# AgentX

> A community directory, comparison and review platform for scientific research AI agents. Verified Run reviews by researchers who actually ran the tools. 208 agents across 10 categories, with live GitHub metrics and evidence-backed community reviews.

## Pages

- [Browse agents](https://agentx.webioinfo.top/agents): search and filter the full registry
- [Compare agents](https://agentx.webioinfo.top/compare): up to four agents side by side
- [Ecosystem reports](https://agentx.webioinfo.top/reports): monthly registry growth and activity
- [Same-name groups](https://agentx.webioinfo.top/samename): disambiguation for colliding project names
- [Contribute](https://agentx.webioinfo.top/contribute): how to get an agent listed
- [About](https://agentx.webioinfo.top/about): review policy and evidence rules
- [Data sources](https://agentx.webioinfo.top/data-sources): where the records come from
- [API docs](https://agentx.webioinfo.top/developers): public JSON API reference

## API

- https://agentx.webioinfo.top/api/agents: the full directory as JSON (q, category, status, limit params)
- https://agentx.webioinfo.top/api/agents/{slug}: one agent with its rating summary

## Agents by category

### Autonomous Research (47)

- [autoresearch](https://agentx.webioinfo.top/agents/autoresearch): AI agents running research on single-GPU nanochat training automatically [karpathy/autoresearch · Python · 96.2k stars · stale]
- [DeepResearch](https://agentx.webioinfo.top/agents/deepresearch): Tongyi Deep Research, the Leading Open-source Deep Research Agent [Alibaba-NLP/DeepResearch · Python · 20.0k stars · stable · paper: arXiv]
- [ARIS](https://agentx.webioinfo.top/agents/aris): ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works w… [wanshuiyin/Auto-claude-code-research-in-sleep · Python · 16.2k stars · active]
- [RD-Agent](https://agentx.webioinfo.top/agents/rd-agent): Research and development (R&D) is crucial for the enhancement of industrial productivity, especially in the AI era, where the core aspects of R&D are mainly focused on data and models. We are committ… [microsoft/RD-Agent · Python · 14.7k stars · active]
- [AI-Scientist (v1)](https://agentx.webioinfo.top/agents/ai-scientist-v1): The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 🧑‍🔬 [SakanaAI/AI-Scientist · Jupyter Notebook · 14.6k stars · stable · paper: arXiv]
- [AutoResearchClaw](https://agentx.webioinfo.top/agents/autoresearchclaw): Fully autonomous & self-evolving research from idea to paper. Chat an Idea. Get a Paper. 🦞 [aiming-lab/AutoResearchClaw · Python · 14.4k stars · active]
- [pi-autoresearch](https://agentx.webioinfo.top/agents/pi-autoresearch): Autonomous experiment loop extension for pi [davebcn87/pi-autoresearch · TypeScript · 8.1k stars · active]
- [AI-Scientist-v2](https://agentx.webioinfo.top/agents/ai-scientist-v2): The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search [SakanaAI/AI-Scientist-v2 · Python · 7.2k stars · stable · paper: Nature]
- [autoresearch (uditgoenka)](https://agentx.webioinfo.top/agents/autoresearch-uditgoenka): Claude Autoresearch Skill — Autonomous goal-directed iteration for Claude Code. Inspired by Karpathy's autoresearch. Modify → Verify → Keep/Discard → Repeat forever. [uditgoenka/autoresearch · Shell · 6.3k stars · active]
- [AgentLaboratory](https://agentx.webioinfo.top/agents/agentlaboratory): Agent Laboratory is an end-to-end autonomous research workflow meant to assist you as the human researcher toward implementing your research ideas [SamuelSchmidgall/AgentLaboratory · Python · 5.9k stars · stable · paper: Findings of EMNLP 2025]
- [AI-Researcher](https://agentx.webioinfo.top/agents/ai-researcher): [NeurIPS2025] "AI-Researcher: Autonomous Scientific Innovation" -- A production-ready version: https://novix.science/chat [HKUDS/AI-Researcher · Python · 5.7k stars · stable · paper: NeurIPS]
- [EvoScientist](https://agentx.webioinfo.top/agents/evoscientist): 🔬 Harness Vibe Research with Self-evolving AI Scientists [EvoScientist/EvoScientist · Python · 4.9k stars · stable · paper: arXiv]
- [OpenScience](https://agentx.webioinfo.top/agents/openscience): The open-source AI workbench for scientific research [synthetic-sciences/openscience · TypeScript · 3.6k stars · active]
- [DeepScientist](https://agentx.webioinfo.top/agents/deepscientist): Now, Stronger AI Pushes Frontiers, Stronger Our Shared Future. [ResearAI/DeepScientist · TypeScript · 3.3k stars · stable · paper: ICLR]
- [FAROS](https://agentx.webioinfo.top/agents/faros): A blueprint-driven AutoResearch runtime for orchestrating AI research workflows from idea generation and experiments to paper writing and peer review. [OpenNSWM-Lab/FAROS · Python · 3.0k stars · active]
- [autoresearch-mlx](https://agentx.webioinfo.top/agents/autoresearch-mlx): Apple Silicon (MLX) port of Karpathy's autoresearch — autonomous AI research loops on Mac, no PyTorch required. [trevin-creator/autoresearch-mlx · Python · 1.8k stars · active]
- [DATAGEN](https://agentx.webioinfo.top/agents/datagen): DATAGEN: AI-driven multi-agent research assistant automating hypothesis generation, data analysis, and report writing. [starpig1129/DATAGEN · Python · 1.8k stars · active]
- [Auto-Deep-Research](https://agentx.webioinfo.top/agents/auto-deep-research): "Your Fully-Automated Personal AI Assistant" [HKUDS/Auto-Deep-Research · Python · 1.7k stars · stable · paper: Findings of ACL 2026]
- [InternAgent](https://agentx.webioinfo.top/agents/internagent): InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery [InternScience/InternAgent · Python · 1.4k stars · active]
- [NanoResearch](https://agentx.webioinfo.top/agents/nanoresearch): 🦞+🔬 NanoResearch: The Autonomous AI Research Assistant [OpenRaiser/NanoResearch · Python · 1.4k stars · active]
