Skip to main content

​
Overview

A Chat Agent is a text-based AI that communicates with users through typed messages. It’s ideal for customer support bots, FAQ assistants, lead qualification flows, or any use case where users interact by typing rather than speaking. Chat agents can be embedded in your website via a web chat channel, integrated into your product, or tested directly from the Subverse dashboard. When to use a Chat Agent:
  • You want a bot that responds to typed messages
  • You need to connect a live chat widget to your website
  • Your use case doesn’t require voice or phone capabilities

​
How to Create

  1. Navigate to Agents in the left sidebar.
  2. Click Add Agent and select Chat as the agent type.
  3. Fill in the Identity fields — name and description.
  4. Write the agent’s system prompt under Instructions.
  5. Configure the LLM Service (provider and model).
  6. Optionally configure Memory, Functions, Pipeline Settings, and Analytics.
  7. Click Save.

​
Deploy: Connect to a Channel

A saved Chat Agent can be tested from the dashboard immediately. To deploy it for real conversations, connect it to a communication channel. Chat agents support two channel types:

WhatsApp Channel

Connect your WhatsApp Business number — customers message your number and the agent replies automatically

WebChat Channel

Enable chat on your website or app via webhooks and APIs
Each channel requires credentials (WhatsApp API key or WebChat credential) stored securely in encrypted form. See Credentials to set these up before creating a channel.

​
How to Test

  1. Open the agent from the Agents list.
  2. Click the Chat icon or open the Chat Sidebar on the right side of the agent page.
  3. Type a message and press Enter.
  4. The agent responds in real time.
  5. Adjust the system prompt or settings and re-test as needed.
You don’t need a live channel to test — the Chat Sidebar lets you interact with the agent directly from the dashboard.

​
Tips & Notes

  • Name format: Agent names must be lowercase with only letters, numbers, and underscores. Example: customer_support_v2.
  • Memory requires both toggles: “Remember” saves data after a session; “Use Memory” reads it back at the start of the next session. You need both enabled for persistent memory to work.
  • Knowledge Base: Attach a KB to let the agent answer questions from your documents without embedding all content in the system prompt.
  • Max Duration: Conversations that exceed maxDuration are cut off. Increase this for complex support flows.

Chat Agent Configuration

See the full configuration reference — every field, type, and default value