Machine learning

Train a model to predict custom properties of molecules or periodic materials using pretrained SelfConditionedDenoisingAtoms (SCD) foundation models.

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

Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction.

Meta's 86M prompt injection and jailbreak detector.

Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL).

Use this skill whenever the user is working with the Pydantic AI framework — including building AI agents, defining structured outputs with Pydantic models…

Build Gradio web UIs and demos in Python.

Look up and read Hugging Face paper pages in markdown, and use the papers API for structured metadata such as authors, linked models/datasets/spaces, Github…

Interactive discovery + implementation workflow that gathers requirements through picker-based questions (intent, scope, constraints, preferences), scans the…

Fast tokenizers optimized for research and production.

Data framework for building LLM applications with RAG.

Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace.

Unified search skill with Intelligent Auto-Routing.

train a Cluster Expansion (CE) for lattice-based Monte Carlo simulation of disordered materials.

Fine-tune Fairchem machine learning interatomic potentials (UMA, ESEN) on custom datasets.

Generate inorganic material structures using MatterGen, a diffusion-based generative model.

Fine-tune MACE machine learning interatomic potentials on custom datasets.

Fine-tune MatGL machine learning interatomic potentials on custom datasets.

Benchmark of inference speed of Machine Learning Interatomic Potentials (MLIPs).

End-to-end ONNX PTQ workflow for AMD Quark — from a `.onnx` file (and calibration data) to a quantized `.onnx` output.

Torch LLM PTQ workflow for AMD Quark — from model selection to quantized output.

Validate workspace shape, model paths, output directories, and repo structure before downstream Quark skills proceed.

GPT-5.6 全栈逆向+渗透融合 Skill（已并入 Code Sentinel）。 覆盖：通用/深度/脱壳/黑盒/JS/APK/Binary Diff 逆向；渗透/API/攻击链； 游戏/卡密/Pwn/IoT/恶意样本/EDR/LLM/供应链/Patch Diff/CTF 全速解题； Kali…