- [OpenResearcher](https://agentx.webioinfo.top/agents/openresearcher): OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis [TIGER-AI-Lab/OpenResearcher · Python · 1.2k stars · active]
- [Dr. Claw](https://agentx.webioinfo.top/agents/dr-claw): A Super AI Lab with massive AI Doctors as Assistants. Best IDE for Research via AI Power. [OpenLAIR/dr-claw · JavaScript · 1.1k stars · active]
- [ScienceClaw (beita6969)](https://agentx.webioinfo.top/agents/scienceclaw-beita6969): 🔬🦞 A self-evolving AI research colleague for scientists. 285 skills, zero hallucination, persistent memory. [beita6969/ScienceClaw · TypeScript · 903 stars · active]
- [data-to-paper](https://agentx.webioinfo.top/agents/data-to-paper): data-to-paper: Backward-traceable AI-driven scientific research [Technion-Kishony-lab/data-to-paper · Python · 826 stars · stable · paper: NEJM AI]
- [freephdlabor](https://agentx.webioinfo.top/agents/freephdlabor): freephdlabor: customizing personalized multiagent systems that researchs 24/7 on your own scientific problem [ltjed/freephdlabor · Python · 723 stars · active]
- [EurekaClaw](https://agentx.webioinfo.top/agents/eurekaclaw): The official repo of EurekaClaw [EurekaClaw/EurekaClaw · Python · 697 stars · active]
- [ScienceClaw (TaichuAI)](https://agentx.webioinfo.top/agents/scienceclaw-taichuai): ScienceClaw is a personal research assistant built with LangChain DeepAgents and AIO Sandbox infrastructure, adopting a completely new architecture beyond OpenClaw. It offers stronger security, bette… [AgentTeam-TaichuAI/ScienceClaw · Python · 666 stars · stale]
- [SciAgents](https://agentx.webioinfo.top/agents/sciagents) [lamm-mit/SciAgentsDiscovery · Python · 639 stars · stable · paper: Advances in Materials]
- [Kosmos](https://agentx.webioinfo.top/agents/kosmos): Kosmos: An AI Scientist for Autonomous Discovery - An implementation and adaptation to be driven by Claude Code or API - Based on the Kosmos AI Paper - https://arxiv.org/abs/2511.02824 [jimmc414/Kosmos · Python · 585 stars · stale · paper: arXiv]
- [InnoClaw](https://agentx.webioinfo.top/agents/innoclaw): An AI research Agent for scientific innovation. [SpectrAI-Initiative/InnoClaw · TypeScript · 393 stars · active]
- [CodeScientist](https://agentx.webioinfo.top/agents/codescientist): CodeScientist: An automated scientific discovery system for code-based experiments [allenai/codescientist · Python · 352 stars · stale]
- [OpenLens AI](https://agentx.webioinfo.top/agents/openlens-ai): OpenLens AI: A Fully Autonomous Multimodal Research Agent｜ OpenLens AI：全自主多模态科研智能体 [jarrycyx/openlens-ai · Python · 282 stars · active]
- [Co-Scientist](https://agentx.webioinfo.top/agents/co-scientist): Open Source Reimplementation of Google Deepmind's Co-Scientist [Kaimen-Inc/Co-Scientist · Python · 260 stars · active]
- [ScienceClaw (MIT LAMM)](https://agentx.webioinfo.top/agents/scienceclaw-mit-lamm) [lamm-mit/scienceclaw · Python · 242 stars · active · paper: arXiv]
- [OmniScientist](https://agentx.webioinfo.top/agents/omniscientist) [tsinghua-fib-lab/OmniScientist · 202 stars · stale]
- [DrClaw (InternScience)](https://agentx.webioinfo.top/agents/drclaw-internscience) [InternScience/DrClaw · Python · 171 stars · stale]
- [NeuriCo](https://agentx.webioinfo.top/agents/neurico): AI Scientist by Chicago Human+AI Lab [ChicagoHAI/NeuriCo · Python · 156 stars · active]
- [SciClaw (drpedapati)](https://agentx.webioinfo.top/agents/sciclaw-drpedapati): Paired-scientist AI assistant for reproducible research — lightweight Go runtime, lifecycle hooks, manuscript integration, 12 baseline scientific skills. PicoClaw-compatible. [drpedapati/sciclaw · Go · 88 stars · active]
- [ScienceClaw (Zaoqu-Liu)](https://agentx.webioinfo.top/agents/scienceclaw-zaoqu-liu): Your AI Research Lab That Never Sleeps. 9 agents, 263 skills, 77 databases — from literature to publication, every discipline, zero boundaries. [Zaoqu-Liu/ScienceClaw · Shell · 60 stars · stale]
- [Open Co-Scientist](https://agentx.webioinfo.top/agents/open-co-scientist): Open Coscientist is an open-source adaptation based on Google Research's AI Co-Scientist research paper. We provide an open implementation that generates, reviews, ranks, and evolves research hypothe… [jataware/open-coscientist · Python · 60 stars · stale]
- [SR-Scientist](https://agentx.webioinfo.top/agents/sr-scientist): [ICLR 2026] SR-Scientist: Scientific Equation Discovery With Agentic AI [GAIR-NLP/SR-Scientist · Python · 56 stars · stale · paper: arXiv]
- [XunZi](https://agentx.webioinfo.top/agents/xunzi) [biocuckooHXH/XunZi · Python · 36 stars · stable · paper: Nature-BME]
- [Agentomics](https://agentx.webioinfo.top/agents/agentomics): Autonomous agentic system for supervised machine learning model development. [BioGeMT/agentomics-ml · Python · 34 stars · active · paper: bioRxiv]
- [LabOS](https://agentx.webioinfo.top/agents/labos): Official Code of LabOS: the AI-XR Co-Scientist that Sees and Works with Humans [zaixizhang/LabOS · Python · 33 stars · stale · paper: arXiv]
- [OpenCLAW Research Platform](https://agentx.webioinfo.top/agents/openclaw-research-platform): OpenCLAW is an ambitious open-source project that transforms the OpenClaw personal AI assistant framework into a fully autonomous, self-improving, multi-agent scientific research platform. The vision… [Agnuxo1/OpenCLAW-Autonomous-Multi-Agent-Scientific-Research-Platform · JavaScript · 10 stars · archived]
- [SciClaw (nahisaho)](https://agentx.webioinfo.top/agents/sciclaw-nahisaho): SciClaw — AI research assistant with Docker sandbox agents [nahisaho/sciclaw · TypeScript · 2 stars · stale]
- [Claude Science](https://agentx.webioinfo.top/agents/claude-science): Anthropic's closed-source AI workbench for scientists — a coordinating agent with 60+ curated skills and connectors across genomics, single-cell, proteomics, structural biology and cheminformatics, w… [anthropic.com/claude-science · 0 stars · no-repo]

