it does
Add a webhookAdd a web address that receives the chatbot's events as they happen: chats, live chat escalations, lead submissions, and users coming online or going offline.
Add an actionAdd an action to a chatbot.
Attach a lead to a conversationAdd or update the customer details (lead) linked to a conversation.
Create and train a chatbotCreate a chatbot using one of Chat Data's medical chat models, your own uploaded data or a custom model.
Delete a chatbotDelete a chatbot and all of its data, including training data, conversations, leads and user monitoring data.
Delete a conversationDelete one of a chatbot's conversations.
Delete a webhookDelete the webhook with the given address and all its event subscriptions.
Delete an actionDelete a specific action from the chatbot by either name or index.
Get a chatbotGet the details of one chatbot, in the same form as listing chatbots gives.
Get a conversationGet a conversation's details, including its messages, metadata and the customer's details when available.
Get all chatbotsList the chatbots your API key can reach: every chatbot in the workspace for a workspace key, or only its own chatbot for a chatbot key.
Get an actionGet a specific action by name or index for the specified chatbot.
Get analyticsGet a chatbot's overall statistics.
Get chatbot training statusGet a chatbot's current training progress.
Get confidence scoresGet confidence scores for the passages a chatbot would use to answer, as an estimate of how well it can answer before you send the message.
Get conversationsList a chatbot's conversations that have at least one user message, most recently active first.
Get customersList the customers (leads) a chatbot has collected, newest first.
Get the conversation portal linkGenerate a shareable portal link for a specific conversation.
Get the current planGet a summary of the workspace's current subscriptions and privileges.
List actionsGet all configured actions for the specified chatbot.
Query the knowledge baseGet the passages from a chatbot's knowledge base that are relevant to a query, for example to give a custom GPT.
Retrain an existing chatbotRetrain a chatbot built from your own uploaded data, adding new information to its knowledge base or removing existing data.
Send a messageSend messages to a chatbot and get its reply.
Update chatbot settingsChange a chatbot's settings.

Add a webhook

Add a web address that receives the chatbot's events as they happen: chats, live chat escalations, lead submissions, and users coming online or going offline. Your plan must include webhooks, and a chatbot can have up to 5. Adding an event to an existing address subscribes it to that event too.

Inputs

Chatbotchatbot_idtextRequired
The ID of the chatbot the webhook belongs to.
URLurltextRequired
The URL endpoint that the chatbot delivers events to.
EventeventtextRequired
The type of event the webhook subscribes to. See [Webhooks](/api-reference/webhooks) for the chat, live-chat-escalation and lead-submission bodies. user-online and user-offline fire when the user joins or leaves a live chat session and carry chatbot_id, chatbot_name, conversation_id and, when known, the user's name, email, phone, source and user_uuid. If a webhook with this url already exists, the event is added to its subscriptions.One of: chat, live-chat-escalation, lead-submission, user-online, user-offline.
Content typecontent_typetext
The Content-Type of the deliveries. Endpoints on hook.make.com, hook.*.make.com, n8n.cloud and *.n8n.cloud always receive text/plain. Re-adding an existing url sets its content type to this value (text/plain when omitted).One of: text/plain, application/json.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "status": "success",
    "webhooks": []
  }
}

Add an action

Add an action to a chatbot. An API action is tested before it is saved: Chat Data calls its address with each parameter's example value and rejects the action if that call fails.

Inputs

Chatbotchatbot_idtextRequired
The ID of the chatbot to which the action will be added.
API actionapi_actionobjectRequired
The action to add.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "action": {
      "bodyParams": [],
      "description": "example",
      "enableDateRestriction": false,
      "enabled": false,
      "headers": [],
      "includeImages": false,
      "includedDomains": [],
      "index": 1,
      "label": "example",
      "link": "example",
      "method": "GET",
      "name": "example",
      "pathParams": [],
      "prioritizeRecentResults": false,
      "queryParams": [],
      "return_file": false,
      "type": "example",
      "url": "example"
    },
    "status": "example"
  }
}

