Xiaoman Zhang

khazix-skills. 数字生命卡兹克开源的 AI Skills 合集 | Agent Skills: leader（帮你定义目标）, neat-freak 洁癖, hv-analysis, khazix-writer & more — Claude Code, Codex & 40+ agents

JESTR1. Joint Embedding Space Technique for Ranking Candidate Molecules for the Annotation of Untargeted Metabolomics Data

frontend-slides. Create beautiful slides on the web using a coding agent's frontend skills

BioMedAgent. Python

RexSonoVQA. A comprehensive benchmark for evaluating vision-language models on ultrasound video understanding tasks.

Office-Word-MCP-Server. A Model Context Protocol (MCP) server for creating, reading, and manipulating Microsoft Word documents. This server enables AI assistants to work with Word documents through a standardized interface, providing rich document editing capabilities.

autoresearch. AI agents running research on single-GPU nanochat training automatically

MinerU-Ecosystem. Python

posterskill. AI-assisted academic posters.

deepagents. The batteries-included agent harness.

alphagenome-pytorch. AlphaGenome PyTorch port

CRIMSON. Python

protein-design-skills. Claude Code skills for protein design

VenusFactory2. 🏭 AI agent platform with skills for protein engineering, the noob-friendly AI tutorial tool for life science professionals.

OpenClaw-Medical-Skills. The largest open-source medical AI skills library for OpenClaw🦞.

academic-research-skills. Academic Research Skills for Claude Code: research → write → review → revise → finalize

hlip. Towards Scalable Language-Image Pre-training for 3D Medical Imaging [TMLR 2026]

scientific-agent-skills. Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 170,000+ scientists worldwide. 158 ready-to-use skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.

skills. Public repository for Agent Skills

cellrank. CellRank: dynamics from multi-view single-cell data

protap. prot_learn is a PyTorch-based machine learning toolbox designed for protein analysis

dplm. The Family of Diffusion Protein Language Models (DPLM)

MAP. Python

openclaw. Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

RadEval. RadEval: A framework for radiology text evaluation

Open-Claude-Cowork. OpenSource Claude Code Cowork lets you can use Claude Cowork with GLM4.7 or MinMax2.1. Claude Code for the rest of your work. Cowork lets you complete non-technical tasks much like how developers use Claude Code.

DALI. A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.

SimVQ. [ICCV 2025] SimVQ: Addressing Representation Collapse in Vector Quantized Models with One Linear Layer

Biomni. Biomni: a general-purpose biomedical AI agent

alphagenome_research. Research code accompanying AlphaGenome

StructTokenBench. A benchmark for comprehensive evaluation on protein structure tokenization methods

ProTrek. ProTrek: illuminating the Protein Universe through Trimodal Protein Language Model (Connecting text and protein)

esm. Jupyter Notebook

ReX-MLE. Python

open-h-embodiment. Open-H-Embodiment is a community‑driven dataset initiative building the open, shared foundation needed to train and evaluate a generalist Vision‑Language‑Action (VLA) model for healthcare robotics

SciToolAgent. SciToolAgent: A Knowledge Graph-Driven Scientific Agent for Multi-Tool Integration

uqlm. UQLM: Uncertainty Quantification for Language Models, is a Python package for UQ-based LLM hallucination detection

KARMA. Leveraging Multi-Agent LLMs for Automated Knowledge Graph Enrichment

pillar-pretrain. This repository contains the pretraining code for the Pillar-0 model.

robot-learning-tutorial. All the source code for "Robot Learning: A Tutorial". Get involved to be featured in the next iteration!

lerobot. 🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning

graphiti. Build Real-Time Knowledge Graphs for AI Agents

AutoReproduce. Python

Paper2Code. Paper2Code: Automating Code Generation from Scientific Papers in Machine Learning

disco_rl. Accompanying code for "Discovering State-of-the-art Reinforcement Algorithms" Nature publication

OpenCE. OpenCE (Open Context Engineering): A community toolkit to implement, evaluate, and combine LLM context strategies (RAG, ACE, Compression). Evolved from the `ACE-open` reproduction.

ML-Master. The official implementation of "ML-Master: Towards AI-for-AI via Integration of Exploration and Reasoning"

state. State is a machine learning model that predicts cellular perturbation response across diverse contexts

InternAgent. InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery

M3Builder. The official codes for "M^3Builder: A Multi-Agent System for Automated Machine Learning in Medical Imaging"

gemini-cli. An open-source AI agent that brings the power of Gemini directly into your terminal.

ai-research-assistant. Aria is Your AI Research Assistant Powered by GPT Large Language Models

CRISP. Nature Computational Science: Predicting drug responses of unseen cell types through transfer learning with foundation models

AgentFlow. AgentFlow: In-the-Flow Agentic System Optimization

PANTHER_baseline. Python

PromoterAI. Python

Scientific-LLM-Survey. Scientific Large Language Models: A Survey on Biological & Chemical Domains

SpliceVarDB. An online database of variants functionally demonstrated to affect (or not affect) splicing.

TissueLab. A Co-evolving Agentic AI System for Medical Imaging Analysis

borzoi. RNA-seq prediction with deep convolutional neural networks.

mle-bench. MLE-bench is a benchmark for measuring how well AI agents perform at machine learning engineering

aira-dojo. AIRA-dojo: a framework for developing and evaluating AI research agents

MLE-STAR-Open. Unofficial, Google‑free reimplementation of MLE‑STAR: a lightweight, local‑friendly multi‑agent ML engineering pipeline that uses OpenAI‑compatible LLMs (OpenRouter/Ollama) and DuckDuckGo search to generate, debug, refine, ensemble, and submit Kaggle‑ready solutions—AutoML for tabular tasks with code generation and iterative refinement.

AutoKaggle. Python

REPRESS. Python

DNABERT. DNABERT: pre-trained Bidirectional Encoder Representations from Transformers model for DNA-language in genome

RNA-FM. Nature Methods: RNA foundation model (together with RhoFold)

linear_perturbation_prediction-Paper. HTML

perturbench. Jupyter Notebook

arc-virtual-cell-atlas. Arc Virtual Cell Atlas

transcriptformer. Jupyter Notebook

agents. Multi-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLI

kaggle-solutions. 🏅 Collection of Kaggle Solutions and Ideas 🏅

gello_mechanical. OpenSCAD

gello_software. Python

i4h-docs. Documentation for Isaac for Healthcare

SAMUS. Python

FLAML. A fast library for AutoML and tuning. Join our Discord: https://discord.gg/Cppx2vSPVP.

DiagGym. A virtual clinical environment for self‑evolving LLM diagnostic agents.

awesome-claude-code. A hand-picked collection of the finest of resources for the most awesome of agents, Claude Code, the undisputed champion of coding companions, from the unstoppable team at Anthropic PBC. A delectable showcase of top tier skills, ambidextrous agents, scintillating status lines, top notch developer tooling, and also we have plugins

UltraBot. [Nature Communications 2025] Towards Expert-level Autonomous Carotid Ultrasonography with Large-scale Learning-based Robotic System

dinov3. Reference PyTorch implementation and models for DINOv3

SonoGym. Jupyter Notebook

PathPT. [Nature Communications, 2026] The official code for "Boosting Pathology Foundation Models via Few-shot Prompt-tuning for Rare Cancer Subtyping"

DeepCode. "DeepCode: Open Agentic Coding (Paper2Code & Text2Web & Text2Backend)"

kaggle-mcp. Kaggle-MCP: Connect Claude AI to the Kaggle API through the Model Context Protocol (MCP), enabling competition, dataset, and kernel operations through the AI interface.

Self-Evolving-Agents.

sawIntuitiveResearchKit. cisst/SAW stack for the da Vinci Research Kit (dVRK)

SurgicAI. Python

IsaacLab. Unified framework for robot learning built on NVIDIA Isaac Sim

cursor-chat-export. Export your chats from Cursor to Markdown.

browser-use. 🌐 Make websites accessible for AI agents. Automate tasks online with ease.

langgraph. Build resilient agents.

MedAgentBench. MedAgentBench: A Realistic Virtual EHR Environment to Benchmark Medical LLM Agents

MedAgentBench-dev. A Comprehensive Benchmark to Evaluate LLMs as Agents in Medical Applications

LesionLocator. LesionLocator is a framework for zero-shot lesion segmentation and longitudinal tumor tracking in 3D full-body imaging.

nnInteractive. nnInteractive is a framework for 3D interactive segmentation, supporting intuitive prompts like points, scribbles, bounding boxes, and lasso. Trained on 120+ diverse 3D datasets, it sets a new standard in accuracy, usability, and adaptability for clinical and research applications.

awesome-medical-mcp-servers. A collection of Medical MCP servers.

gemini-fullstack-langgraph-quickstart. Get started with building Fullstack Agents using Gemini 2.5 and LangGraph

mindshub. Make AI do actual work. Swap the model anytime — keep everything you've built.

AgentGPT. 🤖 Assemble, configure, and deploy autonomous AI Agents in your browser.

vibe-tools. Give Cursor Agent an AI Team and Advanced Skills

scDrugMap. Python

EvoAgentX. 🚀 EvoAgentX: Building a Self-Evolving Ecosystem of AI Agents

Awesome-LLM-Scientific-Discovery. [EMNLP2025] From Automation to Autonomy: A Survey on Large Language Models in Scientific Discovery

dcurves. Python package for Andrew Vickers' Decision Curve Analysis method to evaluate prediction models and diagnostic tests

TITAN. Multimodal Whole Slide Foundation Model for Pathology - Nature Medicine

chexprompt. Expert-level AI radiology report evaluator

Merlin. [Nature 2026] Merlin is a 3D VLM for computed tomography that leverages both structured electronic health records (EHR) and unstructured radiology reports for pretraining.

FactCheXcker. This repository contains code to run the FactCheXcker pipeline on model-generated chest X-ray reports that contain quantifiable metrics, such as endotracheal tube placements.

sleepfm-codebase. Python

oard-react. Jupyter Notebook

unsloth. Unsloth is a local UI for training and running Kimi K3, Gemma 4, Qwen3.6, DeepSeek, GLM and other models.

MCMED. Jupyter Notebook

octotools. OctoTools: An agentic framework with extensible tools for complex reasoning

awesome-ai-agents. A list of AI autonomous agents

autogen. A programming framework for agentic AI

dinov2. PyTorch code and models for the DINOv2 self-supervised learning method.

dspy. DSPy: The framework for programming—not prompting—language models

ehrshot-benchmark. A benchmark for few-shot evaluation of foundation models for electronic health records (EHRs)

Patho-Bench. Standardized benchmark for computational pathology foundation models.

MDS-ED. Repository for the paper 'Enhancing Clinical Decision Support with Physiological Waveforms — A Multimodal Benchmark in Emergency Care'.

r1_vlm. Build your own visual reasoning model

DINO. [ICLR 2023] Official implementation of the paper "DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection"

align-anything. Align Anything: Training All-modality Model with Feedback

headCT_foundation. Foundation 3D ViT model for volumetric head CT - Nature Biomedical Engineering

Awesome-LLM-Long-Context-Modeling. 📰 Must-read papers and blogs on LLM based Long Context Modeling 🔥

open-deep-research. An open source deep research clone. AI Agent that reasons large amounts of web data extracted with Firecrawl

Qwen3-VL. Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

R1-V. Witness the aha moment of VLM with less than $3.

CT-FM. CT-FM: A 3D Image-Based Foundation Model for Computed Tomography

simpleRL-reason. Simple RL training for reasoning

DeepSeek-R1.

Janus. Janus-Series: Unified Multimodal Understanding and Generation Models

open-r1. Fully open reproduction of DeepSeek-R1

CTseg. CTseg: A Tool for Brain CT Segmentation, Spatial Normalisation, and Volumetrics

Quasar-mass-machine-learning. Using machine learning to predict the mass of quasar supermassive black holes

ViFM_Survey. Foundation Models for Video Understanding: A Survey

AgentLaboratory. Agent Laboratory is an end-to-end autonomous research workflow meant to assist you as the human researcher toward implementing your research ideas

awesome-ai4s. AI for Science 论文解读合集（持续更新ing），论文/数据集/教程下载：hyper.ai

large_concept_model. Large Concept Models: Language modeling in a sentence representation space

RadABench. The official codes for "Can Modern LLMs Act as Agent Cores in Radiology Environments?"

my-research-wrapped. Automated LLM summary of your research papers

HeadCTSegmentation. Multi-class U-Net for head CT segmentation

ToolkenGPT. ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings - NeurIPS 2023 (oral)

TokenPacker. The code for "TokenPacker: Efficient Visual Projector for Multimodal LLM", IJCV2025

unmasked_teacher. [ICCV2023 Oral] Unmasked Teacher: Towards Training-Efficient Video Foundation Models

GMAI-VL. GMAI-VL & GMAI-VL-5.5M: A Large Vision-Language Model and A Comprehensive Multimodal Dataset Towards General Medical AI.

open_clip. An open source implementation of CLIP.

PPLLaVA. Official GPU implementation of the paper "PPLLaVA: Varied Video Sequence Understanding With Prompt Guidance"

Medical-SAM-Adapter. A lightweight adapter bridges SAM with medical imaging [MedIA]

blast-ct. Brain Lesion Analysis and Segmentation Tool for Computed Tomography

AbdomenAtlas. [NeurIPS 2023] AbdomenAtlas 1.0 (5,195 CT volumes + 9 annotated classes)

unintentional-unalignment. [ICLR 2025] Unintentional Unalignment: Likelihood Displacement in Direct Preference Optimization

awesome-vlm-architectures. Famous Vision Language Models and Their Architectures

learning_research. 本人的科研经验

voxel. Torch utilities for 3D medical imaging

NeMo-Aligner. Scalable toolkit for efficient model alignment

vllm. A high-throughput and memory-efficient inference and serving engine for LLMs

self-correction-llm-papers. This is a collection of research papers for Self-Correcting Large Language Models with Automated Feedback.

HeadCT-ONE. Python

LLMs-Preference-Optimization.

eureka-ml-insights. A framework for standardizing evaluations of large foundation models, beyond single-score reporting and rankings.

torchxrayvision. TorchXRayVision: A library of chest X-ray datasets and models. Classifiers, segmentation, and autoencoders.

textgrad. TextGrad: Automatic ''Differentiation'' via Text -- using large language models to backpropagate textual gradients. Published in Nature.

LlamaFactory. Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

mteb. MTEB: State-of-the-art evaluation of embeddings across languages and modalities

MedS-Ins. [npj digital medicine] The official codes for "Towards Evaluating and Building Versatile Large Language Models for Medicine"

ReXKG. Python

ProbMed. Official repository for the ACL 2025 Findings paper "Worse than Random? An Embarrassingly Simple Probing Evaluation of Large Multimodal Models in Medical VQA"

AutoRG-Brain. The official codes for "AutoRG-Brain: Grounded Report Generation for Brain MRI".

CHIEF. Clinical Histopathology Imaging Evaluation Foundation Model

gpt_paper_assistant. GPT4 based personalized ArXiv paper assistant bot

FORTE. The original code for paper "Towards a Holistic Framework for Multimodal LLM in 3D Brain CT Radiology Report Generation"

BiomedGPT. BiomedGPT: A Generalist Vision-Language Foundation Model for Diverse Biomedical Tasks

spark-nlp. State of the Art Natural Language Processing

AutoRG-Brain_Annotation_Platform. The code implementation for the annotation platform in AutoRG-Brain

ReXrank. HTML

GREEN. [EMNLP, Findings 2024] a radiology report generation metric that leverages the natural language understanding of language models to identify and explain clinically significant errors in candidate reports

R2Gen. Python

craft-md. Python

RaDialog_v2. LLaVa Version of RaDialog

RaTEScore. [EMNLP 2024] RaTEScore: A Metric for Radiology Report Generation

SQuAD-explorer. Visually Explore the Stanford Question Answering Dataset

CXR-Report-Metric. Python

R2-LLM. The official GitHub repository of the AAAI-2024 paper "Bootstrapping Large Language Models for Radiology Report Generation".

cvt2distilgpt2. Improving Chest X-Ray Report Generation by Leveraging Warm-Starting

ifcc. The code of Improving Factual Completeness and Consistency of Image-to-text Radiology Report Generation

chexpert-plus. Python

FineRadScore. Python

Knowledge-Graph-And-Visualization-Demo. There are 2D-Search and 3D-Graph-View for knowledge graph visualization.

imageclefmedical_caption_23. MedICap: Code for the participation of team CSIRO at the ImageCLEFmedical Caption task of 2023.

MedMNIST. [pip install medmnist] 18x Standardized Datasets for 2D and 3D Biomedical Image Classification

Awesome-Foundation-Models-for-Advancing-Healthcare. We present a comprehensive and deep review of the HFM in challenges, opportunities, and future directions. The released paper: https://arxiv.org/abs/2404.03264

self-llm. 《开源大模型食用指南》针对中国宝宝量身打造的基于Linux环境快速微调（全参数/Lora）、部署国内外开源大模型（LLM）/多模态大模型（MLLM）教程

lit-llama. Implementation of the LLaMA language model based on nanoGPT. Supports flash attention, Int8 and GPTQ 4bit quantization, LoRA and LLaMA-Adapter fine-tuning, pre-training. Apache 2.0-licensed.

HuggingFace-Download-Accelerator. 利用HuggingFace的官方下载工具从镜像网站进行高速下载。

CT-CLIP. Developing Generalist Foundation Models from a Multimodal Dataset for 3D Computed Tomography

recognize-anything. Open-source and strong foundation image recognition models.

flamingo-mini. Implementation of the deepmind Flamingo vision-language model, based on Hugging Face language models and ready for training