Voice Customer Support

Overview

Let callers ask questions about your product and get spoken answers based on your support documents. The agent searches the support namespace in your environment's Connic Retrieval, then uses the returned passages to answer. When an issue needs follow-up, the prompt instructs it to confirm the details and ask for consent before calling the mock ticket tool. The tool returns a demo receipt without storing a ticket or notifying a support team. Customize the YAML prompt, voice settings, and Python tools, and manage your knowledge through Retrieval. The template uses Gemini Live with your own provider credentials. Connic runs the voice session, search, and tools.

Use cases

Architecture

Main template files

Install these files with connic init my-project --templates=voice-customer-support. The CLI places agent definitions in a folder named after the template under agents/.

voice-customer-support/
  agents/
    voice-customer-support.yaml
  tools/
    voice_support.py
  requirements.txt
  README.md

Get started

Install the template, create a Connic project, and deploy through a connected Git repository or the CLI.

1. 1

   Install and scaffold

   Install the SDK and create a project from this template.

   terminal

   Copy

   pip install connic-composer-sdk
   connic init my-project --templates=voice-customer-support

   Then cd my-project

2. 2

   Deploy

   Choose a deployment method. A connected Git repository deploys on push; use the CLI for projects without a connected Git repository.

   Git integration

   1. 1

      In Connic: Project Settings → Git & Environments, connect the GitHub repository

   2. 2

      On the same page, map a branch (e.g. main) to the Production environment

   3. 3

      Push the scaffolded project to that repository

   terminal

   Copy

   git add .
   git commit -m "Add Voice Customer Support template"
   git push origin main

   CLI deploy

   1. 1

      In Connic: Project Settings → API Keys & MCP Auth, create an API key and copy project ID

   2. 2

      Run connic login in the project folder

   3. 3

      Check the configuration with connic lint, then deploy with connic deploy. Test voice behavior in a live phone call; connic test does not support voice sessions.

   terminal

   Copy

   connic login
   connic lint
   connic deploy

3. 3

   Connect and configure

   Before deploying, save Gemini credentials under Project Settings → LLM Provider for gemini/gemini-3.8-live. In the deployment environment, open Retrieval and add your support documents as text, file uploads, or a synced source using the support namespace. Wait for indexing to finish and check results with Retrieval Search. Link the deployed agent to a Connic connector that supports voice and call the configured number. Connect the mock ticket tool to your helpdesk API when you are ready to create real tickets.

   Full connector setup guideQuickstart guideVoice docs