Skip to main content

​
Overview

LiveKit Agents is a framework for building server-side AI agents that interact with users in real time. By integrating with AKOOL’s Streaming Avatar, you can build AI-powered avatar experiences using LiveKit’s agent infrastructure. This guide walks you through setting up the AKOOL LiveKit Agent server locally and testing it with the LiveKit Playground. Repository: AKOOL LiveKit Agents

​
Prerequisites

  • Python (LTS version)
  • uv package manager
  • A LiveKit Cloud account (for API key, secret, and server URL)
  • An OpenAI API key
  • AKOOL Client ID and Client Secret (available from your AKOOL dashboard)

​
Setup

1

Clone the repository

git clone https://github.com/AKOOL-Official/livekit_agents.git
cd livekit_agents
2

Switch to the feature branch

git checkout feature/livekit-plugins-akool
3

Install dependencies

Run the following command in the project root directory:
uv sync
4

Configure environment variables

Navigate to the examples folder and create a .env file from the template:
cd examples
cp env.example .env
Fill in the following credentials in the .env file:
LIVEKIT_API_KEY=""
LIVEKIT_API_SECRET=""
LIVEKIT_URL=""
OPENAI_API_KEY=""
AKOOL_CLIENT_ID=""
AKOOL_CLIENT_SECRET=""
5

Start the agent server

uv run examples/avatar_agents/akool/agent_worker.py dev
6

Test in the playground

Open the LiveKit Agents Playground and sign in with your LiveKit account to interact with the agent.