Query ATLAS Open Data Monte Carlo samples using the atlasopenmagic MCP tools. Load this skill specifically for ATLAS DSIDs, `physics_short` names, cross-sections, k-factors, filter efficiencies, MC weights, or file URLs tied to a named ATLAS release (e.g. 2024r-pp, 2025r-evgen-13tev). For CMS / LHCb / ALICE / OPERA records, or portal-wide queries by recid / DOI / title, load the `cern-opendata` skill instead.
Use this skill only when the user is clearly working with ATLAS Monte
Carlo samples or an ATLAS Open Data release. For non-ATLAS experiments,
portal-wide record searches, or records identified by recid / DOI /
title, use the sibling cern-opendata skill (backed by the
cernopendata MCP). The two servers cooperate — you can start from a
portal record via cod_* tools and drop into ATLAS-specific metadata
with atlas_* tools when the analysis calls for it.
The atlasopenmagic MCP server provides these tools (prefixed with atlasopenmagic_):
atlas_available_releases — List all available data releasesatlas_get_current_release — Show the currently active releaseatlas_set_release — Switch to a different releaseatlas_available_datasets — List all datasets in the current releaseatlas_available_skims — List available skim typesatlas_available_keywords — List physics keywords for filteringatlas_get_metadata — Get metadata for a specific dataset (by DSID or physics_short)atlas_get_metadata_fields — List all available metadata fieldsatlas_get_all_info — Get comprehensive info for a datasetatlas_match_metadata — Search datasets by metadata field values or keywordsatlas_get_urls — Get ROOT file URLs for a dataset (protocols: root, https, eos)atlas_get_weights — Get MC weight metadata for a datasetatlas_get_all_weights_for_release — Get weight metadata for all datasets in a releaseatlas_get_current_release or atlas_set_releaseatlas_match_metadata (search by keywords, process, generator) or atlas_available_datasetsatlas_get_metadata or atlas_get_all_infoatlas_get_urlsatlas_get_weights| Release | Description |
|---|---|
| 2016e-8tev | 2016 education release, 8 TeV pp collisions |
| 2020e-13tev | 2020 education release, 13 TeV pp collisions |
| 2024r-pp | 2024 research release, proton-proton collisions |
| 2024r-hi | 2024 research release, heavy-ion collisions |
| 2025e-13tev-beta | 2025 education beta release, 13 TeV pp collisions |
| 2025r-evgen-13tev | 2025 research event-generation release, 13 TeV |
| 2025r-evgen-13p6tev | 2025 research event-generation release, 13.6 TeV |
Each Monte Carlo sample is identified by a numeric dataset number (DSID), e.g. 301204, and a human-readable physics_short name, e.g. Sh_2211_Zee_maxHTpTV2_BFilter.
The physics_short packs generator, tune/PDF, process, and filters into a structured name. Parts are separated by underscores.
Part 1 — Generator abbreviations (always first):
| Abbrev | Generator | Abbrev | Generator |
|---|---|---|---|
| Sh | Sherpa | H7 | Herwig7 |
| Ph | Powheg | Ag | Alpgen |
| Py8 | Pythia8 | EG | EvtGen |
| MG | MadGraph (LO) | PG | ParticleGun |
| aMC | aMC@NLO | HepMC | HepMC files |
Part 2 — Tune / PDF / Sherpa version:
222 = 2.2.2, 2211 = 2.2.11, 2212 = 2.2.12Remaining parts — Process and filters:
Process abbreviations:
tchan/schan = t/s-channelZee = Z->ee, Zmumu = Z->mumu, Wenu = W->enuincl = inclusive, dil = di-lepton, nonallhad = at least one lepton, allhad = all-hadronicProduction features:
FxFx = FxFx merging, DS/DR = diagram subtraction/removalFilters (usually at the end):
BFilter = b-quark filter, BVetoCFilter = no b, has c, BVetoCVeto = no b or cmaxHTpTV2 = HT and pT(V) filter, MET200 = 200 GeV MET filterExample: Sh_2211_Zee_maxHTpTV2_BFilter = Sherpa 2.2.11, Z->ee, maxHTpTV2 filter, b-quark filter
Pre-filtered subsets: exactly4lep, 3lep, etc. Use noskim for the full dataset.
Core: dataset_number, physics_short, e_tag
Physics: cross_section_pb, genFiltEff, kFactor, nEvents, sumOfWeights, sumOfWeightsSquared
Generation: process, generator, keywords, description, GenTune, PDF, CoMEnergy
Files: file_list, skims (each with skim_type and file_list)
root — XRootD streaming (default, best for analysis)https — Web-accessible via opendata.cern.cheos — EOS POSIX mount pathFor MC normalisation:
weight = cross_section_pb * 1000 * kFactor * genFiltEff * mcWeight / sumOfWeights * luminosity_fb