New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

ML Property Predictor

by learningmatter-mit in learningmatter-mit/AtomisticSkills

Train a property predictor head on top of a Machine Learning Interatomic Potential (MLIP) backbone (MACE or MatGL) to predict custom intensive or extensive properties from crystal or molecular…

MITAuto-check passedData & Analytics

Install ML Property Predictor

skills CLI
$ npx skills add learningmatter-mit/AtomisticSkills --skill ml-property-predictor -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install learningmatter-mit/AtomisticSkills ml-property-predictor --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/learningmatter-mit/AtomisticSkills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/ml-property-predictor .claude/skills/ml-property-predictor && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
ml-property-predictor
GitHub stars
176
Token cost
~1.3k tokens
SKILL.md length
492 words
Files
10 (incl. scripts)
Skills in repo
129
Repo updated
First seen
Licence
MIT

At a glance

Train a property predictor head on top of a Machine Learning Interatomic Potential (MLIP) backbone (MACE or MatGL) to predict custom intensive or extensive properties from crystal or molecular…

  • Works in 3 steps: Prepare Data: Build a .json or .xyz… → Determine Property Type: Determine if… → Execute Script: Run the MACE or MatGL…
  • Tasks that involve Machine learning
  • SKILL.md covers Goal, Overview, Workflow and Example 1: Training a MACE…, plus 3 more sections
  • Runs Python scripts from its folder

What it does

ML Property Predictor is an agent skill from learningmatter-mit/AtomisticSkills. Train a property predictor head on top of a Machine Learning Interatomic Potential (MLIP) backbone (MACE or MatGL) to predict custom intensive or extensive properties from crystal or molecular structures.

Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 14 other files, including scripts (for example `examples/mace_bulk_modulus/README.md`, `examples/mace_bulk_modulus/run_mace.py` and `examples/matgl_bulk_modulus/README.md`).

It sits in Data & Analytics, covering Machine learning. The repository describes itself as: Integrating AtomisticSkills into Agentic IDEs (Cursor, Claude Code, Codex, Google Antigravity, Hermes Agent, etc). The licence is MIT.

When your agent uses it

  • Tasks that involve Machine learning

Example prompts

  • “/ml-property-predictor”

Requirements

  • Python 3

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. Prepare Data: Build a .json or .xyz dataset containing structures and the corresponding scalar property labels. JSON datasets should be…
  2. Determine Property Type: Determine if the property is "intensive" (e.g. Bandgap, Bulk Modulus) or "extensive" (e.g. Total Energy).
  3. Execute Script: Run the MACE or MatGL property prediction script in the mlip environment.

What it can do on your machine

Read from SKILL.md and the folder at commit 6257444. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 2 files in scripts/ (Python), which the agent can run.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Links to these hosts (documentation or services it may open):

    • doi.org
    • github.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

ML Property Predictor loads about 1.3k tokens when it runs. Until then it costs about 57 tokens; SKILL.md has 492 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~57
When it runs · the whole SKILL.md, loaded when a task matches
~1.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from learningmatter-mit/AtomisticSkills at commit 6257444, republished under its MIT licence (© learningmatter-mit). 492 words, ~1,278 tokens.

Download SKILL.mdSave it as .claude/skills/ml-property-predictor/SKILL.md (or your agent's skills folder). This skill also uses 9 other files; get the full folder from GitHub.
name
ml-property-predictor
description
Train a property predictor head on top of a Machine Learning Interatomic Potential (MLIP) backbone (MACE or MatGL) to predict custom intensive or extensive properties from crystal or molecular structures.
metadata.category
machine-learning, materials, chemistry
metadata.venv
mlip

MLIP Property Predictor Training

Goal

To leverage pre-trained GNN representations from MLIPs to train an independent readout head for any custom scalar target property (e.g., bulk modulus, bandgap, formation energy, or spin states) directly from crystal or molecular structures.

Overview

This skill allows you to leverage pre-trained GNN representations from MLIPs to train an independent readout head for any custom scalar target property, such as bulk modulus, bandgap, formation energy, or spin states.