### Literature & Scientific Writing (18)

- [Research-Claw](https://agentx.webioinfo.top/agents/research-claw): In the AI era, why can't everyone be a Principal Investigator? [wentorai/Research-Claw · TypeScript · 850 stars · active]
- [Prismer](https://agentx.webioinfo.top/agents/prismer): Open Source Research Claw [Prismer-AI/Prismer · TypeScript · 794 stars · active]
- [ResearchClaw (ymx10086)](https://agentx.webioinfo.top/agents/researchclaw-ymx10086): ResearchClaw is a personal AI assistant built for research: fast to set up, easy to run locally or in the cloud, and ready to integrate with the chat apps you already use. With extensible skills, it… [ymx10086/ResearchClaw · Python · 313 stars · stale]
- [CitationClaw](https://agentx.webioinfo.top/agents/citationclaw): 让每一次引用都成为可解释的影响力 Turning Every Citation into Explainable Impact [VisionXLab/CitationClaw · Python · 313 stars · active]
- [research-claw (nanoAgentTeam)](https://agentx.webioinfo.top/agents/research-claw-nanoagentteam): A self-hosted AI assistant for academic research — manages your papers, searches literature, tracks deadlines, and answers you on the channels you already use. [nanoAgentTeam/research-claw · Python · 292 stars · active]
- [PaperClaw (guhaohao0991)](https://agentx.webioinfo.top/agents/paperclaw-guhaohao0991): an OpenClaw skill that can generate paper search-review-critque expert-agent relevant to specific topics (we use Scientific ML and 3D geometry surrogate modeling as a demo). [guhaohao0991/PaperClaw · Python · 248 stars · stale]
- [RS-PaperClaw](https://agentx.webioinfo.top/agents/rs-paperclaw): 🦞 Stop reading remote sensing papaers manually — RS-PaperClaw auto-tracks, auto-analyzes, and auto-digests remote sensing papers. [thinson/RS-PaperClaw · Python · 212 stars · active]
- [ClawPhD](https://agentx.webioinfo.top/agents/clawphd): ClawPhD is an agent for research that can turn academic papers into publication-ready diagrams, posters, videos, and more. [ZhihaoAIRobotic/ClawPhD · Python · 157 stars · stale]
- [PaperClaw (meowscles69)](https://agentx.webioinfo.top/agents/paperclaw-meowscles69): 27 OpenClaw skills for academic research teams — literature reviews, hypothesis versioning. GitHub account deleted; last known state ★141. [meowscles69/PaperClaw · 141 stars · gone]
- [PaperClaw (1692775560)](https://agentx.webioinfo.top/agents/paperclaw-1692775560) [1692775560/PaperClaw · Python · 70 stars · stale]
- [ResearchClaw (Noietch)](https://agentx.webioinfo.top/agents/researchclaw-noietch): AI-Powered Research Desktop App [Noietch/ResearchClaw · TypeScript · 65 stars · gone]
- [paper_claw (PigeonDan1)](https://agentx.webioinfo.top/agents/paper_claw-pigeondan1): Paper Claw sends personalized daily research digests from arXiv and beyond straight to your inbox, featuring customizable categories, intelligent classification, and agent-based multilingual summarie… [PigeonDan1/paper_claw · Python · 37 stars · active]
- [PaperClaw (AkaliKong)](https://agentx.webioinfo.top/agents/paperclaw-akalikong) [AkaliKong/PaperClaw · Python · 22 stars · archived]
- [MetaBeeAI](https://agentx.webioinfo.top/agents/metabeeai): Main MetaBeeAI pipeline - tested for bees & pesticides literature [MetaBeeAI/MetaBeeAI · Python · 9 stars · active · paper: bioRxiv]
- [PaperClaw (0xMerl99)](https://agentx.webioinfo.top/agents/paperclaw-0xmerl99) [0xMerl99/PaperClaw · 7 stars · archived]
- [TeroSeek](https://agentx.webioinfo.top/agents/teroseek): An intelligent dialogue platform exclusively dedicated to terpenoids research [qmclab/teroseek · 4 stars · stable · paper: Plant-Communications]
- [PaperClaw (hurtjan)](https://agentx.webioinfo.top/agents/paperclaw-hurtjan) [hurtjan/PaperClaw · Python · 4 stars · stale]
- [PaperClaw (cocoafun)](https://agentx.webioinfo.top/agents/paperclaw-cocoafun) [cocoafun/paperclaw · TypeScript · 2 stars · archived]

