System Entities

References the [[SystemLanguage]] entity. Stores all available languages for the AI Agent, including the AI Agent languages and any additional languages used through the conversation time machine translation.

As [[ChatUser]].AvailableLanguages is an Entity List that references [[SystemLanguage]], each element within the list is an object with the following properties:

• code (String). The standard language code (e.g., "en-US", "fr").
• friendlyName (String). The name of the language (e.g., "English", "French").

[[ChatUser]].AvailableLanguages = [
        {"code": "ro", "friendlyName": "Romanian"}, 
        {"code": "en-US", "friendlyName": "English"}, 
        {"code": "fr", "friendlyName": "French"}
]                                    

The unique identifier for the channel where the conversation is taking place. The value will vary based on the channel:

• [[ChatUser]].ChannelId == “directline”
• [[ChatUser]].ChannelId == “msteams”
• [[ChatUser]].ChannelId == “meta-whatsapp”
• [[ChatUser]].ChannelId == “nexmo-whatsapp”
• [[ChatUser]].ChannelId == “facebook”
• [[ChatUser]].ChannelId == “slack”
• [[ChatUser]].ChannelId == “googlechat”
• [[ChatUser]].ChannelId == “instagram”
• [[ChatUser]].ChannelId == “line”
• [[ChatUser]].ChannelId == “email”
• [[ChatUser]].ChannelId == “outlook”
• [[ChatUser]].ChannelId == “wechat”
• [[ChatUser]].ChannelId == “zendesk”
• [[ChatUser]].ChannelId == “ringcentral”
• [[ChatUser]].ChannelId == “telegram”
• [[ChatUser]].ChannelId == “sms”
• [[ChatUser]].ChannelId == “viber”
• [[ChatUser]].ChannelId == “audiocodes”
• [[ChatUser]].ChannelId == “voicetext”
• [[ChatUser]].ChannelId == “webex”
• [[ChatUser]].ChannelId == “workplace”
• [[ChatUser]].ChannelId == “cuedesk”
• [[ChatUser]].ChannelId == “vonage-sms”

The ID of a live chat conversation.

Set value to "true" to block the file upload at conversation level,  by simply manipulating the conversation context, even for a limited duration (e.g., only during live chat session). The default value is "false".

Use cases for blocking file upload:

• Block file upload on a specific conversation area.
• Prevent users from sending files during a live chat session. Block file upload when placing the user in a queue and unlock when user goes to user-AI Agent chat.
• Block file upload for specific users based on conversation parameters.

Determines where the system predicts user intent within a conversation. By default, the system predicts using NLP if [[ChatUser]].PredictDestination is null.

Available values:

• "KB" – Predicts only using the knowledge base (KB), without NLP.
• "NLP" – Predicts only using natural language processing (NLP), without KB.
• "NLP,KB" – Predicts first using NLP. If no result is found, it predicts using KB.
• "KB,NLP" – Predicts first using KB. If no result is found, it predicts using NLP.

By setting this parameter in the flow, you can control prediction at the conversation level or within specific conversation segments. The prediction destination remains the same throughout a conversation unless explicitly changed.

For example, if you set the prediction destination in the Welcome flow, it applies to all subsequent user messages until modified. This parameter ensures a structured execution pipeline, reducing the need for separate intent-matching actions.

Allows you to restrict the Druid Conductor prediction to a specific set of AI Agents. Set [[ChatUser]].RestrictedTargetBots to an array of AI Agent names in the conversation context before the prediction occurs.

This limits the scope of the prediction, ensuring that only the specified AI Agents are considered for engagement. The field uses a JArray type, so you can provide a list of AI Agent names, such as: [[ChatUser]].RestrictedTargetBots = ["HR", "Finance"]. To remove the restriction and allow the Druid Conductor to engage any eligible AI Agent, set [[ChatUser]].RestrictedTargetBots to null.

Stores the intervention ID from RingCentral, enabling you to track agent engagements within a thread.

You can use this field in authoring flows in conjunction with the RingCentral AI Agentcomplete event (via REST integration) to automatically close an intervention once the conversation is resolved.

The SAML authentication response returned by the IdP after the user completes the login process. This XML contains the user’s identity information and is used by Druid to update the session.

The period for which an anonymous user will not be placed on the waiting queue within the current conversation (the user is blocked). It is the sum of the time when the user was blocked for Live Chat and the BlockIntervalMinutes set in the HelpdeskClient.IsBlockedForLiveChat internal action.

For more information, see Block Chat Users.

You can use this field (set to null or in the past) to author flow paths for unblocking anonymous users.

References the [[Email]] system entity to store information for conversations via the Office 365 channel.

Stores the conversation transcript as text.

--> My washing machine is leaking water //user message
<-- Unplug it and then check and tighten hoses. //agent message
--> It worked! // user message

Sets how the AI Agent replies to an incoming email. When true, the reply is sent to all original recipients (the combined To and Cc lists). When false, the reply is sent only to the sender (From). The default value is true.

The destination email address (or addresses, separated by semicolons) for forwarding. When this field is populated, the system forwards the email to the specified recipients and ignores the ReplyToAll field.

The authorization token of the Portal user.

The description of the flow matching the user intent.

You can use it to filter multiple intents.

The ID of the matching the user intent.

Indicates if the user says is comprised of multiple sentences or not.

Use it to set complex metadata filters within the conversation context (Set Variables > Code Extension). For more information, see KB Metadata.

Stores information regarding the matched flow(s).

This field references the system entity [[MatchingFlow]].

Stores all the intents predicted by the NLP engine. This field references the system entity [[MatchingFlow]].

Stores the ID of the data source that contains the article.

Stores the question of a knowledge base article.

Stores the URL of the document from public or unstructured data sources, such as file repositories or websites. You can use this field in Citations to provide clickable links to these documents.

Stores the URL of the document from private data sources, such as SharePoint or file shares, that require an authentication token to access. You can use this field in Citations to provide clickable links to these documents without requiring users to sign in.

Druid sets the FriendlyName to the name of the conversation's detected language in that specific language.

Examples:

• If the conversation name is English and it detects language change to French, than [[ChatUser].DetectedLanguage will be set to "French".
• If the conversation name is Romanian and it detects language change to French, than [[ChatUser].DetectedLanguage will be set to "Franceza".