# Claw4Science

> Curated directory of AI agent projects and skill libraries for scientific research, bioinformatics, and biomedicine.

## Website
- Homepage: https://claw4science.org/
- Skills Directory: https://claw4science.org/skills
- Skill Hubs: https://claw4science.org/skill-hubs
- Blog: https://claw4science.org/blog
- About: https://claw4science.org/about
- Developers: https://claw4science.org/developers
- API: https://claw4science.org/api/projects

## Projects (125+ total)

### Core Platform & Lightweight Variants
- [OpenClaw](https://github.com/openclaw/openclaw): The core platform. Personal AI assistant connecting 20+ messaging apps (WhatsApp, Telegram, Slack, Discord, Signal, iMessage, Teams, Matrix, Feishu, LINE). Voice support on macOS/iOS/Android with live Canvas rendering.
- [NanoBot](https://github.com/HKUDS/nanobot): Ultra-lightweight OpenClaw in Python. 99% less code, full agent capabilities. Supports MCP, multi-model (OpenAI, Anthropic, Gemini, Ollama), memory, cron, sub-agents. Install: pip install nanobot-ai.
- [PicoClaw](https://github.com/sipeed/picoclaw): Go-based agent running on $10 hardware. <10MB RAM, <1s boot time. 99% less memory than OpenClaw. Supports x86_64, ARM64, MIPS, RISC-V, and LoongArch architectures.
- [MicroClaw](https://github.com/microclaw/microclaw): Agentic AI assistant built with Rust. Runs on <1MB RAM targeting $2-5 MCU hardware like ESP32. The smallest member of the Claw family for sensor-level embedded AI.
- [ZeroClaw](https://github.com/zeroclaw-labs/zeroclaw): Fast, small, and fully autonomous AI assistant infrastructure. Deploy anywhere, swap anything. Written in Rust with extreme performance focus — the infrastructure layer for self-hosted AI agents.
- [NanoClaw](https://github.com/qwibitai/nanoclaw): Lightweight alternative to OpenClaw that runs in containers for security. Connects to WhatsApp, Telegram, Slack, Discord, Gmail. Built on Anthropic Agents SDK with memory and scheduled jobs.
- [EdgeClaw](https://github.com/Openbmb/EdgeClaw): Edge-Cloud Collaborative AI Assistant built on OpenClaw. Three-tier privacy system (Passthrough/Desensitization/Local) with dual-engine edge detector. Keeps sensitive data local while routing 80% of requests to cost-effective cloud models. By THUNLP, RUC, OpenBMB.

### Team & Evolution
- [Clawith](https://github.com/dataelement/Clawith): OpenClaw for Teams. Multi-agent collaboration with persistent identity, long-term memory, and autonomous 'Aware' consciousness system. 6 trigger types: cron, once, interval, poll, on_message, webhook.
- [MetaClaw](https://github.com/aiming-lab/MetaClaw): Self-learning agent that evolves from every conversation. LoRA-based continual learning + reinforcement learning, no GPU cluster needed. Supports OpenClaw, NanoBot, PicoClaw backends.
- [ClawX](https://github.com/ValueCell-ai/ClawX): Desktop app providing a graphical interface for OpenClaw AI agents. Turns CLI-based AI orchestration into a visual desktop experience without using the terminal.

### Science & Research (Claw Ecosystem)
- [EvoScientist](https://github.com/EvoScientist/EvoScientist): Self-evolving multi-agent AI research system. 6 sub-agents (plan, research, code, debug, analyze, write) for end-to-end scientific discovery. #1 on DeepResearch Bench II. ICAIS 2025 Best Paper Award.
- [MedgeClaw](https://github.com/xjtulyc/MedgeClaw): AI research assistant for biomedicine. Built on OpenClaw + Claude Code with 140 K-Dense Scientific Skills. Covers RNA-seq, drug discovery, clinical analysis. Integrated RStudio & JupyterLab.
- [OmicsClaw](https://github.com/TianGzlab/OmicsClaw): Multi-omics AI analysis tool from TianGz Lab. Focused on genomics, transcriptomics, and proteomics data processing with intelligent agent workflows.
- [LabClaw](https://github.com/wu-yc/LabClaw): Operating layer for LabOS — Stanford-Princeton AI Co-Scientists project. Automated scientific co-discovery workflows for laboratory research.
- [MedClaw](https://github.com/zteyesreal/medclaw): Early-stage medical AI project exploring agent-based approaches for clinical decision support and medical data analysis.
- [ClawBio](https://github.com/ClawBio/ClawBio): The first bioinformatics-native AI agent skill library. Local-first, reproducible, built on OpenClaw. Covers genomics, population genetics, and equity-focused bioinformatics workflows.
- [BioClaw](https://github.com/Runchuan-BU/BioClaw): AI-powered bioinformatics research assistant built on OpenClaw. Integrates PubMed literature search and PyMOL molecular visualization for streamlined biological research workflows.
- [MolClaw](https://github.com/IDEA-XL/MolClaw): Light, practical, memory-enabled AI orchestrator for science. Built on NanoClaw. Adds OpenAI/OpenRouter providers, Discord/WhatsApp channels, runtime skills, durable session memory, and real-time agent execution dashboard.