Attach a lead to a conversation

Add or update the customer details (lead) linked to a conversation.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier of the chatbot.
Conversation IDconversation_idtextRequired
The unique identifier of the conversation.
Lead IDlead_idtext
Lead UUID. Can be used alone to find an existing lead, but is not sufficient for creating a new lead. Important: Once a lead exists, its uuid can never be changed. Attempting to change it returns a 400 error. When creating a new lead, this value becomes its uuid; otherwise a new UUID is generated.
Namenametext
Lead name. Can be used to find an existing lead with an exact, case-sensitive string match. If several leads share the same name, one of them is used, and which one is not guaranteed. Use unique identifiers for deterministic matching.
Emailemailtext
Lead email address. Must be a valid, deliverable email address of at most 254 characters (RFC 5321).
Phonephonetext
Lead phone number. It must be a valid phone number in international format, starting with +.
Psidpsidtext
Page-scoped ID (typically for Facebook Messenger).
Ig usernameig_usernametext
Instagram username.
Usernameusernametext
Slack or Telegram username. A platform-specific identifier, distinct from ig_username (Instagram) and psid (Messenger).
Infoinfotext
Additional information about the lead, up to 1,000,000 characters. It may be truncated to 1,000 tokens if "User Info Truncation" is turned on in Settings > Limits Control.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "lead": {
      "email": "example",
      "ig_username": "example",
      "info": "example",
      "leadId": "example",
      "name": "example",
      "phone": "example",
      "psid": "example",
      "source": "example",
      "username": "example"
    },
    "status": "success"
  }
}

Create and train a chatbot

Create a chatbot using one of Chat Data's medical chat models, your own uploaded data or a custom model. The model is required.

Inputs

Chatbot namechatbot_nametext
The name of the chatbot to be created. Defaults to an empty name.
Source textsource_texttext
Text data for the chatbot, subject to character limits based on your plan. Relevant only if the model is custom-data-upload.
URLS to scrapeurls_to_scrapelist
A list of URLs from which Chat Data extracts text content. Because the amount of text on a website cannot be known in advance, crawling stops when either the list of URLs is exhausted or the chatbot's character limit is reached. URLs must start with http:// or https://, for example https://www.chat-data.com. Relevant only if the model is custom-data-upload.
Productsproductslist
A list of products for training the chatbot.
Q and asq_and_aslist
A list of Q&As for the chatbot to learn specific topic answers. Items without both a question and an answer are skipped.
Custom backendcustom_backendtext
The URL of your own backend, starting with http:// or https://. Required when model is custom-model; ignored for other models.
Bearerbearertext
The bearer token sent to your custom backend. Only used when model is custom-model.
ModelmodeltextRequired
The model the chatbot uses: - For training with your own data: custom-data-upload - For human medical questions: medical-chat-human - For veterinary medical questions: medical-chat-vet - For a custom model: custom-modelOne of: custom-data-upload, medical-chat-human, medical-chat-vet, custom-model.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "chatbotId": "example",
    "consumedCharacters": 1,
    "dataSources": [],
    "remainingCharacters": 1,
    "status": "success"
  }
}

Delete a chatbot

Delete a chatbot and all of its data, including training data, conversations, leads and user monitoring data. This can't be undone.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier allocated to the chatbot.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "status": "success"
  }
}

Delete a conversation

Delete one of a chatbot's conversations.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier of the chatbot.
Conversation IDconversation_idtextRequired
The unique identifier of the conversation to be deleted.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "message": "example",
    "status": "success"
  }
}

Delete a webhook

Delete the webhook with the given address and all its event subscriptions. Returns the chatbot's remaining webhooks, and changes nothing if no webhook has that address.

