K Dense BYOK
An AI co-scientist powered by Scientific Agent Skills running on your desktop.
What is K Dense BYOK?
K Dense BYOK is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to ai co-scientist powered by scientific agent skills running on your desktop.
An AI co-scientist powered by Scientific Agent Skills running on your desktop.
This server falls under the Data Science & ML category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- An AI co-scientist powered by Scientific Agent Skills runnin
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx k-dense-byokConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use K Dense BYOK
K-Dense BYOK is a local-first AI co-scientist desktop application that runs 140+ scientific skills and 326 workflow templates across genomics, proteomics, drug discovery, materials science, and 18 other disciplines using your own API key (BYOK) via OpenRouter or local Ollama models. It connects to 229 scientific and financial databases including PubMed-scale biomedical resources, and supports 21 specialized sub-agents for citation checking, statistical review, and peer review. Researchers use it to run differential expression analyses, literature reviews, figure generation, and manuscript checking entirely on their own machine without data leaving their environment.
Prerequisites
- Node.js 18+ and npm or pnpm installed
- An OpenRouter API key (free tier available) or Ollama installed for local models
- Git to clone the repository
- An MCP client such as Claude Desktop or Claude Code
- Sufficient local compute for Ollama if avoiding cloud API keys
Clone the repository
Clone the K-Dense BYOK repository to your local machine.
git clone https://github.com/K-Dense-AI/k-dense-byok.git
cd k-dense-byokConfigure your API key
Copy the example environment file and add your OpenRouter API key. OpenRouter gives access to OpenAI, Anthropic, Google, xAI, and Qwen models through one key.
cp .env.example .env
# Edit .env and add:
# OPENROUTER_API_KEY=sk-or-your-key-hereStart the application
Run the start script to launch the K-Dense BYOK web interface locally.
./start.shAccess the web interface
Open your browser and navigate to http://localhost:3000. The initial setup will configure available scientific skills and databases.
Configure Claude Desktop as MCP client
Register K-Dense BYOK as an MCP server in your Claude Desktop config to use its scientific skills directly in Claude conversations.
{
"mcpServers": {
"k-dense-byok": {
"command": "npx",
"args": ["k-dense-byok"],
"env": {
"OPENROUTER_API_KEY": "sk-or-your-key-here"
}
}
}
}K Dense BYOK Examples
Client configuration
Claude Desktop configuration for K-Dense BYOK using an OpenRouter API key.
{
"mcpServers": {
"k-dense-byok": {
"command": "npx",
"args": ["k-dense-byok"],
"env": {
"OPENROUTER_API_KEY": "sk-or-your-openrouter-key"
}
}
}
}Prompts to try
Example scientific workflows you can run using K-Dense BYOK's 140+ skill library.
- "Run a differential expression analysis on counts.csv comparing treated vs control and plot a volcano plot"
- "Search PubMed for recent papers on CRISPR base editing in hematopoietic stem cells and summarize the top 5"
- "Check my manuscript introduction.docx for citation accuracy and flag unsupported claims"
- "Generate a figure showing protein structure comparison between these two PDB IDs"
- "Run a statistical power analysis for my planned clinical trial with n=50 per arm"Troubleshooting K Dense BYOK
start.sh fails with permission denied
Make the script executable with 'chmod +x start.sh' before running it.
OpenRouter API calls fail with 401 unauthorized
Verify that OPENROUTER_API_KEY is correctly set in .env (starts with 'sk-or-'). Check your OpenRouter dashboard to confirm the key is active and has available credits.
Scientific databases return no results
Some databases require internet access and may be rate-limited. Ensure you are connected to the internet and try reducing the scope of your query. Check the k-dense-byok GitHub issues for known database outages.
Frequently Asked Questions about K Dense BYOK
What is K Dense BYOK?
K Dense BYOK is a Model Context Protocol (MCP) server that ai co-scientist powered by scientific agent skills running on your desktop. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install K Dense BYOK?
Follow the installation instructions on the K Dense BYOK GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with K Dense BYOK?
K Dense BYOK works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is K Dense BYOK free to use?
Yes, K Dense BYOK is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
K Dense BYOK Alternatives — Similar Data Science & ML Servers
Looking for alternatives to K Dense BYOK? 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 K Dense BYOK 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 K Dense BYOK?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.