### Education
- [OpenMAIC](https://github.com/THU-MAIC/OpenMAIC): Open Multi-Agent Interactive Classroom by Tsinghua University. One-click immersive AI classroom with AI teachers and peers. Slides, quizzes, whiteboard, PBL. Published in JCST'26.

### Bioinformatics & Omics
- [PopGenAgent](https://github.com/ai4nucleome/POPGENAGENT): Tool-aware, reproducible, report-oriented AI workflows for population genomics. Template-driven execution with publication-quality figures, full provenance tracking, and cost-aware LLM routing.
- [CARIBOU](https://github.com/OpenTechBio/CARIBOU): Computational AI Research Interface for Bioinformatics, Omics, and Unifying Agents. Multi-agent bioinformatics system by OpenTechBio.
- [AutoBA](https://github.com/JoshuaChou2018/AutoBA): An AI agent for fully automated multi-omic analyses. Supports RNA-seq, scRNA-seq, spatial transcriptomics, WGS/WES, and ChIP-seq with automated code repair.
- [CellAgent](https://github.com/lsq2wal/CellAgent): LLM-driven multi-agent framework for automated scRNA-seq analysis. Three-role architecture: Planner, Executor, and Evaluator for single-cell RNA sequencing workflows.
- [SpatialAgent](https://github.com/Genentech/SpatialAgent): An AI agent for spatial biology by Genentech. Processes multimodal spatial genomics data for research insights.
- [CRISPR-GPT](https://github.com/cong-lab/crispr-gpt-pub): LLM agent for CRISPR genome engineering. Assists with guide RNA design, experiment planning, and protocol optimization.
- [BioMaster](https://github.com/ai4nucleome/BioMaster): AI agent system for various bioinformatics tasks with specialized agent coordination. From the same group as POPGENAGENT.
- [CellVoyager](https://github.com/zou-group/CellVoyager): AI CompBio agent that autonomously analyzes biological data and generates new insights. From Stanford Zou Group. Published in Nature Methods. Specializes in scRNA-seq with self-debugging and hypothesis generation.

### Drug Discovery & Molecular
- [AIAgents4Pharma](https://github.com/VirtualPatientEngine/AIAgents4Pharma): Multi-agent platform for pharmaceutical R&D. Covers drug discovery, drug development, and clinical pipeline automation. Built by VirtualPatientEngine.
- [ChemGraph](https://github.com/argonne-lcf/ChemGraph): Agentic framework for computational chemistry and materials science workflows. From Argonne National Laboratory (ALCF). Orchestrates chemistry simulation pipelines.
- [Virtual Lab](https://github.com/zou-group/virtual-lab): A virtual lab of LLM agents for science research. Multi-agent system for nanobody design with principal investigator agent coordination.
- [MDCrow](https://github.com/ur-whitelab/MDCrow): Molecular dynamics simulations with an LLM agent. Automates simulation setup, execution, and analysis using natural language.

### General Research Agents
- [RD-Agent](https://github.com/microsoft/RD-Agent): Microsoft R&D automation platform. AI-driven data and model research — automates hypothesis generation, experiment design, and iterative improvement for industrial and scientific R&D.
- [AI-Scientist-v2](https://github.com/SakanaAI/AI-Scientist-v2): Sakana AI's autonomous scientific discovery system. Workshop-level paper generation via agentic tree search. End-to-end: idea to experiment to paper. Successor to the 10K-star AI-Scientist v1.
- [DATAGEN](https://github.com/starpig1129/DATAGEN): AI-driven multi-agent research assistant automating hypothesis generation, data analysis, and report writing. Full research pipeline from question to publication-ready report.
- [OpenAGS](https://github.com/openags/OpenAGS): Autonomous Generalist Scientist — AI agent targeting all scientific fields. Broad-scope autonomous research from hypothesis to verification.
- [Open Co-Scientist](https://github.com/jataware/open-coscientist): Open-source adaptation of Google DeepMind's AI Co-Scientist. Generates, reviews, ranks, and evolves research hypotheses using multi-agent architecture.
- [AI-Researcher](https://github.com/HKUDS/AI-Researcher): Autonomous scientific innovation system from HKUDS. NeurIPS 2025 paper. End-to-end research automation: literature review, hypothesis generation, experiment design, and paper writing. Production version at novix.science.

### Specialized Science
- [OpenLens AI](https://github.com/jarrycyx/openlens-ai): Fully autonomous multimodal research agent. Combines vision and text understanding for scientific discovery. Chinese academic origin with bilingual support.
- [BioAgents](https://github.com/bio-xyz/BioAgents): AI scientist framework for autonomous deep research in biological sciences. Multi-agent system combining literature analysis with data scientist agents for iterative scientific discovery.
- [SR-Scientist](https://github.com/GAIR-NLP/SR-Scientist): Scientific equation discovery with agentic AI. Published at ICLR 2026. Symbolic regression via AI agents from GAIR Lab (Shanghai Jiao Tong University).
- [BioDiscoveryAgent](https://github.com/snap-stanford/BioDiscoveryAgent): LLM-based AI agent for closed-loop design of genetic perturbation experiments. Autonomous reasoning for biological discovery.

Note: The full list of 82 projects is available at https://claw4science.org/ and via the API at https://claw4science.org/api/projects.

## Skill Hubs (29 total)

The Skill Hubs directory at https://claw4science.org/skill-hubs catalogs 21 major skill registries and community collections for AI coding agents, including ClawHub (6,300+ skills), Awesome OpenClaw Skills (5,400+), Skills.sh (89,000+), Skills.lc (250K+), OpenClaw Medical Skills (869), and more.

## Curated Skills (125+ total)

The Skills directory at https://claw4science.org/skills highlights 15 hand-picked skill libraries spanning scientific research, bioinformatics, academic writing, medical AI, and general-purpose agent workflows. Featured collections include K-Dense Scientific Skills (170+ across 17 domains), AI Research Skills (86 across 22 categories), Academic Research Skills, and ClawBio.

## Skill Libraries (29 total)

- [ClawHub](https://github.com/openclaw/clawhub): The official OpenClaw skill registry. Publish, version, search, and install text-based agent skills. 6,300+ skills with vector search, starring, commenting, and admin curation.
- [Awesome OpenClaw Skills](https://github.com/VoltAgent/awesome-openclaw-skills): 5,400+ community skills filtered and categorized from the official Skills Registry. Install via clawhub install <slug> or manual copy. Maintained by VoltAgent.
- [K-Dense Scientific Skills](https://github.com/K-Dense-AI/claude-scientific-skills): 170+ ready-to-use skills across 17 scientific domains: bioinformatics, cheminformatics, proteomics, clinical research, materials science, physics, and more. Connected to 250+ databases.
- [AI Research Skills](https://github.com/Orchestra-Research/AI-Research-SKILLs): 86 skills across 22 categories for autonomous AI research. Covers ideation, ML paper writing, fine-tuning, distributed training, inference, safety, agents, RAG, multimodal, and more.
- [Academic Research Skills](https://github.com/Imbad0202/academic-research-skills): Full academic pipeline: 13-agent deep research team, 12-agent paper writing (LaTeX/DOCX/MD), multi-perspective peer review with scoring rubrics, and 10-stage pipeline orchestrator.
- [OpenClaw Medical Skills](https://github.com/FreedomIntelligence/OpenClaw-Medical-Skills): The largest open-source medical AI skills library. 869 curated skills covering clinical, genomics, drug discovery, bioinformatics. Connected to PubMed, ClinicalTrials.gov, ChEMBL, DrugBank.
- [ClawBio](https://github.com/ClawBio/ClawBio): Bioinformatics-native AI agent skill library. 24 skills + 8,000 Galaxy tools. Local-first, reproducible. Features pharmacogenomics, GWAS lookup, polygenic risk scores, UK Biobank queries.
- [SkillHub](https://skillhub.tencent.com): Tencent official OpenClaw skills community, optimized for Chinese users. Browse, share, and install curated AI agent skills with WeChat integration.
- [STELLA](https://github.com/zaixizhang/STELLA): Self-Evolving LLM Agent for Biomedical Research. Autonomous multimodal agent platform from Princeton, Stanford, UCLA. Performs biomedical research tasks without API keys. Published on bioRxiv.
- [ARIS (Auto-Research-In-Sleep)](https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep): Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework lock-in — works with Claude Code, Codex, OpenClaw.
- [Antigravity Awesome Skills](https://github.com/sickn33/antigravity-awesome-skills): 1000+ battle-tested agentic skills for Claude Code, Antigravity, and Cursor. Includes official skills from Anthropic and Vercel. Covers engineering, analysis, and more.
- [Awesome Agent Skills](https://github.com/VoltAgent/awesome-agent-skills): 500+ agent skills from official dev teams and community, compatible with Claude Code, Codex, Antigravity, Gemini CLI, Cursor, and more.
- [Claude Skills Collection](https://github.com/alirezarezvani/claude-skills): 192 Claude Code skills and agent plugins for Claude Code, Codex, Gemini CLI, Cursor, and 8+ coding agents. Covers engineering, marketing, product, compliance, and advisory.
- [Baoyu Skills](https://github.com/jimliu/baoyu-skills): Curated collection of high-quality Claude Code skills by Baoyu. 9.6K+ stars. Covers coding, writing, research, and productivity workflows.
- [Skills.sh](https://skills.sh): The Agent Skills Directory by Vercel. 89,000+ skills for 20+ AI agents (Claude Code, Copilot, Cursor, Gemini, Codex). One-command install: npx skills add <repo>.
- [Skills.lc](https://github.com/Harries/skills-desktop): The open ecosystem for AI agent skills. 250K+ skills with 114M+ installs. Discover, share, and install reusable skills for Claude Code, Codex, Cursor, Windsurf, and Roo Code.