Inputs

Chatbotchatbot_idtextRequired
The ID of the chatbot the webhook belongs to.
URLurltextRequired
The URL endpoint that the chatbot delivers events to.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "status": "success",
    "webhooks": []
  }
}

Delete an action

Delete a specific action from the chatbot by either name or index. Send exactly one of name or index.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier of the chatbot.
Namenametext
The name of the action to delete (mutually exclusive with index).
Indexindexnumber
The index of the action to delete (mutually exclusive with name).At least 0.0.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "deletedAction": {
      "bodyParams": [],
      "description": "example",
      "enableDateRestriction": false,
      "enabled": false,
      "headers": [],
      "includeImages": false,
      "includedDomains": [],
      "index": 1,
      "label": "example",
      "link": "example",
      "method": "GET",
      "name": "example",
      "pathParams": [],
      "prioritizeRecentResults": false,
      "queryParams": [],
      "return_file": false,
      "type": "example",
      "url": "example"
    },
    "status": "success"
  }
}

Get a chatbot

Get the details of one chatbot, in the same form as listing chatbots gives.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier allocated to the chatbot.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "chatbot": {
      "baseModel": "gpt-4o-mini",
      "basePrompt": "example",
      "bearer": "example",
      "bubbleSize": 1,
      "calCom": {
        "brandColor": "example",
        "calendarMessage": "example",
        "hideEventTypeDetails": false,
        "layout": "example",
        "showCalendar": false,
        "username": "example"
      },
      "calendar": {
        "calendarMessage": "example",
        "calendarUrl": "example",
        "showCalendar": false
      },
      "chatbotId": "example",
      "chatbotName": "example",
      "collectedCustomerInformation": [],
      "createdAt": 1,
      "customBackend": "example",
      "dataSources": [],
      "disallowedUrls": [],
      "domains": [],
      "enableBouncingAnimation": false,
      "enableBubbleSound": false,
      "ignoreDataSource": false,
      "imageDescriptionPrompt": "example",
      "initialMessages": [],
      "inputPlaceholderMsg": "example",
      "lastMessageAt": 1,
      "lastTrainedAt": 1,
      "leadsConfig": {
        "collectEmail": false,
        "collectName": false,
        "collectPhone": false,
        "emailMessage": "example",
        "message": "example",
        "nameMessage": "example",
        "phoneMessage": "example"
      },
      "model": "custom-data-upload",
      "notification": {
        "conversationEmails": [],
        "enableConversationEmails": false,
        "enableLeadsEmail": false,
        "leadsEmails": []
      },
      "numberOfConversations": 1,
      "onlyAllowOnAddedDomains": false,
      "padding": {
        "x": 1,
        "y": 1
      },
      "rateLimitMessage": "example",
      "rateLimitPoints": 1,
      "rateLimitTimeframe": 1,
      "status": 0,
      "storeChat": false,
      "styles": {
        "actionButtonColor": "example",
        "alignChatButton": "left",
        "autoOpenChatWindowAfter": 1,
        "bubblePictureFile": "example",
        "buttonColor": "example",
        "displayName": "example",
        "profilePictureFile": "example",
        "theme": "light",
        "userMessageColor": "example"
      },
      "suggestedMessages": [],
      "team": [],
      "temperature": 0,
      "toolsParams": [],
      "trackIpAddress": false,
      "visibility": "private",
      "webhooks": []
    },
    "status": "success"
  }
}

Get a conversation

Get a conversation's details, including its messages, metadata and the customer's details when available.

Inputs

Conversation IDconversation_idtextRequired
The unique identifier of the conversation.
Chatbotchatbot_idtext
The unique identifier of the chatbot to retrieve the conversation from. If not provided, the conversation is retrieved from any chatbot in your workspace.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "conversation": {
      "browserAgent": "example",
      "chatbotId": "example",
      "conversationId": "example",
      "createdAt": 1,
      "customer": {
        "createdAt": 1,
        "email": "example",
        "ig_username": "example",
        "info": "example",
        "ipAddress": "example",
        "leadId": "example",
        "name": "example",
        "phone": "example",
        "psid": "example",
        "source": "widget",
        "userCountry": "example",
        "username": "example"
      },
      "ipAddress": "example",
      "lastMessageAt": "2026-10-24T12:00:00Z",
      "liveChatEnded": false,
      "messages": [],
      "overalRating": 1,
      "requestedLiveChat": false,
      "source": "widget",
      "urlPath": "example",
      "userCountry": "example"
    },
    "status": "success"
  }
}

Get all chatbots

List the chatbots your API key can reach: every chatbot in the workspace for a workspace key, or only its own chatbot for a chatbot key.

