#913 · Primary category: AI Agents & Automation

ScienceClaw

ai-agent bioinformatics data-analysis docker drug-discovery fastapi feishu langchain langgraph llm mcp multi-agent pubmed research-assistant sandbox scientific-research self-hosted tool-use vue wechat

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, better transparency, and a more user-friendly experience.

Project last updated:05/09/26

GitHub Stars

659

Forks

71

Contributors

3

License

MIT

Why we included this project

ScienceClaw is for researchers and analysts who want an AI assistant that can carry a scientific workflow from start to finish, not just answer questions. It packs more than 1,900 scientific tools behind a multi-agent system that runs inside an isolated Docker sandbox, so code execution never touches your host machine and generated data stays in a local workspace. That local, private setup makes it comfortable to hand over PubMed lookups, data crawling, and report drafting without sending your data anywhere. Every step, from web search through tool calls to the final report, is logged and traceable, which matters when you need to explain how a result was produced. With a web frontend, backend, task scheduler, and sandbox components, it reads as a complete self-hosted research assistant rather than a single script, and teams doing literature review or early-stage discovery work will find it a solid base for automating repetitive research chores.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category