# CERN Open Data MCP for AI Agents AI Agent Connect > CERN Open Data is an Connector that gives your AI agent direct access to over 66,000 particle physics datasets from the Large Hadron Collider and LEP experiments. Use it to search for research records, retrieve metadata, find analysis software, and look up technical physics terms from the official CERN glossary. ## Overview - **Category:** the-unthinkable - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_6c8vQb5D0ezH2BgoO0uep3ryrQC6NBo9IIKIfkWa/ai-agent-connect - **Tags:** particle-physics, open-data, research-datasets, large-hadron-collider, scientific-data ## Description Stop manually digging through the CERN Open Data Portal's web interface to find the specific research you need. This Connector lets you talk to your agent to pull data directly into your workspace. You can ask for specific collision types, search for Higgs boson records, or find the exact software needed to reproduce a published result. It handles the heavy lifting of navigating complex categories like Exotica or heavy-ion collisions, giving you clean metadata, file URIs, and abstracts in seconds. If you're looking for the technical definitions of terms like luminosity or b-tagging, you can pull those from the official glossary too. Vinkius makes it easy to plug this into your existing setup so you can focus on the physics instead of the filing. It's a direct line to the world's largest repository of open particle physics data, turning your agent into a specialized research assistant. ## Tools ### get_portal_statistics Get a high-level overview of all datasets and record types on the portal. It shows the scope of available data. ### list_categories See all research categories like Higgs physics or Exotica. It shows the total dataset counts for each topic. ### list_experiments View the list of experiments and their total dataset counts. Use this to see which collaborations have the most data. ### list_record_files Get the file names, sizes, and URIs for a specific dataset. This helps you see what's inside a record before downloading. ### search_by_category Filter the repository by specific research topics and subcategories. It lets you drill down into specific physics areas. ### search_by_collision_energy Find data based on specific TeV or GeV energy levels. This is a primary way to scope LHC Run data. ### search_datasets Run a full-text search across all records with multiple filters. This is your primary tool for finding specific data. ### search_software Locate analysis frameworks and reconstruction tools. Use this to find software like CMSSW for your research. ### search_supplementaries Find technical configuration files needed to reproduce results. These are essential for extending published analyses. ### check_cern_opendata_status Verify that the connection to the InvenioRDM framework is active. This ensures your queries will work correctly. ### get_glossary Retrieve definitions for physics terms like luminosity or transverse momentum. This is great for accurate science communication. ### get_record_by_doi Find a specific dataset using its official DOI link. This is useful when you have a reference from a publication. ### get_record Pull full metadata, abstracts, and file summaries for a specific record. Use this to see the details of a search result. ### search_by_collision_type Filter results by collision types like proton-proton or lead-lead. Use this to focus on specific topologies. ### search_by_experiment Scope your search to a specific collaboration like CMS or ATLAS. This is the fastest way to filter by organization. ### search_documentation Find guides and policies for using specific datasets. This helps you understand the data processing workflows. ## Prompt Examples **Prompt:** ``` I need to find some datasets from the CMS experiment that focus on dark matter. ``` **Response:** ``` I found several dark matter datasets from the CMS experiment. These records include: - **Simulated Dark Matter Signal** (NanoAODSIM format) - **Missing Transverse Energy Analyses** - **Mono-jet and Mono-photon samples** Would you like me to pull the full metadata or file list for any of these? ``` **Prompt:** ``` What does the term 'transverse momentum' mean in the context of particle physics? ``` **Response:** ``` According to the CERN physics glossary, transverse momentum (pT) is a measure of the momentum of a particle perpendicular to the beam axis. It's a key parameter used to identify high-energy particles and study their decay patterns. ``` **Prompt:** ``` Are there any software tools available for reconstructing Higgs boson events? ``` **Response:** ``` I found several analysis frameworks and reconstruction tools for Higgs physics. Notable examples include: - **CMSSW** (CMS Software) - **Analysis Frameworks** for Higgs searches - **Validation Tools** for reconstruction I can provide more specific links or documentation for any of these. ``` ## Capabilities ### Filter datasets by experiment