### Bioinformatics & Omics (53)

- [Biomni](https://agentx.webioinfo.top/agents/biomni): Biomni: a general-purpose biomedical AI agent [snap-stanford/Biomni · Python · 3.9k stars · stable · paper: Science]
- [OmicVerse](https://agentx.webioinfo.top/agents/omicverse): A python library for multi omics included bulk, single cell and spatial RNA-seq analysis. [omicverse/omicverse · Python · 1.2k stars · stable · paper: bioRxiv]
- [Wisp Science](https://agentx.webioinfo.top/agents/wisp-science): Open-source, local-first desktop AI research workbench for scientific computing with Python/R, MCP bioinformatics tools, SSH/WSL/GPU runtimes, and OpenAI/Anthropic models. [xuzhougeng/wisp-science · Rust · 1.1k stars · active]
- [ClawBio](https://agentx.webioinfo.top/agents/clawbio): 🦖 ClawBio - The first bioinformatics-native AI agent skill library. Local-first. Reproducible. Open. Free. [ClawBio/ClawBio · Python · 1.1k stars · active]
- [mLLMCelltype](https://agentx.webioinfo.top/agents/mllmcelltype): Cell type annotation for single-cell RNA-seq using multi-LLM consensus [cafferychen777/mLLMCelltype · Python · 661 stars · stable · paper: Communications-Biology]
- [PantheonOS](https://agentx.webioinfo.top/agents/pantheonos): A general, evolvable, and distributed agent framework & harness for data science. [aristoteleo/PantheonOS · Python · 485 stars · active]
- [BioClaw](https://agentx.webioinfo.top/agents/bioclaw): AI-Powered Bioinformatics Research Assistant. Built on OpenClaw. [Runchuan-BU/BioClaw · TypeScript · 408 stars · active]
- [CellVoyager](https://agentx.webioinfo.top/agents/cellvoyager) [zou-group/CellVoyager · Python · 263 stars · stable · paper: Nature-Methods]
- [AutoBA](https://agentx.webioinfo.top/agents/autoba): An AI Agent for Fully Automated Multi-omic Analyses [JoshuaChou2018/AutoBA · Python · 238 stars · stale · paper: bioRxiv]
- [CellWhisperer](https://agentx.webioinfo.top/agents/cellwhisperer): CellWhisperer bridges the gap between transcriptomics data and natural language, enabling intuitive interaction with scRNA-seq datasets [epigen/cellwhisperer · Jupyter Notebook · 220 stars · stable · paper: Nature-Biotechnology]
- [SpatialAgent](https://agentx.webioinfo.top/agents/spatialagent): A harness for spatial biology [Genentech/SpatialAgent · Python · 211 stars · stale · paper: bioRxiv]
- [BioAgents](https://agentx.webioinfo.top/agents/bioagents): BioAgents: An AI scientist framework for autonomous deep research in biological sciences. Multi-agent system combining literature analysis agents with data scientist agents to enable deep research au… [bio-xyz/BioAgents · TypeScript · 184 stars · gone · paper: arXiv]
- [CRISPR-GPT](https://agentx.webioinfo.top/agents/crispr-gpt): CRISPSR-GPT LLM Agent Public Welcome Page [cong-lab/crispr-gpt-pub · Python · 176 stars · stale · paper: bioRxiv]
- [OmicsClaw](https://agentx.webioinfo.top/agents/omicsclaw): Conversational & memory-enabled AI research partner for multi-omics analysis. CLI + Desktop App (installers in Releases). From biological idea to full research paper. [TianGzlab/OmicsClaw · Python · 161 stars · active]
- [STELLA](https://agentx.webioinfo.top/agents/stella): STELLA: Self-Evolving LLM Agent for Biomedical Research [zaixizhang/STELLA · Python · 154 stars · active · paper: arXiv]
- [BioMedAgent](https://agentx.webioinfo.top/agents/biomedagent) [BOBQWERA/BioMedAgent · Python · 138 stars · stable · paper: Nature-BME]
- [BioDiscoveryAgent](https://agentx.webioinfo.top/agents/biodiscoveryagent): BioDiscoveryAgent is an LLM-based AI agent for closed-loop design of genetic perturbation experiments [snap-stanford/BioDiscoveryAgent · Python · 127 stars · stable · paper: ICLR]
- [BioMaster](https://agentx.webioinfo.top/agents/biomaster) [ai4nucleome/BioMaster · TypeScript · 113 stars · active · paper: bioRxiv]
- [CASSIA](https://agentx.webioinfo.top/agents/cassia): CASSIA: A Multi-Agent LLM-Based Single-Cell Cell Type Annotation Framework [ElliotXie/CASSIA · Python · 106 stars · stable · paper: Nature-Communications]
- [BioDSA](https://agentx.webioinfo.top/agents/biodsa): BioDSA: Framework for Vibe Prototyping of AI Agents for Biomedicine [RyanWangZf/BioDSA · Python · 93 stars · stable · paper: Nature-BME]
- [NeuroClaw](https://agentx.webioinfo.top/agents/neuroclaw) [CUHK-AIM-Group/NeuroClaw · Python · 87 stars · active]
- [SCREADER](https://agentx.webioinfo.top/agents/screader): The screader is a soure code reading tool based the libclang. It is implemented by the C. [shining1984/screader · C · 74 stars · stable · paper: ICDMW]
- [BRAD](https://agentx.webioinfo.top/agents/brad): Bioinformatics Retrieval Augmented Digital (BRAD) Assistant - A LLM powered agent for bioinformatics 🤖 [Jpickard1/BRAD · Python · 64 stars · stable · paper: Bioinformatics]
- [Bioinfor-Claw](https://agentx.webioinfo.top/agents/bioinfor-claw): Your 24/7 bioinformatics copilot who make your analysis easy [MDhewei/bioinfor-claw · HTML · 64 stars · stale]
- [scExtract](https://agentx.webioinfo.top/agents/scextract): Automatic annotation of single-cell RNA-seq data from the literature [yxwucq/scExtract · Python · 48 stars · stable · paper: Genome-Biology]
- [Lobster AI](https://agentx.webioinfo.top/agents/lobster-ai): The self-evolving agentic framework for bioinformatics [the-omics-os/lobster · Python · 46 stars · active]
- [InstructCell](https://agentx.webioinfo.top/agents/instructcell): A Multi-Modal AI Copilot for Single-Cell Analysis with Instruction Following [zjunlp/InstructCell · Jupyter Notebook · 33 stars · stable · paper: arXiv]
- [SCassist](https://agentx.webioinfo.top/agents/scassist): SCassist : An AI Based Workflow Assistant for Single-Cell Analysis [NIH-NEI/SCassist · HTML · 33 stars · stable · paper: Bioinformatics]
- [OmicClaw](https://agentx.webioinfo.top/agents/omicclaw): omicverse agent skill [omicverse/omicclaw · JavaScript · 33 stars · active]
- [MolClaw](https://agentx.webioinfo.top/agents/molclaw): Light, Practical, memory-enabled AI orchestrator for science. Built on NanoClaw. [IDEA-XL/MolClaw · TypeScript · 30 stars · stale]
- [OLAF](https://agentx.webioinfo.top/agents/olaf): The Open Life Science Analysis Platform [OpenTechBio/Olaf · Python · 26 stars · archived · paper: arXiv]
- [Opentest](https://agentx.webioinfo.top/agents/opentest): Claw-like Agent for bioinformatics [AndyZhuang/Opentest · Python · 22 stars · archived]
- [ProtChat](https://agentx.webioinfo.top/agents/protchat) [SIAT-code/ProtChat · Python · 19 stars · stable · paper: JCIM]
- [IAN](https://agentx.webioinfo.top/agents/ian): IAN: An Intelligent System for Omics Data Analysis and Discovery [NIH-NEI/IAN · R · 18 stars · stable · paper: Cell-Reports-Methods]
- [PopGenAgent](https://agentx.webioinfo.top/agents/popgenagent) [ai4nucleome/POPGENAGENT · Python · 15 stars · archived]
- [AstroAgents](https://agentx.webioinfo.top/agents/astroagents): AstroAgents: Multi-Agent AI for Hypothesis Generation from Mass Spectrometry Data [amirgroup-codes/AstroAgents · Python · 15 stars · archived · paper: arXiv]
- [GPTBioInsightor](https://agentx.webioinfo.top/agents/gptbioinsightor): Gain biological insights using LLM. [huang-sh/GPTBioInsightor · Python · 14 stars · stable · paper: Bioinformatics-Advances]
- [Bio-Copilot](https://agentx.webioinfo.top/agents/bio-copilot) [lyyang01/Bio-Copilot · Python · 13 stars · stable · paper: Briefings-in-Bioinformatics]
- [scChat](https://agentx.webioinfo.top/agents/scchat) [li-group/scChat · Python · 13 stars · active · paper: bioRxiv]
- [EpiClaw](https://agentx.webioinfo.top/agents/epiclaw) [GBA-BI/EpiClaw · Python · 10 stars · stale]
- [scAgent](https://agentx.webioinfo.top/agents/scagent): This is the repository of scAgent. [Miiyuu/scAgent · Python · 8 stars · archived · paper: arXiv]
- [Darwin](https://agentx.webioinfo.top/agents/darwin): Darwin is a biomedical generative agent forked from and heavily modified on top of feynman [yejunbin/darwin · TypeScript · 7 stars · stale]
- [Nano Bio-Agents](https://agentx.webioinfo.top/agents/nanobioagent): Nano Bio-Agent (NBA): SLM agent for Genomics [georgesshong/nanobioagent · Python · 6 stars · archived · paper: arXiv]
- [CARIBOU](https://agentx.webioinfo.top/agents/caribou): Computational AI Research Interface for Bioinformatics, Omics, and Unifying Agents [rose-research-group/CARIBOU · Jupyter Notebook · 6 stars · active]
- [MErlin](https://agentx.webioinfo.top/agents/merlin): Stepwise grimoire designed as a modular workflow PacBio/ONT methylome data: from raw reads to advanced statistics and multi-omics crosstalk. [IacopoPasseri/MErlin · 4 stars · active · paper: bioRxiv]
- [CellTypeAgent](https://agentx.webioinfo.top/agents/celltypeagent): CellTypeAgent: Trustworthy cell type annotation with Large Language Models [jianghao-zhang/CellTypeAgent · Python · 4 stars · archived · paper: arXiv]
- [MetaClaw (BGI)](https://agentx.webioinfo.top/agents/metaclaw-bgi): MetaClaw: an auditable AI agent for end-to-end, multi-directional shotgun metagenomic analysis [BGI-AI-models/MetaClaw · Python · 3 stars · active · paper: bioRxiv]
- [MetaMuse](https://agentx.webioinfo.top/agents/metamuse) [Ekansh-Mittal1/MetaMuse · Python · 3 stars · active · paper: bioRxiv]
- [cGSA](https://agentx.webioinfo.top/agents/cgsa) [ncbi-nlp/cGSA · Python · 2 stars · archived]
- [BioOS-Claws](https://agentx.webioinfo.top/agents/bioos-claws) [GBA-BI/BioOS-Claws · JavaScript · 1 stars · stale]
- [autoBioResearch](https://agentx.webioinfo.top/agents/autobioresearch) [JMGamma/autoBioResearch · Python · 1 stars · stale]
- [BioBrain](https://agentx.webioinfo.top/agents/biobrain) [Wendyzzz1/BioBrain · Python · 0 stars · archived]
- [CLAW-MRM](https://agentx.webioinfo.top/agents/claw-mrm) [chopralab/CLAW_MRM · Jupyter Notebook · 0 stars · stable · paper: Analytical-Chemistry]