Inputs

None.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "chatbots": [],
    "status": "success"
  }
}

Get an action

Get a specific action by name or index for the specified chatbot. Send exactly one of name or index.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier allocated to the chatbot.
Namenametext
The name of the action to retrieve (mutually exclusive with index).
Indexindexnumber
The index of the action to retrieve (mutually exclusive with name).At least 0.0.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "action": {
      "bodyParams": [],
      "description": "example",
      "enableDateRestriction": false,
      "enabled": false,
      "headers": [],
      "includeImages": false,
      "includedDomains": [],
      "index": 1,
      "label": "example",
      "link": "example",
      "method": "GET",
      "name": "example",
      "pathParams": [],
      "prioritizeRecentResults": false,
      "queryParams": [],
      "return_file": false,
      "type": "example",
      "url": "example"
    },
    "status": "success"
  }
}

Get analytics

Get a chatbot's overall statistics.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier allocated to the chatbot.
Start timestampstart_timestampdate and time
Start of the time range (inclusive), in milliseconds since the Unix epoch. Statistics are stored per UTC day, each dated at 00:00 UTC.
End timestampend_timestampdate and time
End of the time range (inclusive), in milliseconds since the Unix epoch.
Sourcesourcelist
Filter analytics by source. Repeat the parameter to match several sources.One or more of: widget, iframe, site, api, Whatsapp, WhatsappWebAPI, Messenger, Instagram, Telegram, Line, and 8 more.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "conversationsDistribution": [],
    "countryDistribution": [],
    "creditsDistribution": [],
    "downVotesDistribution": [],
    "hourDistribution": [],
    "leadsCountDistribution": [],
    "liveChatEmailsDistribution": [],
    "liveChatEscalationsDistribution": [],
    "liveChatSmsDistribution": [],
    "messagesDistribution": [],
    "ratingDistribution": [],
    "sentimentsDistribution": [],
    "status": "success",
    "topicsDistribution": [],
    "upVotesDistribution": [],
    "workflowEmailsDistribution": []
  }
}

Get chatbot training status

Get a chatbot's current training progress.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier allocated to the chatbot.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "progress": 0,
    "status": "success"
  }
}

Get confidence scores

Get confidence scores for the passages a chatbot would use to answer, as an estimate of how well it can answer before you send the message. Results are cached, so sending the same messages afterwards stays fast.

Inputs

MessagesmessageslistRequired
The messages of the conversation, oldest first. On this endpoint, non-image files are not downloaded or parsed (only a provided content is used), and extra files are not dropped.
Base modelbase_modeltext
Overrides the chatbot's baseModel setting when processing the RAG context for this request.One of: gpt-4o-mini, gpt-4o, gpt-4, gpt-4.1-nano, gpt-4.1-mini, gpt-4.1, gpt-5-nano, gpt-5-mini, gpt-5, gpt-5.1, and 37 more.
Chatbotchatbot_idtextRequired
The ID of the chatbot.

Outputs

responseobject
When the chatbot uses `custom-model` or ignores its data source, the body is only `{ "status": "success", "score": 0 }`.
Example output
{
  "response": {
    "averageScore": 0,
    "embeddingText": "example",
    "score": 0,
    "scoresArray": [],
    "sources": [],
    "status": "success"
  }
}

