Agent Blocks process data, interact with customers, and execute AI-driven decisions.

​
Voice Agent

Call details for voice agent’s outgoing calls with customers.

Inputs

  • Bot_Number: Number voice agent will use to make a call. This can be linked to a connecting block or selected from dropdown option. Data linkage will override pre-selected dropdown option.
  • Use_Case: Define voice agent to be used. This can be linked to a connecting block or selected from dropdown option. Data linkage will override pre-selected dropdown option.
  • Customer_Number: Customer’s phone number.
  • Customer_Details: Customer’s details such as name, email in JSON format
  • Schedule_Time: Schedule customer’s call at a specific future time. Follow ISO date format YYYY-MM-DDTHH:MM:SSZ. Default value: current time, to trigger the action immediately.
  • Agent_Options: If provided, these will override settings as defined in the Voice Agents section. E.g. if you want to change the language, voice, initial prompt, etc.

Outputs

  • Bot_Number: Number voice agent used to make a call.
  • Use_Case: Voice agent that handled the call.
  • Customer_Number: Customer’s phone number.
  • Customer_Details: Customer’s details such as name, email in JSON format.
  • Call_Time: Time at which customer dialed vocie agent. This follows ISO date format YYYY-MM-DDTHH:MM:SSZ.
  • Call_Status: Status of the call - connected or couldn't connect.
  • Call_Duration: Status of the call in seconds.
  • Call_Direction: Direction of the call - outbound.
  • Transcript: Transcript of the call with speaker tags.

​
AI Query Agent

Enables automated decision-making to perform multiple actions (e.g. scheduling callbacks, sending emails, generating tickets) based on dynamic conditions, mimicking human-like intelligence.

Inputs

  • Query: User input or query requiring AI processing. Example:
1. Generate a ticket with comments as per discussion and assign to relevant department.
2. Send an email to the customer with ticket details, CC to [email protected]
3. If the customer asked to call later / callback, schedule a callback accordingly
4. If the customer was willing to speak but the call got hung up abruptly, schedule a callback in 5 minutes
  • Tools: Select the tools you want to use to perform specific actions, such as ending an email or generating a ticket.
  • System_Context: Provide additional context about the user or situation, such as a call transcript or customer details, to help the AI make more informed decisions and keep the conversation truly personalized. Example text prompt. Remember, you can also link with multiple outputs from other blocks.
You are a backend agent for SubVerse AI to perform multiple tasks based on customer's discussion with virtual agent.
  • Knowledge_Base: Option to Select Knowledge Base to pull relevant information from your company documents.

Outputs

  • LLM_Response: The AI Query Agent generates a response based on the user’s query, using the context and knowledge base you provided.
  • Tool Outputs: The block also produces outputs from the selected tools, which can be used as inputs to function blocks, allowing you to automate follow-up tasks or notifications.