### Chemistry & Drug Discovery (25)

- [ChemCrow](https://agentx.webioinfo.top/agents/chemcrow): Chemcrow [ur-whitelab/chemcrow-public · Python · 957 stars · stale]
- [BioNeMo Framework](https://agentx.webioinfo.top/agents/bionemo-framework): BioNeMo Recipes: For building and adapting AI models in drug discovery at scale [NVIDIA/bionemo-framework · Python · 857 stars · active · paper: Nvidia's blog]
- [Virtual Lab](https://agentx.webioinfo.top/agents/virtual-lab): A virtual lab of LLM agents for science research [zou-group/virtual-lab · Jupyter Notebook · 735 stars · stable · paper: Nature]
- [TxAgent](https://agentx.webioinfo.top/agents/txagent): TxAgent: An AI Agent for Therapeutic Reasoning Across a Universe of Tools [mims-harvard/TxAgent · Python · 653 stars · stale · paper: arXiv]
- [BloClaw](https://agentx.webioinfo.top/agents/bloclaw) [qinheming/BIoClaw · Python · 378 stars · stale · paper: arXiv]
- [MDCrow](https://agentx.webioinfo.top/agents/mdcrow): Molecular dynamics simulations with an LLM agent [ur-whitelab/MDCrow · Jupyter Notebook · 247 stars · stale]
- [OriGene](https://agentx.webioinfo.top/agents/origene): OriGene: A Self-Evolving Virtual Disease Biologist for Mechanism-Guided Therapeutic Target Discovery [GENTEL-lab/OriGene · Python · 225 stars · stale · paper: bioRxiv]
- [ChemGraph](https://agentx.webioinfo.top/agents/chemgraph): Agentic framework for computational chemistry and materials science workflows [argonne-lcf/ChemGraph · Python · 157 stars · active]
- [DrugClaw (DrugClaw)](https://agentx.webioinfo.top/agents/drugclaw-drugclaw): 💊 AI Research Assistant for Accelerated Drug Discovery. 🦞 [DrugClaw/DrugClaw · Rust · 125 stars · stale]
- [DrugClaw (QSong)](https://agentx.webioinfo.top/agents/drugclaw-qsong): 🦀 Agentic RAG for drug intelligence · 57 skills · 15 task categories · DTI · ADR · DDI · PGx · Repurposing · Powered by LangGraph [QSong-github/DrugClaw · Python · 116 stars · active]
- [AIAgents4Pharma](https://agentx.webioinfo.top/agents/aiagents4pharma): AI Agents for drug discovery, drug development, and other pharmaceutical R&D [VirtualPatientEngine/AIAgents4Pharma · Python · 100 stars · stale]
- [MatClaw](https://agentx.webioinfo.top/agents/matclaw): MatClaw: an open materials-science agent that turns natural-language tasks into reproducible simulation workflows. [DingyangLyu/MatClaw · TypeScript · 92 stars · stale]
- [ProtAgents](https://agentx.webioinfo.top/agents/protagents) [lamm-mit/ProtAgents · Jupyter Notebook · 53 stars · stable · paper: Digital-Discovery]
- [ChemClaw](https://agentx.webioinfo.top/agents/chemclaw): ChemClaw — The first chemistry-native AI agent skill library. [InternScience/ChemClaw · Jupyter Notebook · 52 stars · stale]
- [DrugDiscoveryBench](https://agentx.webioinfo.top/agents/drugdiscoverybench): Opensource repository containing task and image data for DrugDiscoveryBench [scaleapi/DrugDiscoveryBench · Python · 29 stars · stable · paper: Technical report]
- [DrugPilot](https://agentx.webioinfo.top/agents/drugpilot): A LLM-based Agent Framework for Drug Discovery [wzn99/DrugPilot · Jupyter Notebook · 23 stars · archived · paper: arXiv]
- [TCM-Agent](https://agentx.webioinfo.top/agents/tcm-agent) [AITCM/TCM-Agent · 23 stars · active]
- [LIDDIA](https://agentx.webioinfo.top/agents/liddia) [ninglab/LIDDIA · Python · 9 stars · archived · paper: arXiv]
- [MutexaGPT](https://agentx.webioinfo.top/agents/mutexagpt): A web application that serves as an interface between a user and EnzyHTP's workflows. [ChemBioHTP/EnzyHTP-GPT · Python · 7 stars · stable · paper: Nature-Computational-Science]
- [DrugEvolve](https://agentx.webioinfo.top/agents/drugevolve): An AI System for Autonomous Algorithm Evolution in Drug Development [GAIR-NLP/DrugEvolve · Python · 5 stars · active · paper: bioRxiv]
- [PyMolClaw](https://agentx.webioinfo.top/agents/pymolclaw): Comprehensive molecular visualization and analysis skill powered by PyMOL. One description → publication-quality figure + analysis. [BioTender-max/PyMolClaw · Python · 4 stars · stale]
- [ChatMolData](https://agentx.webioinfo.top/agents/chatmoldata): A multimodal agent for automatic molecular data processing [YiYuDL/ChatMolData · Python · 4 stars · stable · paper: Advanced-Intelligent-Systems]
- [AgentMol](https://agentx.webioinfo.top/agents/agentmol): AgentMol: multimodel AI system for automatic drug-target identification and molecule development. [golempharm/agentmol · Python · 1 stars · stable · paper: Methods-and-Protocols]
- [LLMsFold](https://agentx.webioinfo.top/agents/llmsfold): Repository for the LLMsFold [tacclab/LLMsFold · Python · 1 stars · active · paper: bioRxiv]
- [Pepti-Agent](https://agentx.webioinfo.top/agents/pepti-agent): AI agent for peptide design: multi-agent pipeline with PDB, UniProt, conservation, Boltz structure prediction, and literature tools. [BioTender-max/peptide-agent · Python · 0 stars · active · paper: arXiv]