To keep the core MLIP wrappers clean, property prediction in AtomisticSkills is handled by standalone training scripts located in the ${CLAUDE_SKILL_DIR}/scripts/ directory.

Workflow

  1. Prepare Data: Build a .json or .xyz dataset containing structures and the corresponding scalar property labels. JSON datasets should be lists of dicts containing a structure key (Pymatgen format) and your target property key.
  2. Determine Property Type: Determine if the property is "intensive" (e.g. Bandgap, Bulk Modulus) or "extensive" (e.g. Total Energy).
    • MatGL logic: For intensive targets, node features undergo a global graph readout (like Set2Set) before passing through an MLP. For extensive targets, the MLP outputs atomic properties which are then sum-pooled.
    • MACE logic: MACE natively supports extensive targets by predicting site-wise scalar outputs and sum-pooling them. When intensive properties are targeted, MACE still sum-pools site-wise outputs, forcing the model to internally learn the intensive invariant.
  3. Execute Script: Run the MACE or MatGL property prediction script in the mlip environment.

Example 1: Training a MACE Property Predictor

MACE property training is handled by scripts/train_mace_property.py. It dynamically patches the mace.cli.run_train module to freeze the backbone (if requested) and inject a custom intensive/extensive property readout.

bash

# Run the standalone MACE property training script
${CLAUDE_SKILL_DIR}/../../venv/run mlip python ${CLAUDE_SKILL_DIR}/scripts/train_mace_property.py \
    --data_path .agents/test/mp_bulk_modulus.json \
    --model_name MACE-OMAT-0-small \
    --target_property bulk_modulus \
    --property_type intensive \
    --epochs 30 \
    --batch_size 16 \
    --lr 0.001 \
    --output_dir custom_mace_results/
  • Adds --freeze_backbone automatically by default to preserve the MACE representation and avoid catastrophic forgetting.
  • The custom weights will be saved to custom_mace_results/.
Show full SKILL.md (207 more words)Show less

Example 2: Training a MatGL (M3GNet) Property Predictor

MatGL property training is handled by scripts/train_matgl_property.py. It loads a pretrained M3GNet model, replaces the data collater to safely handle graph caching, and trains the property explicitly.

bash

# Run the standalone MatGL property training script
${CLAUDE_SKILL_DIR}/../../venv/run mlip python ${CLAUDE_SKILL_DIR}/scripts/train_matgl_property.py \
    --data_path .agents/test/mp_bulk_modulus.json \
    --model_name M3GNet-PES-MatPES-PBE-2025.2 \
    --target_property bulk_modulus \
    --property_type intensive \
    --epochs 30 \
    --batch_size 32 \
    --lr 0.001 \
    --freeze_backbone \
    --output_dir custom_matgl_results/
  • --freeze_backbone is optional (defaults to False for MatGL). If passed, only the final readout MLP layers of the M3GNet model will be fine-tuned.
  • Intensive extensive targets are handled seamlessly without breaking the pretrained model architecture.
  • Model checkpoints are securely saved in custom_matgl_results/matgl_model/.

Constraints

  • Environments: The MACE and MatGL predictors both require mlip. Each code block MUST specify the environment.
  • Data Format: The dataset must be .json or XYZ formatted with the raw structures or ASE atoms.
  • Subprocess Dependency: The train_mace_property.py script spawns an underlying mace.cli.run_train subprocess to maintain compatibility with MACE's native optimizers.
  • Pre-trained Architecture: For MatGL, changing the intensive/extensive nature of a pre-trained model changes the head dimensions. Extensive model predictions are mathematically scaled down by the number of atoms dynamically at training time if an intensive property is targeted.

References

  • Batatia et al., "MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields", NeurIPS, 2022. DOI
  • Chen et al., "Universal potential energy machine learning models", Nat. Comput. Sci., 2022. DOI

Author: Bowen Deng Contact: GitHub @learningmatter-mit

© learningmatter-mit, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 9 other files (scripts) in skills/ml-property-predictor of learningmatter-mit/AtomisticSkills.