or collision type Search the repository using specific filters like CMS, ATLAS, or proton-proton collisions. ### Pull full metadata and file lists Get abstracts, author ORCIDs, and direct ROOT/EOS URIs for any specific record. ### Resolve publication DOIs Convert a DOI from a research paper into a direct link to its corresponding open dataset. ### Search the official physics glossary Retrieve precise definitions for technical terms like transverse momentum or pseudorapidity. ### Browse analysis software and guides Find reconstruction frameworks, analysis tools, and documentation for reproducing results. ### Get portal-wide statistics View the total distribution of datasets across different energies, years, and formats. ## Use Cases ### Finding specific collision data A physicist needs to find CMS data for 13 TeV proton-proton collisions. They ask the agent to search by collision energy and experiment to get a list of relevant records. ### Defining technical terms A student wants to know what pseudorapidity means. They ask the agent to check the glossary for the definition and examples. ### ML dataset discovery A data scientist is looking for dark matter datasets. They use the agent to search by category and find specific files for anomaly detection training. ### Reproducing research A researcher wants to reproduce a paper. They ask the agent to find the supplementary materials and documentation for a specific DOI. ## Benefits - Skip the manual web navigation by using search_datasets to find specific physics records instantly. - Get precise metadata including ORCID authors and DOI links using get_record to save time on bibliography work. - Access the official physics glossary with get_glossary to ensure your reports use correct technical definitions. - Identify the right software for your analysis by using search_software to find frameworks like CMSSW. - Quickly find the exact files you need for your ML model by listing URIs with list_record_files. - Understand the scale of the repository using get_portal_statistics to see how many datasets exist per energy level. ## How It Works The bottom line is you get direct access to over 66,000 datasets without leaving your chat interface. 1. Connect the CERN Open Data MCP to your AI client through Vinkius. 2. Ask your agent to search for specific physics categories, experiments, or collision energies. 3. Receive a list of records with metadata and direct links to the data files. ## Frequently Asked Questions **How can I find Higgs boson data with CERN Open Data?** You can ask your agent to search by category or experiment. It will scan the 66,000+ available records and return a list of relevant datasets, including metadata like event counts and file sizes. **Does CERN Open Data help me find research software?** Yes, you can search for specific analysis frameworks, reconstruction tools, and validation software. It provides titles, descriptions, and direct links to the software used in published results. **Can CERN Open Data include the physics glossary?** Yes, it includes over 1,000 glossary entries. Your agent can pull definitions for technical terms like luminosity, b-tagging, and various detector components. **How do I get the files for a specific DOI?** You can simply provide the DOI to your agent. It will resolve the DOI to the correct CERN Open Data record and provide you with the file distribution summary and URIs. **Can I see how many datasets are in the CMS experiment?** Yes, your agent can pull portal statistics to show you the total count of datasets for CMS, along with other experiments like ATLAS and ALICE. **Is there a way to find data for specific collision energies?** You can filter searches by specific energy levels, such as 13 TeV or 7 TeV. This helps you quickly isolate data from specific LHC runs. **Do I need an API key to use this server?** No. The CERN Open Data Portal API is completely public and requires no authentication. Simply subscribe to this server and enter any placeholder value in the API key field to start querying particle physics datasets immediately. **What kind of data can I access from CERN?** You can access over 66,000 datasets from major LHC experiments (CMS, ATLAS, ALICE, LHCb) and legacy experiments (DELPHI, OPERA). This includes real collision data, Monte Carlo simulations, derived datasets, analysis software, physics glossary entries, and detailed documentation. Data covers Higgs boson searches, Dark Matter studies, exotic particle searches, heavy-ion physics, and more. **Can I use CERN data for machine learning projects?** Absolutely. CERN provides labeled datasets specifically designed for ML applications, including particle identification, jet classification, event reconstruction, and anomaly detection. Use the search tools with queries like 'machine learning' or filter by file type 'csv' or 'nanoaodsim' to find ML-ready formats. The CMS experiment alone has published thousands of simulated datasets with known physics labels.