### Clinical & Healthcare (7)

- [MedgeClaw](https://agentx.webioinfo.top/agents/medgeclaw): Open-source AI research assistant for biomedicine — chat to run RNA-seq, drug discovery, clinical analysis, and more. Built on Claude Code with 140 K-Dense scientific skills, real-time dashboard, and… [xjtulyc/MedgeClaw · TeX · 641 stars · stale]
- [HealthClaw](https://agentx.webioinfo.top/agents/healthclaw): HealthClaw: Open-source Self-Evolving Personal Health Copilot [HC-Guo/HealthClaw · Python · 151 stars · active]
- [MedClaw (zteyesreal)](https://agentx.webioinfo.top/agents/medclaw-zteyesreal) [zteyesreal/MedClaw · Python · 67 stars · stale]
- [MedClaw (Paulzhang)](https://agentx.webioinfo.top/agents/medclaw-paulzhang): 为医疗领域定制的小龙虾（open claw) [Paulzhang2023/medclaw · TypeScript · 14 stars · active]
- [HemaGuide](https://agentx.webioinfo.top/agents/hemaguide): Clinical decision support in hematological malignancies using a case-grounded AI agent [Friedrich-Lab/HemaGuide · Python · 9 stars · stable · paper: Nature Medicine]
- [ChatRadio-Valuer](https://agentx.webioinfo.top/agents/chatradio-valuer) [DPX-CN/ChatRadio-Valuer · 1 stars · stable · paper: IEEE-TBME]
- [EAS-Diagnosis](https://agentx.webioinfo.top/agents/eas-diagnosis): Code for the paper: Integrating clinical reasoning into large language model-based diagnosis through etiology-aware attention steering [Twilight-sp/EAS-Diagnosis · 0 stars · archived · paper: arXiv]

