One-click install
npx skills add https://github.com/cartesia-ai/skills --skill line-voice-agent-cartesia-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: line-voice-agent
Source: https://github.com/cartesia-ai/skills/tree/main/skills/line
Command: npx skills add https://github.com/cartesia-ai/skills --skill line-voice-agent-cartesia-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you build and deploy production voice agents that can handle speech-to-text, text-to-speech, interruptions, and multi-agent handoffs without manually managing the full audio orchestration stack.

Core Features & Use Cases

  • Real-time voice agent deployment: Create a Python voice agent using the Cartesia Line SDK and deploy it with the cartesia CLI for managed auto-scaling.
  • LLM tool calling and multi-agent workflows: Use LiteLLM-backed LLM providers plus built-in and custom tools to coordinate tasks and transfer conversations between specialized agents.
  • Interruption-safe conversations: Support real-time barge-in behavior so the agent can stop and recover when the caller interrupts.

Quick Start

Tell your AI agent to explain how to create a minimal Line SDK voice agent and deploy it using cartesia, including the required main.py entry point and cartesia.toml configuration.

Frequently Asked Questions about line-voice-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I deploy a real-time voice agent with Python that handles speech-to-text and text-to-speech?

Deploy a real-time voice agent by building a Python service with the Cartesia Line SDK and using the `cartesia` CLI for managed auto-scaling. The platform handles STT, TTS, and audio orchestration automatically.

Can I use LiteLLM model strings and external LLM providers for tool calling in my voice agent?

Yes, you can use LiteLLM model strings and provider API keys to enable LLM tool calling. The SDK supports built-in and custom tools for coordinating tasks and generating LLM-grounded responses.

How do real-time voice agents handle caller interruptions during a conversation?

Real-time voice agents support barge-in interruptions natively. The audio orchestration stops and recovers automatically when a caller interrupts, ensuring smooth conversational flow.

What's the best way to set up multi-agent handoffs for specialized voice tasks?

Set up multi-agent handoffs using the Line SDK's built-in loopback, passthrough, and handoff tool patterns. These deterministic patterns transfer conversations between specialized agents based on caller needs.

Do I need a specific configuration file to deploy my voice agent with the cartesia CLI?

Yes, you need a `main.py` entry point containing your `VoiceAgentApp`-based Python service and a `cartesia.toml` configuration file to deploy your voice agent using the `cartesia` CLI.

Does the Line SDK support websocket integration for web and mobile voice applications?

Yes, the Line SDK supports websocket integration for web and mobile deployments. It also supports phone and telephony integrations through its managed audio orchestration stack.