ScienceClaw
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
What is ScienceClaw?
ScienceClaw is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to 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...
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
This server falls under the Data Science & ML category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- ScienceClaw is a personal research assistant built with Lang
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx scienceclawConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use ScienceClaw
ScienceClaw is a personal research assistant built on the OpenClaw engine and LangChain DeepAgents that provides scientists with access to 25+ academic databases including PubMed, Semantic Scholar, and UniProt. It features a 4-layer persistent memory system, 285 domain-specific skills across biomedicine, genomics, chemistry, and economics, and enforces a zero-hallucination protocol by requiring every citation to be verified against live tool results. Researchers use ScienceClaw to run multi-hour literature reviews, meta-analyses, and cross-disciplinary investigations that would otherwise require manual database querying.
Prerequisites
- Node.js 18 or later and pnpm installed
- Python 3.10 or later (required by underlying agent infrastructure)
- An OpenClaw-compatible MCP client
- Git to clone the repository
Clone the ScienceClaw repository
Download the project to your local machine.
git clone https://github.com/beita6969/ScienceClaw.git
cd ScienceClawRun the automated setup script
The setup script installs Node, Python dependencies, MCP servers, and research skills in one step.
chmod +x setup.sh && ./setup.shInstall dependencies manually (alternative)
If you prefer a manual install, use pnpm and then onboard via OpenClaw.
pnpm install && npx openclaw onboardConfigure OpenClaw for research features
Edit ~/.openclaw/openclaw.json to enable persistent memory and extend session timeouts above the default 10-minute limit.
{
"gateway": { "mode": "local" },
"plugins": {
"slots": { "memory": "memory-core" },
"entries": {
"memory-core": { "enabled": true },
"memory-lancedb": { "enabled": true }
}
},
"agents": {
"defaults": {
"heartbeat": { "interval": 1800 }
}
}
}Start a research session
Launch OpenClaw with ScienceClaw skills active. The agent will have access to 285 research skills and all configured academic databases.
npx scienceclawScienceClaw Examples
Client configuration
MCP client configuration to launch ScienceClaw via npx.
{
"mcpServers": {
"scienceclaw": {
"command": "npx",
"args": ["scienceclaw"]
}
}
}Prompts to try
Research-oriented prompts for ScienceClaw once connected.
- "Search PubMed for meta-analyses on CRISPR off-target effects published after 2022"
- "Retrieve the UniProt entry for human TP53 and summarize its known isoforms"
- "Conduct a literature review on mRNA vaccine stability and store findings in memory"
- "Cross-reference Semantic Scholar results on RNA sequencing with my saved notes"Troubleshooting ScienceClaw
setup.sh fails midway through installation
Check that Node.js, pnpm, and Python 3.10+ are all on your PATH before running setup.sh. Run each tool individually to confirm versions.
Memory plugin is not persisting between sessions
Confirm that memory-core and memory-lancedb are both set to enabled: true in ~/.openclaw/openclaw.json and that lancedb has write access to the data directory.
Academic database searches return no results
Some database skills require internet access to external APIs. Verify your network connection and that no firewall is blocking outbound HTTPS requests to PubMed or Semantic Scholar.
Frequently Asked Questions about ScienceClaw
What is ScienceClaw?
ScienceClaw is a Model Context Protocol (MCP) server that 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 It connects AI assistants to external tools and data sources through a standardized interface.
How do I install ScienceClaw?
Follow the installation instructions on the ScienceClaw GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with ScienceClaw?
ScienceClaw works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is ScienceClaw free to use?
Yes, ScienceClaw is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
ScienceClaw Alternatives — Similar Data Science & ML Servers
Looking for alternatives to ScienceClaw? Here are other popular data science & ml servers you can use with Claude, Cursor, and VS Code.
Ultrarag
★ 5.6kA Low-Code MCP Framework for Building Complex and Innovative RAG Pipelines
RocketRide
★ 3.1k📇 🏠 - MCP server that exposes RocketRide AI pipelines as t
Aix Db
★ 2.1kAix-DB 基于 LangChain/LangGraph 框架，结合 MCP Skills 多智能体协作架构，实现自然语言到数据洞察的端到端转换。
NeMo Data Designer
★ 1.9k🎨 NeMo Data Designer: Generate high-quality synthetic data from scratch or from seed data.
PaperBanana
★ 1.7kOpen source implementation and extension of Google Research’s PaperBanana for automated academic figures, diagrams, and research visuals, expanded to new domains like slide generation.
MiniMax
★ 1.5kBridges MiniMax AI capabilities to the Model Context Protocol, enabling AI agents to perform image understanding, text-to-image generation, and speech synthesis. It provides a standardized interface for accessing MiniMax's core tools via JSON-RPC.
Browse More Data Science & ML MCP Servers
Explore all data science & ml servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up ScienceClaw in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use ScienceClaw?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.