### Platforms & Infrastructure (30)

- [OpenClaw](https://agentx.webioinfo.top/agents/openclaw): The AI that really does things. Any OS. Any Platform. The lobster way. 🦞 [openclaw/openclaw · TypeScript · 389.9k stars · active]
- [Hermes Agent](https://agentx.webioinfo.top/agents/hermes-agent): The agent that grows with you [NousResearch/hermes-agent · Python · 246.3k stars · active]
- [Pi](https://agentx.webioinfo.top/agents/pi): AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI [earendil-works/pi · TypeScript · 106.5k stars · active]
- [NanoBot](https://agentx.webioinfo.top/agents/nanobot): Ultra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, MCP, multi-agent workflows, automation, and chat apps [HKUDS/nanobot · Python · 48.3k stars · active]
- [ZeroClaw](https://agentx.webioinfo.top/agents/zeroclaw): Fast, small, and fully autonomous AI personal assistant infrastructure, any OS, any platform — deploy anywhere, swap anything 🦀 [zeroclaw-labs/zeroclaw · Rust · 32.9k stars · active]
- [NanoClaw](https://agentx.webioinfo.top/agents/nanoclaw): A lightweight alternative to OpenClaw that runs in containers for security. Connects to WhatsApp, Telegram, Slack, Discord, Gmail and other messaging apps,, has memory, scheduled jobs, and runs direc… [nanocoai/nanoclaw · TypeScript · 30.8k stars · active]
- [PicoClaw](https://agentx.webioinfo.top/agents/picoclaw): Tiny, Fast, and Deployable anywhere — automate the mundane, unleash your creativity [sipeed/picoclaw · Go · 30.0k stars · active]
- [NemoClaw](https://agentx.webioinfo.top/agents/nemoclaw): Run agents like Hermes, LangChain Deep Agents, and OpenClaw more securely inside NVIDIA OpenShell with managed inference [NVIDIA/NemoClaw · TypeScript · 22.5k stars · active]
- [SkillOpt](https://agentx.webioinfo.top/agents/skillopt): SkillOpt is a text-space optimizer that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits, validation-gated updates, and deployable best_skill.md artifacts. [microsoft/SkillOpt · Python · 17.2k stars · stable · paper: arXiv]
- [OpenHarness](https://agentx.webioinfo.top/agents/openharness): "OpenHarness: Open Agent Harness with a Built-in Personal Agent--Ohmo!" [HKUDS/OpenHarness · Python · 15.8k stars · active]
- [IronClaw](https://agentx.webioinfo.top/agents/ironclaw): IronClaw is an Agent OS focused on privacy, security and extensibility [nearai/ironclaw · Rust · 12.6k stars · active]
- [NullClaw](https://agentx.webioinfo.top/agents/nullclaw): Fastest, smallest, and fully autonomous AI assistant infrastructure written in Zig [nullclaw/nullclaw · Zig · 8.1k stars · active]
- [OpenSpace](https://agentx.webioinfo.top/agents/openspace): "OpenSpace: The Skill Management Layer for AI Agents" -- https://open-space.cloud/ [HKUDS/OpenSpace · Python · 7.7k stars · active]
- [ClawX](https://agentx.webioinfo.top/agents/clawx): ClawX is a desktop app that provides a graphical interface for OpenClaw AI agents. It turns CLI-based AI orchestration into a desktop experience without using the terminal. China website is https://c… [ValueCell-ai/ClawX · TypeScript · 7.6k stars · active]
- [autoagent](https://agentx.webioinfo.top/agents/autoagent): autonomous harness engineering [kevinrgu/autoagent · Python · 4.6k stars · stale]
- [TextGrad](https://agentx.webioinfo.top/agents/textgrad): TextGrad: Automatic ''Differentiation'' via Text -- using large language models to backpropagate textual gradients. Published in Nature. [zou-group/textgrad · Python · 3.7k stars · stable · paper: Nature]
- [SkillClaw](https://agentx.webioinfo.top/agents/skillclaw): Let Skills Evolve Collectively with Agentic Evolver [AMAP-ML/SkillClaw · Python · 2.6k stars · stable · paper: arXiv]
- [ClawManager](https://agentx.webioinfo.top/agents/clawmanager): A Kubernetes-native control plane for AI agent instance management, with governed AI access, runtime orchestration, and reusable resources across multiple agent runtimes. [Yuan-lab-LLM/ClawManager · Go · 1.9k stars · active]
- [ToolUniverse](https://agentx.webioinfo.top/agents/tooluniverse): Democratizing AI scientists with ToolUniverse [mims-harvard/ToolUniverse · Python · 1.7k stars · active]
- [Memento-Skills](https://agentx.webioinfo.top/agents/memento-skills): Memento-Skills: Let Agents Design Agents [Memento-Teams/Memento-Skills · Python · 1.6k stars · stable · paper: arXiv]
- [AlphaClaw](https://agentx.webioinfo.top/agents/alphaclaw): The ultimate setup harness for OpenClaw. Deploy in minutes. Stay running for months. No CLI required. [chrysb/alphaclaw · JavaScript · 1.5k stars · active]
- [FastClaw](https://agentx.webioinfo.top/agents/fastclaw): Multi-Agent Framework [fastclaw-ai/fastclaw · Go · 1.3k stars · active]
- [EdgeClaw](https://agentx.webioinfo.top/agents/edgeclaw): EdgeClaw: Edge-Cloud Collaborative Personal AI Assistant based on OpenClaw [OpenBMB/EdgeClaw · TypeScript · 1.2k stars · stale]
- [LabClaw](https://agentx.webioinfo.top/agents/labclaw): LabClaw – Operating Layer for LabOS (Stanford-Princeton AI Co-Scientists) [wu-yc/LabClaw · 1.1k stars · stale]
- [Alita](https://agentx.webioinfo.top/agents/alita) [CharlesQ9/Alita · 880 stars · stale · paper: arXiv]
- [MicroClaw](https://agentx.webioinfo.top/agents/microclaw): 🦀An agentic AI assistant that lives in your chats, inspired by nanoclaw and incorporating some of its design ideas. Built with Rust 🦀 [microclaw/microclaw · Rust · 735 stars · active]
- [StatsClaw](https://agentx.webioinfo.top/agents/statsclaw): A workflow framework for statistical package development [statsclaw/statsclaw · 93 stars · active · paper: https://yiqingxu.org/papers/2026_statsclaw/statsclaw.pdf]
- [GeneClaw](https://agentx.webioinfo.top/agents/geneclaw): Self-evolving AI agent framework with 5-layer safety gatekeeper. Agents observe failures, propose fixes, and safely apply them. Built on HKUDS/nanobot. [Clawland-AI/Geneclaw · Python · 45 stars · stale]
- [PythonClaw](https://agentx.webioinfo.top/agents/pythonclaw): 🐍 Personal AI agent in pure Python — OpenClaw reimagined. Memory, RAG, skills marketplace, cron, voice. Telegram/Discord/WhatsApp/Web. Works with DeepSeek, Claude, Gemini, Kimi, GLM, Ollama. [ericwang915/PythonClaw · Python · 43 stars · active]
- [MedClaw (Suzano)](https://agentx.webioinfo.top/agents/medclaw-suzano): Universal framework for domain-specific AI agents (Medical, Legal, Finance, Sports) [suzano-ai/medclaw · TypeScript · 1 stars · stale]

