How to Use the Databricks MCP in AutoGen

Q: How do I configure an AutoGen agent to use the Databricks listclusters tool?

You use the mcpservertools function with your Vinkius URL to get a list of available tools. Then, you pass that tool list into the AssistantAgent constructor when you create your agent.

Q: Can I create two AutoGen agents that discuss the output of a Databricks job run?

Absolutely. You can have one agent call listjobruns and post the result into the group chat. The other agent can then analyze that output and continue the conversation, which is a core AutoGen pattern.

Q: What's the best way to give my AutoGen agent group access to my Databricks account?

You don't give them direct access. You provide the mcpservertools function with your single Vinkius endpoint URL. The Vinkius platform securely manages the connection to Databricks on your behalf.

Build teams of AutoGen agents that debate and manage your Databricks resources collaboratively.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Databricks MCP to AutoGen

Create your Vinkius account to connect Databricks to AutoGen 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 Databricks with AutoGen

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Have Agents Debate Databricks Cluster Usage

Create a team of AutoGen agents to watch your compute. A "Monitor" agent can use `list_clusters` to get the current state, while an "Analyst" agent uses `list_job_runs` to see what's actually running. They don't just report facts, they talk. The Monitor might state, "Cluster 'job-cluster-01' is at 95% utilization." The Analyst can then reply, "That's because the 'backfill_task' is running. Let's flag it for review." This conversational approach helps your agents find the root cause of issues together.

Use AutoGen Agents to Audit Unity Catalog

Set up a "Governance" agent that periodically calls `list_catalogs` and `list_schemas`. You can pair it with a "Security" agent that reviews the output based on a set of rules you define. When the Governance agent finds a new schema, it starts a conversation. "I found a new schema named 'test_data'. Does this follow our policy?" The Security agent can then check and respond, creating an automated, conversational audit log of your Databricks environment.

Let AutoGen Agents Oversee Databricks Jobs

An "Operator" agent can use `list_jobs` and `list_job_runs` to track job health. A "Finance" agent, armed with the `list_warehouses` tool, can monitor the associated compute costs. This is all possible with one MCP Server. When the Operator reports, "The 'daily_aggregation' job has failed again," the Finance agent can add context. "It's running on our most expensive SQL warehouse. We should escalate this." The team uses the tools from this server to inform their collaborative decisions.

Setup guide

Set up Databricks MCP in AutoGen

Prerequisites

Python 3.10+ installed
autogen-ext[mcp] package
Active Vinkius subscription with a valid endpoint token

1

Install AutoGen with MCP
Run pip install "autogen-ext[mcp]" autogen-agentchat. The MCP extension includes mcp_server_tools for stateless tool access.
2

Fetch tools from the MCP
Call mcp_server_tools(SseServerParams(url=...)) with your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Run your agent
Pass the tools to AssistantAgent and call agent.run(). The agent invokes Databricks tools and returns structured results.

agent.py

from autogen_ext.tools.mcp import SseServerParams, mcp_server_tools
from autogen_agentchat.agents import AssistantAgent
from autogen_ext.models.openai import OpenAIChatCompletionClient

server_params = SseServerParams(
    url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
)

tools = await mcp_server_tools(server_params)

agent = AssistantAgent(
    name="Databricks_assistant",
    model_client=OpenAIChatCompletionClient(model="gpt-4o"),
    tools=tools,
)

result = await agent.run("List recent Databricks data")
print(result.messages[-1].content)

Get your connection token →

Prerequisites

Python 3.10+ installed
autogen-ext[mcp] + autogen-agentchat
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Same packages as above. McpWorkbench is ideal when your agent needs stateful sessions across multiple tool calls.
2

Use McpWorkbench as context manager
Wrap your agent in async with McpWorkbench(...) to maintain shared state and resources. The workbench manages the full MCP session lifecycle.
3

Run with workbench
Pass workbench=workbench to your agent. State is preserved across multiple tool calls within the same session.

agent.py

from autogen_ext.tools.mcp import McpWorkbench, SseServerParams
from autogen_agentchat.agents import AssistantAgent
from autogen_ext.models.openai import OpenAIChatCompletionClient

server_params = SseServerParams(
    url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
)

async with McpWorkbench(server_params) as workbench:
    agent = AssistantAgent(
        name="Databricks_assistant",
        model_client=OpenAIChatCompletionClient(model="gpt-4o"),
        workbench=workbench,
    )

    result = await agent.run("List recent Databricks data")
    print(result.messages[-1].content)

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 Databricks. 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 Databricks 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 Databricks MCP in AutoGen

You use the `mcp_server_tools` function with your Vinkius URL to get a list of available tools. Then, you pass that tool list into the `AssistantAgent` constructor when you create your agent.

Absolutely. You can have one agent call `list_job_runs` and post the result into the group chat. The other agent can then analyze that output and continue the conversation, which is a core AutoGen pattern.

You don't give them direct access. You provide the `mcp_server_tools` function with your single Vinkius endpoint URL. The Vinkius platform securely manages the connection to Databricks on your behalf.

The LLM backing the agent makes the decision based on the conversation history and the user's prompt. AutoGen's framework is designed to have the agent select the most appropriate tool for the task at hand.

The agents access specific metadata you allow, such as cluster status, job run logs, and user information via `get_me`. Vinkius isolates every single tool call in a dedicated V8 Isolate sandbox which is destroyed immediately after the call completes, ensuring no data ever persists.

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