How to Use the Inworld AI MCP in OpenAI Agents SDK

Q: Can my OpenAI Agents SDK agent update Inworld AI routing rules dynamically?

Yes, your agent uses updaterouter and createrouter to modify character paths during a live session. The OpenAI tracing dashboard logs these state changes automatically so you can debug the transitions.

Q: How does voice generation work with Inworld AI and OpenAI Agents SDK?

Your agent calls synthesizespeechstream to stream audio bytes directly to the client. This bypasses the need for local speech generation models, saving you CPU cycles on your hosting environment.

Give your OpenAI Agents SDK production-grade speech synthesis and character routing with zero-config MCP tool discovery.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Inworld AI MCP to OpenAI Agents SDK

Create your Vinkius account to connect Inworld AI to OpenAI Agents SDK and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Inworld AI with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Real-time voice cloning in OpenAI Agents SDK

The `clone_voice` tool immediately establishes custom voice profiles directly from your agentic workflows. Your OpenAI Agents SDK pipeline sends raw audio samples to the server, creating a cloned voice that is ready for instant playback. To generate the actual speech, your agent calls `synthesize_speech_stream` or `synthesize_speech_sync` to output audio files. The OpenAI dashboard traces these calls automatically, ensuring you can monitor latency and token usage for every generated voice file.

Dynamic character routing via MCP Server

The `create_router` tool lets your agent set up LLM routers that direct user queries to specific character personalities. Your agent can inspect existing configurations with `get_router` and update rules on the fly with `update_router`. This MCP Server integration allows complex multi-agent handoffs within the OpenAI framework. The agent evaluates the user's intent and dynamically selects the correct character routing path without manual code changes.

Direct WebRTC streaming for low-latency calls

The `create_realtime_call` tool opens a direct WebRTC connection between your user and the Inworld platform. Your agent initiates this call to bypass standard HTTP polling, bringing voice latency down to sub-200ms ranges. While the call is active, the agent uses `transcribe_audio` to convert incoming speech to text on the fly. This keeps the conversational state synchronized perfectly with your backend systems.

Setup guide

Set up Inworld AI MCP in OpenAI Agents SDK

Prerequisites

Python 3.10+ installed
openai-agents package (pip install openai-agents)
Active Vinkius subscription with a valid endpoint token

1

Install the SDK

Run pip install openai-agents to install the OpenAI Agents SDK. The MCP integration is built-in — no extra dependencies needed.
2

Connect via SSE transport

Use MCPServerSse with your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. The SDK auto-discovers all Inworld AI tools at runtime.
3

Create your Agent

Pass the MCP to Agent(mcp_servers=[server]). The agent receives Inworld AI tools as native definitions — JSON schemas resolve automatically.
4

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate Inworld AI tools and returns structured results. Copy the full example on the right to get started.

agent.py

import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerSse

async def main():
    async with MCPServerSse(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    ) as server:
        agent = Agent(
            name="Inworld AI Agent",
            instructions="You have access to Inworld AI tools.",
            mcp_servers=[server],
        )
        result = await Runner.run(agent, "List recent transactions")
        print(result.final_output)

asyncio.run(main())

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Inworld AI. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Inworld AI now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Inworld AI MCP in OpenAI Agents SDK

Install openai-agents, set up MCPServerStreamableHttp with the Vinkius endpoint, and pass it to your Agent constructor. The SDK auto-discovers all 19 tools immediately. You can set cacheToolsList to true to keep things fast.

Yes, your agent uses `update_router` and `create_router` to modify character paths during a live session. The OpenAI tracing dashboard logs these state changes automatically so you can debug the transitions.

Your agent calls `synthesize_speech_stream` to stream audio bytes directly to the client. This bypasses the need for local speech generation models, saving you CPU cycles on your hosting environment.

You can restrict tool access within the Agent constructor configuration. However, the default setup registers all 19 Inworld endpoints for maximum agent flexibility out of the box.

The server processes your raw audio samples and voice profiles in an ephemeral, zero-trust V8 sandbox on Vinkius. Your voice data is never cached or used to train public models, keeping your custom assets secure.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript