ml-generative-adit

Generate periodic crystal and molecular structures as CIF or XYZ outputs.

144|21|Updated Jan 8, 2026

Search Tags:#generate #MCP tools #materials discovery #crystal structures #molecule generation #diffusion model

Ready to run

npx skills add https://github.com/learningmatter-mit/AtomisticSkills --skill ml-generative-adit

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: ml-generative-adit
Source: https://github.com/learningmatter-mit/AtomisticSkills/tree/main/.agents/skills/ml-generative-adit
Command: npx skills add https://github.com/learningmatter-mit/AtomisticSkills --skill ml-generative-adit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It enables rapid generation of novel atomistic structures—both periodic crystals and non-periodic molecules—so researchers can expand candidate sets for downstream simulation and screening instead of relying only on hand-designed starting points.

Core Features & Use Cases

Crystal structure generation (CIF): Produces periodic materials candidates from the MP20 learned distribution using ADiT’s latent diffusion model.
Molecule generation (XYZ): Produces non-periodic molecular candidates from the QM9 learned distribution using the same unified model.
Workflow integration: Feeds generated structures directly into relaxation, stability checks (e.g., E_hull), and property prediction steps.

Quick Start

Use the ADiT MCP tool to generate 10 crystal structures into your chosen output directory.

Frequently Asked Questions about ml-generative-adit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate novel crystal structures for materials discovery?▼

You can generate novel crystal structures using a latent diffusion model that samples from the MP20 learned distribution, outputting periodic material candidates directly as CIF files for downstream atomistic simulation and screening.

Can I generate non-periodic molecules for computational chemistry workflows?▼

Yes, you can generate non-periodic molecules by sampling from the QM9 learned distribution using the same unified diffusion model, outputting molecular candidates as XYZ files for computational chemistry workflows.

Do I need a CUDA-enabled environment to run diffusion models for structure generation?▼

Yes, running the ADiT MCP tool for crystal and molecule generation requires a CUDA-enabled environment to execute the latent diffusion model and produce the desired CIF or XYZ output files with metadata.

What is the best way to expand candidate sets for atomistic materials research?▼

The best way to expand candidate sets is using a generative diffusion model to rapidly produce novel atomistic structures, replacing hand-designed starting points with data-driven candidates for downstream stability and property analysis.

How do generated crystal structures integrate with downstream stability checks?▼

Generated crystal structures feed directly into downstream workflow steps including structural relaxation, stability checks like E_hull calculation, and property prediction to validate the novel material candidates.

What are the limitations of using generative models for materials discovery?▼

Generated structures from diffusion models represent candidates from learned distributions like MP20 and QM9, requiring subsequent structure validation, relaxation, and stability analysis rather than serving as immediately finalized materials.