Skip to content
Vapi MCP Server

Vapi MCP Server

by Vapi
Released
2 views
v0.0.11
MIT
Repository Docs npm
Developer Tools

Summary

Vapi's official MCP server for building voice agents — create assistants, place and schedule outbound phone calls, and manage phone numbers from your AI client.

Features

  • Create and configure Vapi voice assistants
  • Place outbound calls with dynamic per-call variables
  • Schedule calls for a future time
  • Manage phone numbers and assistants
  • Local stdio server and hosted streamable-HTTP endpoint
  • Companion Agent Skill included in the repository

Installation

Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.

Any MCP-compatible agent

Most agents (Claude, Cursor, Windsurf, VS Code, and more) read a standard mcpServers configuration.

  1. Open your agent's MCP configuration file.
  2. Merge the snippet below into it, filling in the environment variables with your own values.
  3. Restart the agent — the "Vapi MCP Server" tools will be available.
{
  "mcpServers": {
    "vapi-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@vapi-ai/mcp-server"
      ],
      "env": {
        "VAPI_TOKEN": "<your_vapi_token>"
      }
    }
  }
}

Description

Vapi builds voice AI agents that answer and place real phone calls, and this is its official MCP server: it puts assistant creation and call control directly into whatever MCP client you already work in.

What you can do with it
  • Create and configure voice assistants — describe the assistant you want and let the agent build it, rather than assembling it in the dashboard.
  • Place outbound calls — dial a number with a named assistant and pass structured variables into the prompt, so the same appointment-reminder assistant can be reused with a different customer name, date and time on every call.
  • Schedule calls for later — hand it a time and it queues the call.
  • Manage phone numbers and assistants across the account.

The dynamic-variable support on create_call is what makes this practical at more than demo scale: the assistant prompt stays a template and the per-call data is passed in, instead of a new assistant being minted per contact.

Two ways to connect

Local (stdio) — npx -y @vapi-ai/mcp-server with VAPI_TOKEN in the environment. In Claude Code that is a single command: claude mcp add -e VAPI_TOKEN=… vapi -- npx -y @vapi-ai/mcp-server.

Remote (streamable HTTP) — point a client that supports remote servers at https://mcp.vapi.ai/mcp with an Authorization: Bearer header. Clients that only speak stdio can bridge to it with mcp-remote.

Also ships a skill

The repository includes a companion Agent Skill under skill/, a set of reusable instructions for hosts that support Agent Skills — useful because voice-agent configuration has enough domain-specific vocabulary that raw tool schemas alone leave a model guessing.

Requires a Vapi API key from the Vapi dashboard. MIT licensed.

Covered in the Weekly

Related MCP Servers

MCP: Arm MCP Server

by Arm

New

Arm's official MCP server for porting and tuning software on Arm — semantic docs search, codebase migration scanning, container architecture checks and LLVM-MCA assembly analysis.

Developer Tools

MCP: AntV Chart MCP

by AntV

New

Official AntV MCP server that turns a description of your data into a rendered chart — 26 visualization types from bar and line to sankey, mind-map, fishbone and geographic maps.

Developer ToolsDatabases & Data
1 views

MCP: Headroom

by Headroom Labs

New

Context-compression MCP server that shrinks tool outputs, logs, files and RAG chunks before they reach the model, cutting 60-95% of tokens on JSON payloads.

Developer Tools
3 views

MCP: Engram

by Gentleman Programming

New

Persistent memory for coding agents as a single Go binary — SQLite and FTS5 behind MCP, an HTTP API, a CLI and a TUI, with no Node, Python or Docker required.

Developer Tools
2 views
Browse all MCP servers →