The best DX for building Voice AI

Everything you need to build voice AI agents, right from your terminal. Drop Vapi into any project, debug webhooks locally, and turn your IDE into a Vapi expert with MCP integration.

Install Vapi CLI View Docs

Source Code Read Announcement →

Terminal

➜  ~ vapi login
✓ Authenticated successfully

➜  ~ vapi init
✓ Detected: Next.js application
✓ Installed @vapi-ai/web SDK
✓ Generated components & examples

➜  ~ vapi mcp setup
✓ Created/updated Cursor config: /Users/hackerman/.cursor/mcp.json
✓ Successfully configured Cursor

➜  ~ vapi auth switch
✓ Using account 'account-1751925877'|

Install in 10 seconds

One command to get started. Works on macOS, Linux, Windows, and Docker.

curl -sSL https://vapi.ai/install.sh | bash

Everything you need, one command away

The Vapi CLI brings the full power of our platform to your terminal, with developer experience as the top priority.

Drop into Any Project

Auto-detect your tech stack and generate working code examples instantly.

vapi init

Complete Terminal Control

Manage assistants, calls, phone numbers, and more without leaving your terminal.

vapi assistant create

Local Webhook Testing

Debug webhooks locally. Forward events to your dev server easily.

vapi listen --forward-to localhost:3000

IDE Intelligence

Turn your IDE into a Vapi expert with MCP integration. No more API hallucinations.

vapi mcp setup

Multi-Account Support

Switch between organizations, staging, and production environments seamlessly.

vapi auth switch production

Logs Observability

View logs, debug issues, and monitor your voice AI infrastructure in real-time.

vapi logs calls

Turn your IDE into a Vapi expert

With MCP (Model Context Protocol) integration, your IDE's AI assistant gains complete, accurate knowledge of Vapi's APIs, features, and best practices.

No More Hallucinations

Get accurate Vapi code suggestions every time

Complete Documentation

Access all Vapi docs directly in your IDE

Real Code Examples

Working examples for every use case

Supported IDEs

  • Cursor (AI-powered code editor)
  • Windsurf (Codeium's AI IDE)
  • VSCode (With GitHub Copilot)

Learn More About MCP

Example Code

const assistant = await vapi.assistants.create({
  name: "Customer Support",
  model: {
    provider: "openai",
    model: "gpt-4"
  },
  voice: {
    provider: "elevenlabs",
    voiceId: "paula"
  }
});

Start building voice AI agents today

Join thousands of developers building the future of voice AI with Vapi.

Install Vapi CLI Read Documentation

Source Code Explore the source code, contribute, and report issues Announcement Read about the launch and what's coming next Community Join our Discord for support and discussions