  • SKILL.md
  • examples/mace_bulk_modulus/README.md
  • examples/mace_bulk_modulus/run_mace.py
  • examples/matgl_bulk_modulus/README.md
  • examples/matgl_bulk_modulus/matgl_model/model.json
  • examples/matgl_bulk_modulus/matgl_model/model.pt
  • examples/matgl_bulk_modulus/matgl_model/state.pt
  • examples/matgl_bulk_modulus/run_matgl.py
  • scripts/train_mace_property.py
  • scripts/train_matgl_property.py

Open the folder on GitHubat commit 6257444

Compare with similar skills

ML Property Predictor next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

ML Property Predictor compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
ML Property Predictor this skilllearningmatter-mit/AtomisticSkills176—~1.3kAutomated safety check: PassMIT
Scikit LearnzLanqing/codex-claude-academic-skills4.6k17 repos~3.9kAutomated safety check: PassBSD-3-Clause
Senior Data ScientistRaidriar7170/hermes-skilleval1256 repos~1.4kAutomated safety check: PassMIT
Agentic Kaggle WorkflowFrankS-IntelLab/agentic-kaggle-skill188—~4kAutomated safety check: PassMIT
Retention Analysisliangdabiao/claude-data-analysis-ultra-main2901 repos~1.3kAutomated safety check: NotesNone
Geomlitalo-goncalves/geoML109—~4.2kAutomated safety check: PassGPL-3.0

Similar skills

  • Scikit Learn

    zLanqing/codex-claude-academic-skills

    Machine learning in Python with scikit-learn. An agent skill from zLanqing/codex-claude-academic-skills.

    4.6k GitHub starsUsed in 17 repos~3.9k tokens
    Data & AnalyticsAuto-check passed

  • Senior Data Scientist

    Raidriar7170/hermes-skilleval

    World-class data science skill for statistical modeling, experimentation, causal inference, and advanced analytics.

    125 GitHub starsUsed in 6 repos~1.4k tokens
    Data & AnalyticsAuto-check passed

  • Agentic Kaggle Workflow

    FrankS-IntelLab/agentic-kaggle-skill

    Takes a Kaggle competition from rules and validation design through baselines, ensembling and notebook architecture to a scored submission.

    188 GitHub stars~4k tokensUpdated 3 mo ago
    Data & AnalyticsAuto-check passed

  • Retention Analysis

    liangdabiao/claude-data-analysis-ultra-main

    Analyze user retention and churn using survival analysis, cohort analysis, and machine learning.

    290 GitHub starsUsed in 1 repo~1.3k tokens
    Data & AnalyticsAuto-check: notes

  • Geoml

    italo-goncalves/geoML

    Working knowledge of the geoML Python package (github.com/italo-goncalves/geoML): variational Gaussian processes for spatial data, implicit geological modelling, block models, drillhole data…

    109 GitHub stars~4.2k tokensUpdated 6 days ago
    Data & AnalyticsAuto-check passed

  • Radiomics ML

    Aperivue/medsci-skills

    A skill your agent uses when building or auditing a radiomics or tabular clinical-ML prediction model with a classical learner (LASSO, SVM, random forest, XGBoost and similar).

    329 GitHub starsUsed in 1 repo~2.7k tokens
    Data & AnalyticsAuto-check passed

More from learningmatter-mit/AtomisticSkills

All 129 skills in this repo

  • Drug Binding Site Definition

    learningmatter-mit/AtomisticSkills

    Define a docking search box (center coordinates + box dimensions in Angstroms) from a co-crystal ligand, binding-site residues, or a saved JSON specification.

    176 GitHub stars~2.9k tokensUpdated today
    Auto-check passed

  • Drug Complex System Builder

    learningmatter-mit/AtomisticSkills

    Build a solvated, charge-neutralized protein-ligand complex for OpenMM molecular dynamics simulation.