### Multi-Agent Orchestration (12)

- [Multica](https://agentx.webioinfo.top/agents/multica): Make humans and AI agents work as one team — open-source and self-hostable. [multica-ai/multica · Go · 50.2k stars · active]
- [Edict](https://agentx.webioinfo.top/agents/edict): 🏛️ 三省六部制 · OpenClaw Multi-Agent Orchestration System — 9 specialized AI agents with real-time dashboard, model config, and full audit trails [cft0808/edict · Python · 16.9k stars · active]
- [Open Multi-Agent](https://agentx.webioinfo.top/agents/open-multi-agent): Self-hosted TypeScript agent runtime with durable approvals and verifiable run records. Own it, approve it, audit it. [open-multi-agent/open-multi-agent · TypeScript · 6.9k stars · active]
- [AgentTeams](https://agentx.webioinfo.top/agents/agentteams): An open-source Collaborative Multi-Agent OS for transparent, human-in-the-loop task coordination via Matrix rooms. [agentscope-ai/AgentTeams · Go · 5.6k stars · active]
- [ClawTeam](https://agentx.webioinfo.top/agents/clawteam): "ClawTeam: Agent Swarm Intelligence" (One Command → Full Automation) [HKUDS/ClawTeam · Python · 5.5k stars · stale]
- [Clawith](https://agentx.webioinfo.top/agents/clawith): Your First AI Agents Company [dataelement/Clawith · Python · 4.2k stars · active]
- [TinyAGI](https://agentx.webioinfo.top/agents/tinyagi): TinyAGI is the agent teams orchestrator for One Person Company. (fka TinyClaw) [TinyAGI/tinyagi · TypeScript · 3.6k stars · stale]
- [MetaClaw (aiming-lab)](https://agentx.webioinfo.top/agents/metaclaw-aiming-lab): 🦞 Just talk to your agent — it learns and EVOLVES 🧬. [aiming-lab/MetaClaw · Python · 3.5k stars · stable · paper: arXiv]
- [ClawTeam-OpenClaw](https://agentx.webioinfo.top/agents/clawteam-openclaw): ClawTeam fork fully adapted for OpenClaw — multi-agent swarm coordination with OpenClaw as the default agent [win4r/ClawTeam-OpenClaw · Python · 1.5k stars · active]
- [ArgusBot](https://agentx.webioinfo.top/agents/argusbot): ArgusBot: A 24/7 supervisor Agent for Codex CLI and Claude Code CLI that keeps agents running, reviewing, and planning until the job is actually done. [waltstephen/ArgusBot · Python · 316 stars · stale]
- [MagiClaw](https://agentx.webioinfo.top/agents/magiclaw): MagiClaw: Conversational Command Center for Your Scientific Agent Team. [sjtu-sai-agents/MagiClaw · Python · 130 stars · stale]
- [GUIA CLI](https://agentx.webioinfo.top/agents/guia-cli) [edwincheunglab/guia-cli · Python · 2 stars · active · paper: bioRxiv]

### Benchmarks (10)

- [PinchBench](https://agentx.webioinfo.top/agents/pinchbench): PinchBench is a benchmarking system for evaluating LLM models as OpenClaw coding agents. Made with 🦀 by the humans at https://kilo.ai [pinchbench/skill · Python · 1.3k stars · active]
- [Claw-Eval](https://agentx.webioinfo.top/agents/claw-eval): Claw-Eval is an evaluation harness for evaluating LLM as agents. All tasks verified by humans. [claw-eval/claw-eval · Python · 773 stars · active]
- [ClawBench](https://agentx.webioinfo.top/agents/clawbench): Open-source benchmark for browser AI agents on daily tasks. [TIGER-AI-Lab/ClawBench · Python · 771 stars · active]
- [ResearchClawBench](https://agentx.webioinfo.top/agents/researchclawbench): 🦞 ResearchClawBench: Evaluating AI Agents for Automated Research from Re-Discovery to New-Discovery [InternScience/ResearchClawBench · Jupyter Notebook · 264 stars · active]
- [BixBench](https://agentx.webioinfo.top/agents/bixbench): Benchmark for LLM-based Agents in Computational Biology [Future-House/BixBench · Python · 145 stars · stale · paper: arXiv]
- [ClawMark](https://agentx.webioinfo.top/agents/clawmark): 🦞 ClawMark: A Living-World Benchmark for Multi-Day, Multimodal Coworker Agents [evolvent-ai/ClawMark · Python · 124 stars · active]
- [RareArena](https://agentx.webioinfo.top/agents/rarearena): A Comprehensive Rare Disease Diagnostic Dataset with nearly 50,000 patients covering more than 4000 diseases [zhao-zy15/RareArena · Python · 53 stars · stable · paper: Lancet-Digital-Health]
- [BioAgent Bench](https://agentx.webioinfo.top/agents/bioagent-bench): Benchmark for evaluating LLM agents in bioinformatics [bioagent-bench/bioagent-bench · Python · 33 stars · active · paper: arXiv]
- [HeurekaBench](https://agentx.webioinfo.top/agents/heurekabench): [ICLR 2026] A framework to "create benchmarks" and "evaluate AI co-scientists" in experimental data-driven real-world scientific research. [mlbio-epfl/HeurekaBench · Python · 21 stars · active · paper: arXiv]
- [BioDSA-1K](https://agentx.webioinfo.top/agents/biodsa-1k): Benchmark for data-driven biomedical hypothesis validation — 1,029 hypothesis-centric tasks with 1,177 analysis plans curated from 300+ published studies on cBioPortal patient data. Includes non-veri… [huggingface.co/zifeng-ai/BioDSA-1K · 0 stars · no-repo · paper: arXiv]

### Safety & Security (4)

- [ClawVault](https://agentx.webioinfo.top/agents/clawvault): OpenClaw Security Vault — Atomic "claw" control: every AI reach, within your sight. [tophant-ai/ClawVault · Python · 1.1k stars · active]
- [ClawSafety (Benchmark)](https://agentx.webioinfo.top/agents/clawsafety-benchmark): Safety benchmark for personal AI agents under realistic prompt injection. 120 test cases across 5 professional domains, 3 injection vectors, and 5 harmful action types. (arXiv 2026) [weibowen555/ClawSafety · Python · 18 stars · stale · paper: arXiv]
- [ClawKeeper](https://agentx.webioinfo.top/agents/clawkeeper): Open-source security scanner for AI agent hosts. Audits macOS and Linux machines for misconfigurations, credential exposure, and network hardening — 42 checks across 5 phases with auto-fix. [rad-security/clawkeeper · TypeScript · 17 stars · archived]
- [ClawSafety (Scanner)](https://agentx.webioinfo.top/agents/clawsafety-scanner): Security scanner for Agent Skills. Protect the Agent-Native ecosystem. [relaxcloud-cn/clawsafety · Rust · 2 stars · active]

### Others (2)

- [OpenMAIC](https://agentx.webioinfo.top/agents/openmaic): Open Multi-Agent Interactive Classroom — Get an immersive, multi-agent learning experience in just one click [THU-MAIC/OpenMAIC · TypeScript · 37.5k stars · stable · paper: Journal of Computational Science and Technology]
- [MathClaw](https://agentx.webioinfo.top/agents/mathclaw): MathClaw is a multimodal learning assistant designed for middle and high school mathematics. [MathClaw-ruc/MathClaw · Python · 374 stars · stale]