How to Use the Karbon MCP in OpenAI Agents SDK

Q: How do I connect Karbon to the OpenAI Agents SDK?

Install openai-agents via pip and configure an MCPServerStreamableHttp instance. Pass that server to your Agent constructor's mcpservers list. The SDK automatically discovers all twelve tools.

Q: Can my OpenAI agent read specific Karbon contact notes?

Yes. The agent can execute listcontactnotes to pull specific client history. You can then use that context to inform how the agent interacts with other tools.

Q: Does the Karbon MCP Server support tool caching in OpenAI Agents SDK?

It does. You should set cacheToolsList=True when initializing the server connection. This prevents the agent from re-fetching the schema for tools like listworktasks on every run.

Q: What is the best way to update engagements using this integration?

Have your agent query current details with getworkitem first. Once it understands the current state, it can fire updateworkitem to modify the project parameters safely.

Build production-grade accounting agents with OpenAI Agents SDK that manage firm workflows and contacts.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Karbon MCP to OpenAI Agents SDK

Create your Vinkius account to connect Karbon 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 Karbon with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Automate firm tasks with OpenAI Agents SDK

Connect your production OpenAI agents directly to your accounting firm's workflow using this MCP Server. You configure the guardrails in Python, and the agent auto-discovers tools like `create_work_item` to spin up new engagements without manual entry. Handoffs between specialized agents become trivial. A client intake agent can run `create_contact` and `list_contact_notes`, then pass the context to a billing agent that uses `update_work_item` to adjust project scope based on those notes.

Audit accounting activity via OpenAI tracing

Every time your agent calls `list_work_items` or `get_work_item`, the OpenAI dashboard logs the exact payload. You see exactly what context the agent pulled before it decided to act on an engagement. This matters when dealing with sensitive firm data. If an agent tries to execute `list_users` to assign a task, you can trace the entire decision tree to ensure it followed your strict internal assignment rules.

Build safe workflow loops

Prototyping is easy, but deploying agents that touch client records requires built-in safety constraints. You can restrict which tools the agent accesses, ensuring it only reads data via `get_contact` or `list_organizations` rather than modifying it. When you do allow writes, the agent validates the required fields. `create_work_item` requires a title and work type, and the SDK ensures your agent provides both before the HTTP request even fires.

Setup guide

Set up Karbon 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 Karbon tools at runtime.
3

Create your Agent

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

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate Karbon 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="Karbon Agent",
            instructions="You have access to Karbon 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 Karbon. 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 Karbon 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 Karbon MCP in OpenAI Agents SDK

Install `openai-agents` via pip and configure an `MCPServerStreamableHttp` instance. Pass that server to your Agent constructor's `mcp_servers` list. The SDK automatically discovers all twelve tools.

Yes. The agent can execute `list_contact_notes` to pull specific client history. You can then use that context to inform how the agent interacts with other tools.

It does. You should set `cacheToolsList=True` when initializing the server connection. This prevents the agent from re-fetching the schema for tools like `list_work_tasks` on every run.

Have your agent query current details with `get_work_item` first. Once it understands the current state, it can fire `update_work_item` to modify the project parameters safely.

Your firm's client names, work items, and internal user lists remain strictly within your environment. Vinkius runs this connection in a V8 Isolate Sandbox, meaning the data your agent pulls via `list_people` is destroyed the moment the session ends.

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