Get conversations

List a chatbot's conversations that have at least one user message, most recently active first. Gets one page; to see more, run it again with the next page it gives.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier allocated to the chatbot.
Start timestampstart_timestampdate and time
Start of the time range (inclusive), in milliseconds since the Unix epoch. A conversation matches when it was created or last active within the range.
End timestampend_timestampdate and time
End of the time range (exclusive), in milliseconds since the Unix epoch.
Startstartnumber
The start position of the results, used for pagination. A negative value disables pagination and returns all results.
Sizesizenumber
The maximum number of results returned after the start position, used for pagination. A negative value returns all remaining results.
Sourcesourcelist
Filter conversations by source. Repeat the parameter to match several sources.One or more of: widget, iframe, site, api, Whatsapp, WhatsappWebAPI, Messenger, Instagram, Telegram, Line, and 5 more.
Lead IDlead_idtext
The UUID of the lead, which can be obtained from the lead-submission webhook event.
Requested live chatrequested_live_chattrue/false
true returns only conversations that requested live chat support, false only those that did not.
Live chat endedlive_chat_endedtrue/false
true returns only conversations whose live chat has ended, false only those whose live chat has not ended.
Overal ratingoveral_ratingnumber
Filter conversations by their overall rating (1-5).From 1.0 to 5.0.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "conversations": [],
    "status": "success",
    "total": 1
  }
}

Get customers

List the customers (leads) a chatbot has collected, newest first. Gets one page; to see more, run it again with the next page it gives.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier allocated to the chatbot.
Start timestampstart_timestampdate and time
Start of the time range (inclusive) for when customers were collected, in milliseconds since the Unix epoch.
End timestampend_timestampdate and time
End of the time range (exclusive), in milliseconds since the Unix epoch.
Startstartnumber
The start position of the results, used for pagination.
Sizesizenumber
The maximum number of results returned after the start position, used for pagination. A negative value returns all remaining results; 0 returns a 400 error.
Sourcesourcelist
Filter customers by the source they were acquired from. Repeat the parameter to match several sources.One or more of: widget, iframe, site, api, Whatsapp, WhatsappWebAPI, Messenger, Instagram, Telegram, Line, and 5 more.
Lead IDlead_idtext
Filter by the lead's UUID (leadId).
Psidpsidtext
Filter by the page-scoped user ID (Messenger).
Ig usernameig_usernametext
Filter by the Instagram username.
Usernameusernametext
Filter by the Slack or Telegram username.
Namenametext
Filter by the customer name (exact match).
Emailemailtext
Filter by the customer email address (exact match).
Phonephonetext
Filter by the customer phone number (exact match of the stored value).

Outputs

responseobject
Response.
Example output
{
  "response": {
    "customers": [],
    "status": "success",
    "total": 1
  }
}

Generate a shareable portal link for a specific conversation. The conversation portal must be enabled for the chatbot. The link is valid for 30 days.

Inputs

Conversation IDconversation_idtextRequired
The UUID of the conversation.
Chatbotchatbot_idtextRequired
The unique identifier of the chatbot.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "portalLink": "example",
    "status": "success"
  }
}

Get the current plan

Get a summary of the workspace's current subscriptions and privileges.

Inputs

None.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "email": "example",
    "name": "example",
    "plan": {
      "api": false,
      "chatbotCharactersLimit": 1,
      "chatbots": 1,
      "chatbotsLimit": 1,
      "customDomain": false,
      "customModel": false,
      "files": 1,
      "gatherLeads": false,
      "gpt4": false,
      "notification": false,
      "refreshingCredits": 1,
      "refreshingTrainingChars": 1,
      "remainingCredits": 1,
      "remainingTrainingChars": 1,
      "removeBrand": false,
      "shareAnalysis": false,
      "subscriptions": {
        "customize_plan_1": 1,
        "entry": 1,
        "entry_legacy": 1,
        "extra_chatbots": 1,
        "extra_messages": 1,
        "free": 1,
        "live_chat": 1,
        "professional": 1,
        "remove_brand": 1,
        "remove_label": 1,
        "scrape": 1,
        "sso_login": 1,
        "standard": 1,
        "storage": 1,
        "webhook_delivery": 1
      },
      "userAnalysis": false
    },
    "status": "success"
  }
}

