How to Use the CERN Open Data MCP in AutoGen

Q: How can an AutoGen team decide which CERN Open Data experiment is best for a query?

You can have a 'Planner' agent use listexperiments to see the options. Then, a 'CMSSpecialist' and an 'ATLASSpecialist' agent can each use searchbyexperiment and argue for the relevance of their respective datasets based on the results.

Q: Can I use AutoGen to have one agent find CERN Open Data and another explain it?

Yes, that's a perfect use case. A 'Searcher' agent can use searchdatasets to find a record, passing the ID to an 'Explainer' agent. The Explainer then uses getrecord and getglossary to break down the abstract and its technical terms.

Deploy AutoGen agent teams that debate and decide how to best explore CERN's open research data.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect CERN Open Data MCP to AutoGen

Create your Vinkius account to connect CERN Open Data 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 CERN Open Data with AutoGen

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Let Agents Debate the Best Search Strategy

An AutoGen setup isn't one agent following steps; it's a conversation. A 'Researcher' agent might propose using `search_datasets` with a broad query like 'exotic particles'. A 'Critic' agent could then challenge that, suggesting it's better to start with `list_categories` to see what's actually available first. This back-and-forth leads to a better plan. The agents might agree to use `search_by_category` for 'Exotica' and then refine the results. They converge on an efficient path through the data by challenging each other's use of the MCP tools.

Have Agents Vet Data Quality for You

Once a dataset is found, the work begins for your agent team. One agent uses `get_record` to pull the metadata. Another agent, a 'Data_Validator', could then use `list_record_files` to check if the file count and formats match expectations for an analysis. A 'Documentation_Specialist' agent could simultaneously run `search_documentation` to find the 'how-to' guide for that specific dataset. The final recommendation you get is a consensus decision from multiple specialists that have each examined a different piece of the puzzle.

AutoGen Agents Decide on Relevant Data

What if you ask for 'the most important datasets for Higgs boson research?' That's subjective. An AutoGen team can tackle it. One agent might use `search_by_collision_energy` to argue for 13 TeV data, while another uses `search_datasets` to find highly cited work. A third agent could use `get_portal_statistics` from this MCP server to add context about which data is most plentiful. They debate the meaning of 'important'—is it collision energy, citation count, or data volume? The final answer is a reasoned conclusion, not just a list of search results.

Setup guide

Set up CERN Open Data 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 CERN Open Data 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="CERN Open Data_assistant",
    model_client=OpenAIChatCompletionClient(model="gpt-4o"),
    tools=tools,
)

result = await agent.run("List recent CERN Open Data 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="CERN Open Data_assistant",
        model_client=OpenAIChatCompletionClient(model="gpt-4o"),
        workbench=workbench,
    )

    result = await agent.run("List recent CERN Open Data 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 CERN Open Data. 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 CERN Open Data 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 CERN Open Data MCP in AutoGen

You can have a 'Planner' agent use `list_experiments` to see the options. Then, a 'CMS_Specialist' and an 'ATLAS_Specialist' agent can each use `search_by_experiment` and argue for the relevance of their respective datasets based on the results.

Yes, that's a perfect use case. A 'Searcher' agent can use `search_datasets` to find a record, passing the ID to an 'Explainer' agent. The Explainer then uses `get_record` and `get_glossary` to break down the abstract and its technical terms.

A 'UserProxyAgent' would state the need. An 'Engineer' agent would use the `search_software` tool with keywords like 'framework' or 'CMSSW'. A 'QA_Agent' could then take the results and use `search_documentation` to find guides for that software, confirming it's well-supported before reporting back.

The first step in any agent conversation should be a sanity check. Have one of your agents call `check_cern_opendata_status` to confirm the API is up before the other agents start planning complex queries.

Yes. Each tool call from your agents, whether for searching datasets or fetching record metadata, is routed through a stateless, ephemeral Vinkius endpoint. Your authentication token authorizes the request, but the content of your queries is not logged or retained.

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