movoice-mcp
OfficialProvides speech recognition (ASR) and text-to-speech (TTS) capabilities for voice agents, allowing them to understand and speak with callers.
Provides high-quality text-to-speech voices for voice agents, enabling natural-sounding speech synthesis.
Allows using OpenAI's language models (e.g., GPT-4) and text-to-speech voices for voice agents, offering flexibility in LLM and TTS choices.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@movoice-mcp Create a customer support agent for my e-commerce store"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Movoice AI MCP Server
Create and manage voice AI agents directly from Claude Desktop or Cursor using natural language — no dashboard required.
"Create a customer support agent for my e-commerce store. Use a friendly female voice and make it handle order status and returns."
That's it. The agent is live and ready to take calls.
What you can do
Create a voice AI agent by describing it in plain English
List all your agents
Update agent prompts, voices, or settings
Delete agents
View call logs and transcripts
Related MCP server: Bolna MCP Server
Quickstart
Option 1 — npx (no install needed)
{
"mcpServers": {
"movoice": {
"command": "npx",
"args": ["-y", "@movoice/mcp"],
"env": {
"MOVOICE_API_KEY": "mv_live_your_key_here"
}
}
}
}Option 2 — Install globally
npm install -g @movoice/mcp{
"mcpServers": {
"movoice": {
"command": "movoice-mcp",
"env": {
"MOVOICE_API_KEY": "mv_live_your_key_here"
}
}
}
}Setup
Step 1 — Get your API key
Go to app.movoice.ai/admin/account → API Keys → Generate Key
Your key looks like: mv_live_xxxxxxxxxxxxxxxx
Step 2 — Add to your client
Claude Desktop
Edit ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"movoice": {
"command": "npx",
"args": ["-y", "@movoice/mcp"],
"env": {
"MOVOICE_API_KEY": "mv_live_your_key_here"
}
}
}
}Restart Claude Desktop.
Cursor
Go to Cursor Settings → MCP → Add Server:
{
"movoice": {
"command": "npx",
"args": ["-y", "@movoice/mcp"],
"env": {
"MOVOICE_API_KEY": "mv_live_your_key_here"
}
}
}VS Code (with MCP extension)
Add to .vscode/mcp.json:
{
"servers": {
"movoice": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@movoice/mcp"],
"env": {
"MOVOICE_API_KEY": "mv_live_your_key_here"
}
}
}
}Example prompts
Once connected, just talk naturally:
"Create a customer support agent for my e-commerce store called ShopBot. It should handle order status, returns, and general questions. Use a friendly female voice."
"Show me all my agents"
"Update the ShopBot's welcome message to say 'Hi! I'm ShopBot, your shopping assistant. How can I help?'"
"Create a sales agent for an insurance company that qualifies leads and books appointments. Make it speak Hindi."
"Show me the last 10 calls and summarise what customers were asking about"
"Delete the test agent"
Available tools
Tool | Description |
| List all your voice agents |
| Create a new voice agent |
| Get details of a specific agent |
| Update agent settings |
| Delete an agent |
| View recent call logs |
| Get full call details + transcript |
create_agent parameters
Parameter | Type | Required | Description |
| string | ✅ | Agent name (e.g. "Sales Bot") |
| string | ✅ | System prompt — personality, knowledge, behavior |
| string | ✅ | First thing the agent says when a call connects |
| string |
| |
| string | e.g. | |
| string |
| |
| string | Voice ID or name (default: rachel) | |
| string |
| |
| string | Language code e.g. | |
| string | Webhook URL to receive call events |
Environment variables
Variable | Required | Description |
| ✅ | Your Movoice API key ( |
| Override API base URL (default: |
Build from source
git clone https://github.com/Movoiceai/movoice-mcp.git
cd movoice-mcp
npm install
npm run buildThen point your MCP client to dist/index.js:
{
"mcpServers": {
"movoice": {
"command": "node",
"args": ["/path/to/movoice-mcp/dist/index.js"],
"env": {
"MOVOICE_API_KEY": "mv_live_your_key_here"
}
}
}
}Links
Movoice AI — Dashboard
Documentation — Full API docs
Issues — Report a bug
MCP Protocol — Learn about MCP
License
MIT © Movoice AI
Available Tools
7 toolscreate_agentA
Create a new voice AI agent on Movoice AI. The agent will be ready to take calls immediately.
Inputs: name (required), prompt (required), welcomeMessage (required), language, +6 more.
delete_agentA
Permanently delete a voice AI agent.
Inputs: id (required).
get_agentA
Get details of a specific voice AI agent by its ID.
Inputs: id (required).
get_callA
Get full details of a specific call including transcript and AI analysis.
Inputs: id (required).
list_agentsA
List all voice AI agents in your Movoice AI account.
Inputs: page, per_page.
list_callsA
List recent call logs for your account. Includes duration, sentiment, and transcript availability.
Inputs: page, per_page.
update_agentA
Update an existing voice AI agent. Only provide fields you want to change.
Inputs: id (required), name, prompt, isActive, +8 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 7 tools
Each tool targets a distinct resource and action: five tools handle agent lifecycle (list/get/create/update/delete), and two handle call logs (list/get). There is no overlap or ambiguity between the tools.
All tool names follow a consistent verb_noun pattern using snake_case, such as list_agents, create_agent, delete_agent, and list_calls. This makes the API predictable and easy for an agent to navigate.
Seven tools is a well-scoped set for a voice AI agent management server. Each tool covers a meaningful operation without unnecessary redundancy or bloat.
The tool surface provides full CRUD coverage for agents and read-only access to call logs, which is exactly what the domain requires. There are no obvious missing operations that would cause agent workflows to fail.
Maintenance
Related MCP Connectors
Lets Claude, Cursor or any MCP client build voice agents and place real phone calls.
Build and manage voice AI agents for real phone lines: assistants, calls, numbers and campaigns.
Create and manage voice agents, discover voices, review conversations, and search Phonic docs.
AI voice agents on real phone numbers: build agents, place and receive calls, read call summaries.
Related MCP Servers
- AlicenseAqualityDmaintenanceManage voice AI agents from Claude Code, Cursor, VS Code, or any MCP-compatible assistant.33MIT
- AlicenseAqualityDmaintenanceManage voice AI agents, make calls, run campaigns, and control phone numbers through natural language.5827 npm1MIT
- AlicenseNot gradedqualityDmaintenanceEnables Claude to launch and manage Cursor Cloud Agents on GitHub repos, including polling status, sending follow-ups, and reading conversations/PRs.MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to access local voice synthesis, zero-shot cloning, and voice catalog tools through native MCP tool calls for applications like Claude and Cursor.AGPL 3.0