Agent skill
mlip-guide
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/mlip-guide
SKILL.md
Machine Learning Interatomic Potentials (MLIPs)
Universal and specialized machine learning interatomic potentials for rapid atomistic simulation without DFT. Covers model selection, usage patterns, validation strategies, and known limitations.
Sub-Skills
| Sub-Skill | Directory | Description |
|---|---|---|
| Universal MLIPs | universal-mlip/ |
MACE-MP-0, CHGNet, M3GNet/MatGL, SevenNet-0: setup, usage, benchmarking, and validation against DFT |
| TorchSim Batch GPU | torchsim-batch/ |
GPU-accelerated batch MD and optimization with TorchSim: 10-100x speedup, auto-batching, parallel relaxation and screening |
Method Decision Guide
What do you need MLIPs for?
Quick geometry relaxation / screening many structures?
--> universal-mlip/ (MACE-MP-0 medium is fastest, CHGNet is a good alternative)
--> torchsim-batch/ (GPU available? 10-100x faster batch relaxation with TorchSim)
Molecular dynamics (phonons, thermal, diffusion)?
--> universal-mlip/ (MACE-MP-0 large for best accuracy)
--> torchsim-batch/ (GPU available? TorchSim for 10-100x faster MD on GPU)
Elastic constants / equation of state?
--> universal-mlip/ (any universal MLIP; validate against DFT for novel systems)
Band gaps / electronic properties / magnetic ordering?
--> MLIPs CANNOT predict these. Use Quantum ESPRESSO DFT instead.
System contains rare elements / extreme conditions?
--> Validate MLIP against DFT first; MLIPs may extrapolate poorly.
Pre-installed vs. Installable
| MLIP | Status | Install Command |
|---|---|---|
| MACE-MP-0 | Pre-installed | -- |
| CHGNet | pip install | pip install chgnet |
| M3GNet (MatGL) | pip install | pip install matgl |
| SevenNet | pip install | pip install sevenn |
| ORB Models | pip install | pip install orb-models |
| TorchSim | pip install | pip install torch-sim (requires PyTorch + CUDA for GPU) |
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
majiayu000/claude-skill-registry
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
majiayu000/claude-skill-registry
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
majiayu000/claude-skill-registry
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
majiayu000/claude-skill-registry
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
majiayu000/claude-skill-registry
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
majiayu000/claude-skill-registry
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?