    176 GitHub stars~2k tokensUpdated today
    Auto-check passed

  • Drug Pocket Detection

    learningmatter-mit/AtomisticSkills

    Identify and rank ligandable pockets on a protein structure or model using geometry (fpocket) or an ML predictor (P2Rank).

    176 GitHub stars~4k tokensUpdated today
    Auto-check passed

  • Chem Bond Dissociation

    learningmatter-mit/AtomisticSkills

    Calculate homolytic and heterolytic bond dissociation energies (BDEs) for all single bonds in a molecule using MLIPs with RDKit fragmentation.

    176 GitHub stars~2.5k tokensUpdated today
    Auto-check passed

  • Chem Conformer Search

    learningmatter-mit/AtomisticSkills

    Generate molecular conformers with RDKit ETKDG, relax with MLIPs, and rank by energy with Boltzmann weighting.

    176 GitHub stars~1.3k tokensUpdated today
    Auto-check passed

  • Chem DB Mof

    learningmatter-mit/AtomisticSkills

    Query multiple MOF databases (QMOF via MPContribs; ARC-MOF DB7/Majumdar et al.

    176 GitHub stars~1.9k tokensUpdated today
    Auto-check passed

Categories

Questions about ML Property Predictor

What does ML Property Predictor do?

Train a property predictor head on top of a Machine Learning Interatomic Potential (MLIP) backbone (MACE or MatGL) to predict custom intensive or extensive properties from crystal or molecular…. ML Property Predictor is an agent skill from learningmatter-mit/AtomisticSkills. Train a property predictor head on top of a Machine Learning Interatomic Potential (MLIP) backbone (MACE or MatGL) to predict custom intensive or extensive properties from crystal or molecular structures.

When should I use ML Property Predictor?

ML Property Predictor fits situations like: tasks that involve Machine learning.

How do I install ML Property Predictor in Claude Code?

Run `npx skills add learningmatter-mit/AtomisticSkills --skill ml-property-predictor -a claude-code`. Or copy the skill folder (skills/ml-property-predictor in learningmatter-mit/AtomisticSkills) into .claude/skills/ml-property-predictor in your project. Claude Code loads it when a task matches its description.

How do I install ML Property Predictor in Codex?

Run `npx skills add learningmatter-mit/AtomisticSkills --skill ml-property-predictor -a codex`. Or copy the skill folder (skills/ml-property-predictor in learningmatter-mit/AtomisticSkills) into .agents/skills/ml-property-predictor in your project. Codex loads it when a task matches its description.

Can I use ML Property Predictor in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add learningmatter-mit/AtomisticSkills --skill ml-property-predictor -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ml-property-predictor, .gemini/skills/ml-property-predictor, .github/skills/ml-property-predictor and .opencode/skills/ml-property-predictor in your project.

What does ML Property Predictor need to run?

Going by SKILL.md and its folder, ML Property Predictor needs Python for the scripts in its folder. Our summary lists: Python 3.

Does ML Property Predictor access the network?

SKILL.md names 2 domains. As links in the text: doi.org and github.com. This is read from the text; nothing was executed.

Is ML Property Predictor safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does ML Property Predictor use?

ML Property Predictor is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does ML Property Predictor use?

About 1.3k tokens (SKILL.md is roughly 5.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to ML Property Predictor?

Skills that share tags, products or a category with ML Property Predictor: Scikit Learn (zLanqing/codex-claude-academic-skills, 4.6k stars), Senior Data Scientist (Raidriar7170/hermes-skilleval, 125 stars), Agentic Kaggle Workflow (FrankS-IntelLab/agentic-kaggle-skill, 188 stars) and Retention Analysis (liangdabiao/claude-data-analysis-ultra-main, 290 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains ML Property Predictor?

learningmatter-mit (a GitHub organization) maintains it in learningmatter-mit/AtomisticSkills, which has 176 GitHub stars. The repository holds 129 skills in this directory. The repository was last updated on October 7, 2026.

Source: learningmatter-mit/AtomisticSkills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.