@vora/voice — Voice for Autonomous AI Agents
Make phone calls, close deals, book meetings. Full-stack voice AI: brain + voice + ears.
Install
npx @vora/voiceOr add to your MCP config:
{
"mcpServers": {
"vora-voice": {
"command": "npx",
"args": ["@vora/voice"]
}
}
}Related MCP server: Agent Negotiation MCP
Tools
Tool | What it does |
| Create account. Vora interviews you about your business (3-5 turns). |
| Build a persistent voice agent for a specific use case. |
| Make an outbound phone call. Returns structured outcome. |
| Check results, history, analytics, AI recommendations. |
| Improve your agent based on call learnings. |
Quick Start
1. vora_register → answer questions → get account_id
2. vora_create_agent → get agent_id
3. vora_call → make calls → get outcomes
4. vora_calls → check analytics
5. vora_update_agent → auto-improvePayment
x402 (USDC on Base): Zero-friction. Set
VORA_X402_ENABLED=true.API Key: Set
VORA_API_KEY=your_key. Get key from vora_register.
Pricing
What | Cost |
Onboarding | Free |
Agent creation | Free |
Outbound calls | $0.06-0.08/min |
Phone numbers | $2/mo |
Environment Variables
VORA_API_URL=https://agent.voicevora.com # API endpoint (default)
VORA_API_KEY=vora_ak_... # API key auth
VORA_X402_ENABLED=true # Enable x402 USDC paymentsLanguages
English, Arabic (all dialects), Hindi, Bengali, Tamil, Telugu, Vietnamese, Turkish, Russian, Serbian, Chinese, Japanese, Korean, Spanish, French, German, Portuguese, and 15+ more.
License
MIT
Available Tools
5 toolsvora_callA
Make an outbound phone call using your voice agent. Your agent handles the entire conversation autonomously — greeting, pitch, objection handling, qualification, closing — and returns a structured outcome.
Calls typically take 1-5 minutes. You can poll vora_calls for the result or provide a callback_url for async delivery.
The more lead_context you provide, the better the conversation. Tell Vora everything you know about who it's calling.
Inputs: agent_id (required), phone (required), lead_context, specific_objective, +1 more.
vora_callsA
Query your call history, results, and analytics. Use to:
Check the result of a specific call (by call_id)
View recent call history for an agent
Get aggregate analytics with conversion rates, top objections, and AI recommendations
The analytics include AI-generated recommendations for improving your calls.
Inputs: call_id, agent_id, status, last, +1 more.
vora_create_agentA
Create a persistent voice agent for a specific use case. Uses your account context (from vora_register) plus any additional details you provide.
The voice agent persists across calls and improves with every conversation it has. Vora automatically:
Compiles product knowledge from your website and registration context
Generates objection handling based on your industry
Selects the optimal voice, language, and speaking style
Builds a conversation workflow (cold call, appointment, follow-up, etc.)
You can create multiple voice agents under one account for different use cases.
Inputs: account_id (required), objective (required), workflow, language, +1 more.
vora_registerA
Create your Vora account through conversational onboarding. Vora asks you questions about your business, products, target customers, and objectives to build a rich profile. This produces dramatically better voice agents than one-shot configuration.
Multi-turn: call this tool, answer the returned questions, call again with your answers. Repeat until status is "complete" (typically 3-5 turns).
On completion you receive an account_id and API key to use with all other Vora tools.
Inputs: session_id, name, url, description, +1 more.
vora_update_agentA
Incrementally improve your voice agent without full recompilation. Use this to:
Add new objection responses based on what you're hearing in calls
Update pricing or product information
Apply AI-recommended improvements from call analytics (apply_learnings)
Refine your objective based on what's working
Force a full recompile when your website has significantly changed
Changes take effect on the very next call. The apply_learnings option is powerful — Vora analyzes all your past calls and automatically updates objection handling, greetings, and qualification criteria based on what's actually converting.
Inputs: agent_id (required), add_objection_response, update_pricing, update_context, +3 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 5 tools
Each tool has a clearly distinct purpose: calling, querying calls, creating an agent, registering an account, and updating an agent. There is no overlap or ambiguity.
All tools use the 'vora_' prefix followed by a verb or verb_noun pattern (call, calls, create_agent, register, update_agent). The naming is consistent and predictable.
With 5 tools, the set is well-scoped for a voice agent platform. Each tool covers a distinct area without being overly numerous or sparse.
Core CRUD operations are covered (create via vora_create_agent, update via vora_update_agent, read via vora_calls, use via vora_call). However, there is no delete tool for agents or calls, and no tool to list all agents.
Maintenance
Related MCP Connectors
LLM Orchestration MCP Agent
AI voice agents on SMB websites — fully autonomous build in 2–3 min. 23 MCP tools. EU, GDPR.
Voice and chat for AI agents — Discord, Teams, Meet, Slack, Zoom, Telegram, WhatsApp, NC Talk, SIP
MCP-Native LLM Orchestration Agent
Related MCP Servers
AlicenseCqualityDmaintenanceBest VoiceAI White Label Platform for Agencies with Full MCP Connectivity to Any Platform for Custom VoiceAi & AI Chat Integrations and You Keep 100% of the Profits32ISC- AlicenseNot gradedqualityCmaintenanceAgent Negotiation - MCP server providing AI-powered tools and automation by MEOK AI Labs7 npm169 PyPIMIT
- AlicenseNot gradedqualityCmaintenanceAgent Delegation - MCP server providing AI-powered tools and automation by MEOK AI Labs6 npm148 PyPIMIT
- AlicenseNot gradedqualityBmaintenanceLets Claude, Cursor, ChatGPT or any MCP client build AI voice agents, place calls and read transcripts, without writing HTTP code.1,149 npmMIT