List actions

Get all configured actions for the specified chatbot. Gets one page; to see more, run it again with the next page it gives.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier allocated to the chatbot.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "actions": [],
    "status": "success"
  }
}

Query the knowledge base

Get the passages from a chatbot's knowledge base that are relevant to a query, for example to give a custom GPT. Results are cached by query.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier allocated to the chatbot.
QueryquerytextRequired
The text used to search the knowledge base by vector (cosine) similarity.
Countcountnumber
The minimum number of context segments to select when available. More are added until a minimum token budget is met, then the list is trimmed to a maximum token budget, so the result can hold more or fewer. Only segments with a positive cosine similarity to the query are returned.

Outputs

responselist
Response.
Example output
{
  "response": []
}

Retrain an existing chatbot

Retrain a chatbot built from your own uploaded data, adding new information to its knowledge base or removing existing data.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier allocated to the chatbot.
Source textsource_texttext
Text data intended for chatbot training, subject to character limits based on your plan. A chatbot has one text source: this value replaces the current text. Send an empty string to remove the text; omit the field to keep the current text.
URLS to scrapeurls_to_scrapelist
A list of URLs provided for text content extraction. The web crawling process concludes once the list of URLs is fully processed or the chatbot's character limit is reached. URLs must begin with 'http://' or 'https://'. Applicable only to chatbots using the custom-data-upload model. A website that already exists with the same URL is recrawled to get the latest content.
Optionsoptionsobject
Configuration options that control how websites are scraped from the URLs provided in the urlsToScrape parameter. These options affect the content extraction process for each URL being crawled.
Productsproductslist
A list of products for training the chatbot.
Q and asq_and_aslist
A list of Q&As for the chatbot to learn specific topic answers. Items without both a question and an answer are skipped, and a pair identical to an existing one is not added twice.
Deletesdeleteslist
A list of knowledge sources to remove from the chatbot's knowledge base. To remove the plain text source, send sourceText as an empty string instead.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "chatbotId": "example",
    "consumedCharacters": 1,
    "dataSources": [],
    "remainingCharacters": 1,
    "status": "success"
  }
}

Send a message

Send messages to a chatbot and get its reply. It can also reply in an OpenAI-compatible format.

Inputs

MessagesmessageslistRequired
The messages of the conversation, oldest first. Every role must be user or assistant, and the last message must not be empty.
Chatbotchatbot_idtextRequired
The ID of the chatbot.
Include reasoninginclude_reasoningtrue/false
Whether to include reasoning in the response. This parameter overrides the chatbot's reasoning settings and has an effect only with models that support reasoning. For Claude models, a chatbot with AI actions shows reasoning only in the answer generated after an action is called. Non-streaming responses start with the reasoning in a tag followed by a blank line, for example {reasoning}\n\n{response}. Streaming responses wrap the reasoning in … for Claude models and … for deepseek-r1.
Base modelbase_modeltext
Overrides the chatbot's baseModel for this request, i.e. the LLM that generates the answer. In a HIPAA workspace, a model not covered by the BAA is rejected.One of: gpt-4o-mini, gpt-4o, gpt-4, gpt-4.1-nano, gpt-4.1-mini, gpt-4.1, gpt-5-nano, gpt-5-mini, gpt-5, gpt-5.1, and 37 more.
Base promptbase_prompttext
Overrides the chatbot's basePrompt (system prompt) for this request. Must be a non-empty string.
Append messagesappend_messagestrue/false
If true, messages must contain exactly one user message, which is appended to the stored messages of the conversation with the same conversationId. If false, it is assumed that the messages parameter contains the entire message history.
Conversation IDconversation_idtext
ID of the current conversation. If omitted, the chatbot generates one, which is not returned in the response.
Skip cacheskip_cachetrue/false
When true, bypasses the response cache lookup for this request and runs the LLM pipeline normally. On success, the new response overwrites the prior cached entry under the same cache key, unless an AI action was triggered or failed (those responses are never cached). This is useful for regenerating answers when the cached response is no longer wanted.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "choices": [],
    "id": "example"
  }
}

Update chatbot settings

Change a chatbot's settings. Only the fields you fill in are changed; empty ones are left as they are.

