QUICKSTART.md•6.68 kB
# Quick Start Guide
Get started with Research MCP in 2 minutes!
## Option 1: Use with npx (Easiest - No Installation!)
### 1. Configure Your MCP Client (1 minute)
**For Claude Desktop:**
Edit your config file:
- **Windows**: `%APPDATA%\Claude\claude_desktop_config.json`
- **macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`
- **Linux**: `~/.config/Claude/claude_desktop_config.json`
Add this (no API keys needed!):
```json
{
"mcpServers": {
"research-arxiv": {
"command": "npx",
"args": ["-y", "researchmcp", "arxiv"]
},
"research-semantic-scholar": {
"command": "npx",
"args": ["-y", "researchmcp", "semantic"]
},
"research-pubmed": {
"command": "npx",
"args": ["-y", "researchmcp", "pubmed"]
}
}
}
```
### 2. Restart & Try!
Restart Claude Desktop and try:
> "Find the 5 most recent papers on quantum computing from arXiv"
**That's it!** All servers work without any API keys.
---
## Option 2: Local Development
Use this if you want to modify the code or contribute.
## 1. Install (2 minutes)
```bash
# Clone the repository
git clone https://github.com/gyash1512/ResearchMCP.git
cd ResearchMCP
# Install dependencies
npm install
# Build the project
npm run build
```
## 2. Configure (Optional - Skip if no API keys)
**All servers work without API keys!** Only create `.env` if you want higher rate limits.
Create a `.env` file (optional):
```bash
cp .env.example .env
```
Edit `.env` (all optional):
```env
SEMANTIC_SCHOLAR_API_KEY=your_key_here # Optional
PUBMED_API_KEY=your_key_here # Optional
PUBMED_EMAIL=your_email@example.com # Optional
```
> **Note**: arXiv works without any API key!
## 3. Test a Server (1 minute)
Start the arXiv server to verify everything works:
```bash
npm run start:arxiv
```
You should see: `arXiv MCP Server running on stdio`
Press `Ctrl+C` to stop.
## 4. Configure Your MCP Client (1 minute)
### For Claude Desktop (using npx)
Edit your Claude Desktop config file:
- **Windows**: `%APPDATA%\Claude\claude_desktop_config.json`
- **macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`
- **Linux**: `~/.config/Claude/claude_desktop_config.json`
**Simple config (no API keys needed):**
```json
{
"mcpServers": {
"research-arxiv": {
"command": "npx",
"args": ["-y", "researchmcp", "arxiv"]
},
"research-semantic-scholar": {
"command": "npx",
"args": ["-y", "researchmcp", "semantic"]
},
"research-pubmed": {
"command": "npx",
"args": ["-y", "researchmcp", "pubmed"]
}
}
}
```
**Or for local development:**
```json
{
"mcpServers": {
"research-arxiv": {
"command": "node",
"args": ["./dist/servers/arxiv-server.js"],
"cwd": "/absolute/path/to/ResearchMCP"
},
"research-semantic-scholar": {
"command": "node",
"args": ["./dist/servers/semantic-scholar-server.js"],
"cwd": "/absolute/path/to/ResearchMCP"
},
"research-pubmed": {
"command": "node",
"args": ["./dist/servers/pubmed-server.js"],
"cwd": "/absolute/path/to/ResearchMCP"
}
}
}
```
> **Note**: Replace `/absolute/path/to/ResearchMCP` with your actual project path.
### For Cline (VS Code Extension)
Open Cline settings in VS Code and add the same configuration.
## 5. Try Your First Query!
Restart your MCP client (Claude Desktop/Cline) and try:
> "Find the 5 most recent papers on quantum computing from arXiv"
or
> "Search for papers about federated learning in healthcare and give me a summary"
## Example Queries to Try
### Simple Searches
1. **arXiv Search**
> "Search arXiv for papers about transformers in NLP published after 2023"
2. **Semantic Scholar Search**
> "Find the most cited papers on graph neural networks from the last 2 years"
3. **PubMed Search**
> "Search PubMed for recent cancer immunotherapy studies"
### Advanced Queries
4. **Multi-Source Comparison**
> "Compare papers on reinforcement learning from arXiv and Semantic Scholar. Which are most cited?"
5. **Citation Analysis**
> "Find papers that cite 'Attention is All You Need' and summarize their contributions"
6. **Trend Analysis**
> "What are the latest developments in diffusion models for image generation?"
7. **Get BibTeX Citations**
> "Find papers on quantum machine learning and give me BibTeX citations for the top 3"
## Common Use Cases
### Academic Research
```
"I'm writing a paper on federated learning. Find recent papers from all sources,
deduplicate them, and give me the top 10 with citations."
```
### Literature Review
```
"Summarize the state of research on graph neural networks for drug discovery.
Include key papers from arXiv, Semantic Scholar, and PubMed."
```
### Finding Influential Papers
```
"What are the most cited papers on transformers from 2023? Get details from
Semantic Scholar and provide BibTeX citations."
```
### Medical Research
```
"Search PubMed for clinical trials on cancer immunotherapy published in the
last year. Give me abstracts and DOIs."
```
## Troubleshooting
### Server won't start
- Check that you ran `npm run build` first
- Verify Node.js 18+ is installed: `node --version`
- Check for errors in the build output
### No results returned
- Verify your search query isn't too specific
- Check API rate limits (wait a few seconds)
- Try without year filters first
### Connection errors
- Check internet connection
- Verify API keys are correct (if using)
- Try arXiv first (no key required)
### Rate limit errors
- Wait 1-2 seconds between requests
- Get API keys for higher limits
- Consider caching results
## Next Steps
1. **Read Full Documentation**: See `README.md` for complete features
2. **API Reference**: Check `API_DOCUMENTATION.md` for detailed API info
3. **Customize**: Modify rate limits, add caching, or extend functionality
4. **Contribute**: Add more research sources or features
## Tips for Best Results
✅ **Do:**
- Start with broad queries, then refine
- Use year filters to get recent papers
- Combine multiple sources for comprehensive results
- Cache or save important results
❌ **Don't:**
- Make too many rapid requests
- Use overly specific queries initially
- Forget to check for duplicate papers across sources
- Ignore citation counts when available
## Getting Help
- **Issues**: Open a GitHub issue
- **Questions**: Check API_DOCUMENTATION.md
- **API Docs**:
- [arXiv API](https://arxiv.org/help/api)
- [Semantic Scholar](https://api.semanticscholar.org/)
- [PubMed](https://www.ncbi.nlm.nih.gov/books/NBK25501/)
---
**Happy Researching! 📚🔬**