harnessing
in
your agentic IDEs
A composable framework for AI-driven atomistic research. Chain modular skills across MLIPs, DFT, drug discovery, and generative AI — in any coding copilot.
Strictly typed, battle-tested operations exposed via MCP servers. Fixed signatures for relax, MD, literature search, DFT I/O, and more.
Mid-level tutorials combining tool calls to solve focused research problems. Each ships with SKILL.md, helper scripts, and runnable examples.
High-level, publication-scope research campaigns chaining multiple skills. From computational materials science workflows to building atomistic ML models.
Click any skill to view its full documentation and example runs.
Pre-built pipelines for complex multi-stage research objectives.
Each environment is a uv project under venv/, created on first use by
venv/run (or run from a container image where the host cannot), so dependency conflicts stay
apart. Servers marked MCP expose tools directly to your AI agent; the others are
used through skill scripts.
MCP Server Environments
| MCP Server | Tools | Environment | Core Libraries | Purpose |
|---|---|---|---|---|
| base | cpu | pymatgen, ase, rdkit, packmol | MP queries, VASP I/O, structure tools | |
| atomate2 | cpu | atomate2, jobflow-remote | Remote DFT workflows via NERSC/HPC | |
| drugdisc | cpu | rdkit, autodock-vina, meeko | Docking, ADMET, fingerprints | |
| smol | cpu | smol | Cluster expansion + Monte Carlo | |
| mace | mlip | mace-torch | MACE models (MP, OMAT, MATPES, MH) | |
| matgl | mlip | matgl (PyTorch Geometric) | CHGNet, TensorNet | |
| fairchem | fairchem | fairchem-core | UMA, eSEN models (OMAT, OMol, OC20) | |
| mattergen | mattergen | MatterGen, PyG, lightning | Generative crystal design | |
| adit | adit | ADiT, lightning, hydra, PyG | All-atom diffusion generation | |
| diffcsp | diffcsp | DiffCSP++, hydra, PyG, pyxtal | Symmetry-constrained crystal generation |
Script Environments (no MCP server —
invoked via venv/run <env>)
| Environment | Core Libraries | Purpose |
|---|---|---|
| cpu | pycalphad, DARA, fipy, nmrsim, scine (x86_64) | CALPHAD, XRD refinement, phase field, NMR, ORCA drivers |
| msms | ms-pred (ICEBERG 2.1), DGL | LC-MS/MS spectrum prediction (x86_64) |
| reactot | React-OT, PyTorch | Transition state generation |
| scd | SelfConditionedDenoisingAtoms, PyG | Property prediction with SCD foundation models (x86_64) |
In-depth guides for developing new skills, managing environments, and data visualization.
AtomisticSkills is designed to work directly with agentic IDEs. Follow these simple steps:
git clone git@github.com:learningmatter-mit/AtomisticSkills.git && cd AtomisticSkills