Inputs

Chatbotchatbot_idtextRequired
The unique identifier allocated to the chatbot.
Chatbot namechatbot_nametext
The new name of the chatbot.
Base promptbase_prompttext
The base prompt (instructions) for the chatbot.
Initial messagesinitial_messageslist
Array of initial messages the chatbot shows when a conversation starts.
Suggested messagessuggested_messageslist
Array of suggested messages shown in the chatbot.
Visibilityvisibilitytext
The visibility of the chatbot. It controls whether your chatbot is accessible to others.One of: private, public.
Only allow on added domainsonly_allow_on_added_domainstrue/false
Whether the chatbot can be embedded (iframe or widget) on other websites; false disables embedding. Allowed domains are set by domains.
Domainsdomainslist
The domains on which the iframe and widget are allowed. Replaces the current list.
Disallowed URLSdisallowed_urlslist
Pages on which the chat widget is not shown. Replaces the current list.
Rate limit pointsrate_limit_pointsnumber
The maximum number of messages one device can send every rateLimitTimeframe seconds. Each message counts as one here, however many message credits its model consumes.
Rate limit timeframerate_limit_timeframenumber
The time window, in seconds, to which the rateLimitPoints message limit applies.
Rate limit messagerate_limit_messagetext
The error message shown to the user when the conversation is rate limited.
Base modelbase_modeltext
The large language model (OpenAI GPT, Anthropic Claude, Google Gemini or DeepSeek) used in our RAG processing pipeline. In a HIPAA workspace, models not covered by a BAA are rejected.One of: gpt-4o-mini, gpt-4o, gpt-4, gpt-4.1-nano, gpt-4.1-mini, gpt-4.1, gpt-5-nano, gpt-5-mini, gpt-5, gpt-5.1, and 37 more.
Temperaturetemperaturenumber
The temperature of the AI model, in the range [0, 1].From 0.0 to 1.0.
Store chatstore_chattrue/false
Indicates whether chat history is stored on our server. All chatbot models, including medical-chat-human and medical-chat-vet, start with storage enabled by default. Storage can be disabled for any model.
Q a linksq_a_linkslist
Links to frequently asked questions pages shown in the chatbot. Replaces the current links; items without both text and link are dropped.
Show calendarshow_calendartrue/false
Specifies whether the Calendly booking button should be displayed in the chatbot widget or iframe.
Calendar URLcalendar_urltext
The Calendly URL where users can schedule appointments.
Calendar messagecalendar_messagetext
The message to display for appointment scheduling.
Cal comcal_comobject
The Cal.com booking button settings. The object replaces the current settings as a whole.
Enable bouncing animationenable_bouncing_animationtrue/false
Whether the chat bubble bounces at the start of a session to attract attention.
Enable bubble soundenable_bubble_soundtrue/false
Whether a sound plays when the initial-message bubble pops up.
Bubble sizebubble_sizenumber
The size of the chat bubble button, in pixels. Values that are not positive numbers are ignored.
Input placeholder msginput_placeholder_msgtext
The placeholder text of the message input.
Rerankreranktrue/false
Whether retrieved knowledge base chunks are reranked before the chatbot answers. Reranking costs extra message credits.
Sourcessourcesobject
Whether and how the sources used for an answer are shown under it.
Footerfootertext
Text shown at the bottom of the chat widget.
Ignore data sourceignore_data_sourcetrue/false
Whether your uploaded data should be ignored when the chatbot responds. If set to true, the chatbot answers like a general-purpose AI chatbot without your knowledge base. This option applies to chatbots based on the custom-data-upload model.
Leads configleads_configobject
The information to be collected from potential customers.
Stylesstylesobject
The appearance settings of the chat widget.

Outputs

responseobject
Response.
Example output
{
  "response": {
    "status": "success"
  }
}

On this page