​
AI Agents in SigmaMind

An Agent in SigmaMind is a conversational workflow builder designed to handle customer interactions intelligently.
Agents are created using nodes, where each node represents a step in the conversation flow. By connecting these nodes together, you can design how the AI agent responds, analyzes, waits, branches, or takes actions across different channels such as voice, chat, and email.

​
Purpose of an Agent

Agents are at the core of how SigmaMind powers intelligent, automated conversations. They enable you to:
  • Build structured conversational workflows.
  • Automate customer support, sales, and service processes.
  • Handle inbound and outbound interactions seamlessly.
  • Integrate external systems and tools into conversations.
  • Ensure conversations follow a logical, predictable flow.
Agents transform business processes into automated, intelligent experiences that scale across channels.

​
How an Agent Works

When a phone number or channel (such as voice, chat, or email) is connected to an Agent, every incoming call or message is processed through that Agent’s workflow.
  1. Triggering an Agent
    • Agents are triggered by events, such as:
      • Conversation Started
      • Conversation Updated
      • Conversation Ended
    • Triggers can also be filtered by event type to control when and how an Agent activates.
  2. Flow Through Nodes
    • Once triggered, the Agent evaluates nodes in sequence.
    • Example node behaviors:
      • SendResponse – Sends a reply to the customer.
      • Analyze – Interprets customer intent from a message.
      • Wait – Pauses until the customer responds.
      • Branch/Jump – Routes the flow based on conditions or logic.
      • AppAction / HelpDeskAction / ViceAction – Executes external actions or tool integrations.
  3. Continuous Conversation
    • The Agent continues to move through nodes as events occur.
    • When a Wait node is active, the Agent pauses until a new customer message arrives.
    • If an Analyze node is used, the Agent evaluates customer input and routes accordingly.
  4. Multiple Actions and Tools
    • Agents can include multiple actions, integrated via tools, allowing them to perform external tasks during a conversation.

​
Agent Flow Example

A typical Agent workflow might follow this sequence:
  1. A customer calls a number connected to SigmaMind.
  2. The Conversation Started event triggers the Agent.
  3. The Agent sends an initial greeting using a SendResponse node.
  4. The customer replies, and the message is evaluated in an Analyze node to detect intent.
  5. Depending on the intent, the flow moves to a Branch node, directing the customer toward the right workflow.
  6. If additional input is required, a Wait node holds until the next customer message.
  7. The Agent may call external APIs or services using an AppAction node.
  8. The conversation continues until it reaches a logical conclusion, ending with a closing message or transferring to a human agent if required.

​
Requirements for Using an Agent

  • At least one communication channel (voice, chat, or email) must be connected.
  • An Agent must be configured with nodes that define the conversation flow.
  • Triggers must be defined so the Agent knows when to start.
  • If external integrations are required, they must be connected through supported actions or webhooks.
A well-structured Agent workflow ensures smooth, predictable conversations and minimizes failure cases.

​
What You Can Achieve with Agents

By building Agents in SigmaMind, you can:
  • Provide 24/7 customer support through intelligent automation.
  • Route conversations dynamically based on customer intent.
  • Integrate with external platforms like CRM, e-commerce, or helpdesk systems.
  • Personalize responses using dynamic variables and context.
  • Reduce human workload by automating repetitive inquiries.

​
Summary

  • Agents are workflow builders that power AI-driven conversations in SigmaMind.
  • They are triggered by events such as conversation start, update, or end.
  • Agents are built with nodes, each handling a specific step in the flow.
  • Agents can analyze customer input, branch into workflows, pause for responses, and call external actions.
  • With proper setup, Agents provide structured, scalable, and intelligent automation across communication channels.
Agents are the foundation of conversational AI in SigmaMind, enabling you to design, control, and optimize how your AI engages with customers.