Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/learningmatter-mit/atomisticskills/ml-generative-mattergennpx skills add learningmatter-mit/AtomisticSkills --skill ml-generative-mattergengit clone --depth 1 https://github.com/learningmatter-mit/AtomisticSkillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/learningmatter-mit/atomisticskills/ml-generative-mattergen)<a href="https://agentmods.dev/skills/learningmatter-mit/atomisticskills/ml-generative-mattergen"><img src="https://agentmods.dev/badge/skills/learningmatter-mit/atomisticskills/ml-generative-mattergen.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00022 | $0.01507 |
| Opus 5 | $0.00011 | $0.00754 |
| Sonnet 5 | $0.00004 | $0.00301 |
| Haiku 4.5 | $0.00002 | $0.00151 |
Grade A, and why
ml-generative-mattergen scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MatterGen Structure Generation Skill
This skill provides tools for generating novel inorganic material structures using MatterGen, a state-of-the-art diffusion-based generative model for crystalline materials.
1. Prerequisites
[!IMPORTANT] ARM/aarch64 Support: MatterGen CAN work on ARM-based systems like NVIDIA DGX Spark. However, PyG dependencies (torch-scatter, torch-cluster) must be compiled from source with CUDA_HOME properly configured. See installation guide below.
- The
mattergen-agentconda environment must be installed and configured. - MatterGen requires Python 3.10 and CUDA 13.0 compatible GPU for efficient generation.
- For ARM/aarch64 systems: See Installing torch-scatter on ARM for detailed installation instructions.
2. Available Models
MatterGen provides several pretrained models:
mattergen_base: Base unconditional generative modelmp_20_base: Materials Project base modeldft_mag_density: Model for magnetic density conditioningchemical_system: Model for chemical system conditioning
3. MCP Tool Usage
The MCP tool automatically loads models when needed - no explicit load step required.
Unconditional Generation
Generate novel structures without conditioning:
from mcp_base import mcp_mattergen_generate_structures
result = mcp_mattergen_generate_structures(
model_name="mattergen_base",
num_structures=10,
batch_size=10,
output_dir="research/my_project/generated"
)
Chemical System Conditioning
Generate structures from a specific chemical system (controls which elements appear):
result = mcp_mattergen_generate_structures(
chemical_system="Li-Fe-P-O", # Automatically uses chemical_system model
guidance_scale=1.0, # Recommended for chemical system conditioning
num_structures=20,
batch_size=10,
output_dir="research/cathode_materials/generated"
)
[!NOTE] Chemical system conditioning controls which elements appear, but NOT the exact stoichiometry. For example,
chemical_system="Li-Zr-Cl"can generate Li3Cl5, LiZrCl4, Li2ZrCl5, etc., but you cannot specify exactly "Li2ZrCl6".
What ships with it
8 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- examples/finetuning/example_finetuned_model/adapter_config.json 0 B
- examples/finetuning/example_finetuned_model/adapter_model.bin 0 B
- examples/finetuning/make_dummy.py 337 B runs code
- examples/finetuning/README.md 1.8 KB
- examples/finetuning/run_finetune_example.sh 1.1 KB runs code
- examples/finetuning/training_data.csv 0 B
- scripts/prepare_training_data.py 5.7 KB runs code
- scripts/run_finetuning.py 11 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 181 lines · 22 tokens per session scan A 737749f597b5
ml-generative-mattergen is a skill published in the GitHub repository learningmatter-mit/AtomisticSkills (158 stars, last pushed yesterday), licensed MIT. It adds 22 tokens to every session and 1,507 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
hugging-face-trackio
Track and visualize ML training experiments with Trackio. Use when logging metrics during training (Python API) or retrieving/analyzing logged metrics (CLI). Supports real-time dashboard visualization, HF Space syncing, and JSON output for automation.
pysr
Use when fitting equations to data with PySR or SymbolicRegression.jl, when a user wants an interpretable formula, symbolic model, scaling law, or empirical relation discovered from numeric data, or when debugging a PySR search that is slow, stuck, or giving poor equations.
cellxgene-census-query
Query CZ CELLxGENE Census (61M+ cells). Filter by cell type/tissue/disease, retrieve expression data, and integrate with scanpy/PyTorch for population-scale single-cell analysis. Use this skill when: (1) Querying single-cell expression data by cell type, tissue, or disease, (2) Exploring available single-cell datasets…
HomeSafe-Bench
VLM indoor safety hazard detection benchmark inspired by HomeSafeBench (arXiv 2509.23690).
depth-estimation
Real-time depth map privacy transforms using Depth Anything v2 (CoreML + PyTorch).
start-experiment
Start the autoresearch optimization loop for a specific model + lane. Resolves the hierarchical program.md (root → model → lane), asks the user for hardware (local TPU VM or GKE cluster of a specified TPU type + topology), discovers available clusters from .env/, checks occupancy with USERPREFIX-aware attribution…