livekit-agents

Build voice AI agents with LiveKit Cloud and the Agents SDK.

3|1|Updated Nov 30, 2025
One-click install
npx skills add https://github.com/PALabs-v1/AI_friend --skill livekit-agents-palabs-v1
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: livekit-agents
Source: https://github.com/PALabs-v1/AI_friend/tree/main/.claude/skills/livekit-agents
Command: npx skills add https://github.com/PALabs-v1/AI_friend --skill livekit-agents-palabs-v1

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Building real-time voice AI agents involves fast-changing SDKs, latency constraints, and workflow complexity that generic coding knowledge handles poorly. This Skill provides opinionated architectural guidance for LiveKit Cloud development and enforces verification against live documentation instead of stale model memory. ## Core Features & Use Cases - Opinionated Architecture Guidance: Covers latency optimization, context minimization, and voice-specific design principles for real-time conversations. - Workflow Structuring: Guides the use of agent handoffs and scoped tasks to avoid monolithic agents with bloated context and tool lists. - Mandatory Testing Workflow: Requires writing behavior tests for every agent implementation using LiveKit's testing framework. - Documentation Verification: Enforces use of the LiveKit MCP server or web search to verify all API signatures before writing code. - Use Case: A developer asks to build a customer support voice agent with a greeting phase, intake phase, and billing specialist handoff; the Skill guides the multi-agent structure, credential setup, and test coverage. ## Quick Start Ask the assistant to build a LiveKit voice agent on LiveKit Cloud and it will set up credentials, structure the workflow, and write tests.

Frequently Asked Questions about livekit-agents

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

FAQPage Schema
How do I build a voice AI agent with LiveKit Cloud?

Create a LiveKit Cloud project, set LIVEKIT_URL, LIVEKIT_API_KEY, and LIVEKIT_API_SECRET environment variables, then build your agent with the Agents SDK. Use LiveKit Inference for AI models so no separate provider API keys are needed.

How do I structure a complex LiveKit voice agent?

Use handoffs to transfer control between specialized agents and tasks for tightly-scoped operations. This avoids monolithic agents with bloated tool lists and system prompts that increase latency.

What is LiveKit Inference and should I use it?

LiveKit Inference is the recommended way to access AI models on LiveKit Cloud, providing model providers through your LiveKit credentials with consolidated billing. It eliminates the need to manage separate API keys for each AI provider.

Does this guidance apply to self-hosted LiveKit?

The skill targets LiveKit Cloud developers. Self-hosted users can apply the architecture and testing principles, but recommendations around LiveKit Inference and managed infrastructure do not apply directly.

Why must LiveKit API calls be verified against documentation?

The LiveKit Agents SDK evolves faster than model training cycles, so memorized API signatures, imports, and configuration options are frequently outdated. The skill requires checking live documentation via the LiveKit MCP server before writing code.

Are tests required for LiveKit agent implementations?

Yes, every agent implementation must include at least one test covering conversation flow, tool invocation, and error handling. Tests are only skipped if the user explicitly requests it, with a warning about production risks.