paper-search-mcp-nodejs

# AIPaper-assisant User Guide ## 🚀 Quick Start ### Prerequisites - Node.js ≥ 18.0.0 - npm or yarn package manager - API keys for premium platforms (optional) ### Installation 1. **Clone the repository** ```bash git clone https://github.com/yourusername/paper-search-nodejs.git cd paper-search-nodejs ``` 2. **Install dependencies** ```bash npm install ``` 3. **Configure environment** ```bash cp .env.example .env # Edit .env with your API keys ``` 4. **Build the project** ```bash npm run build ``` 5. **Start the MCP server** ```bash npm start ``` ## 🔧 Configuration ### Environment Variables Create a `.env` file in the project root: ```bash # Required for Web of Science WOS_API_KEY=your_wos_api_key_here # Required for Scopus SCOPUS_API_KEY=your_scopus_api_key_here # Required for ScienceDirect SCIENCEDIRECT_API_KEY=your_elsevier_api_key_here # Required for Springer SPRINGER_API_KEY=your_springer_api_key_here # Required for Wiley TDM WILEY_API_KEY=your_wiley_api_key_here # Optional: MCP server configuration MCP_SERVER_NAME=paper-search-mcp MCP_SERVER_VERSION=0.2.3 ``` ### Getting API Keys #### Web of Science 1. Visit [Clarivate Developer Portal](https://developer.clarivate.com/) 2. Register for an account 3. Subscribe to Web of Science API 4. Copy your API key #### Scopus 1. Visit [Elsevier Developer Portal](https://dev.elsevier.com/) 2. Create an account 3. Register your application 4. Get your API key #### ScienceDirect & Scopus 1. Both use the same Elsevier API key 2. Ensure your subscription includes the required APIs #### Springer 1. Visit [Springer Nature API Portal](https://dev.springernature.com/) 2. Register your application 3. Obtain API credentials #### Wiley 1. Contact Wiley directly for TDM API access 2. Provide your use case details 3. They will provision API access ## 📖 Basic Usage ### With Claude Desktop 1. **Configure Claude Desktop** Add to your `claude_desktop_config.json`: ```json { "mcpServers": { "paper-search": { "command": "node", "args": ["/path/to/paper-search-nodejs/dist/server.js"], "env": { "WOS_API_KEY": "your_wos_api_key", "SCOPUS_API_KEY": "your_scopus_api_key" } } } } ``` 2. **Restart Claude Desktop** 3. **Use natural language to search** Example queries: - "Search for papers about machine learning on arXiv" - "Find recent quantum computing research from 2023" - "Download the paper with DOI 10.1000/example" ### Programmatic Usage ```javascript import { McpClient } from '@modelcontextprotocol/sdk'; const client = new McpClient(); await client.connect('paper-search'); // Search for papers const results = await client.callTool('search_papers', { query: 'machine learning transformers', platform: 'arxiv', maxResults: 10 }); // Download a paper const download = await client.callTool('download_paper', { doi: '10.1000/example', platform: 'arxiv' }); ``` ## 🔍 Search Examples ### Basic Search **Search across all platforms:** ``` Search for papers about climate change ``` **Search specific platform:** ``` Search for quantum computing papers on arXiv ``` **Limit results:** ``` Find 5 papers about CRISPR on PubMed ``` ### Advanced Search **Year range filtering:** ``` Search for AI papers from 2020-2024 on Web of Science ``` **Multiple platforms:** ``` Search for COVID-19 papers on both PubMed and bioRxiv ``` **Citation-focused search:** ``` Find highly cited papers about blockchain on Scopus ``` ## 📥 Download Examples ### Download by DOI ``` Download the paper with DOI 10.1038/s41586-023-06083-9 ``` ### Download from specific platform ``` Download a machine learning paper from arXiv ``` ### Download with title hint ``` Download the paper titled "Attention Is All You Need" from arXiv ``` ## 🛠️ Platform-Specific Features ### arXiv - **Categories**: Search within specific categories (cs.AI, physics.gen-ph, etc.) - **Date sorting**: Get newest papers first - **PDF download**: Direct access to full papers Example: ``` Search for recent computer vision papers in cs.CV category on arXiv ``` ### Web of Science - **Multi-topic search**: Use boolean operators (AND, OR, NOT) - **Citation tracking**: Find highly cited papers - **Research areas**: Filter by field of study Example: ``` Search for "machine learning" AND "healthcare" on Web of Science, sorted by citations ``` ### PubMed - **MeSH terms**: Medical Subject Headings for precise search - **Clinical queries**: Specialized search filters - **Gene/protein search**: Biological entity recognition Example: ``` Search for papers about BRCA1 gene mutations on PubMed ``` ### Google Scholar - **Comprehensive coverage**: Includes journals, conferences, theses - **Citation network**: Related papers and citing articles - **Author profiles**: Researcher publication lists Example: ``` Find all papers by Geoffrey Hinton on Google Scholar ``` ### Sci-Hub - **Universal access**: Works with any DOI - **Mirror management**: Automatic healthy mirror selection - **Legal notice**: Use only for research purposes Example: ``` Access the paper with DOI 10.1126/science.abc123 through Sci-Hub ``` ## ⚠️ Important Notes ### Rate Limiting - Be respectful of platform resources - Default limits are conservative - Higher limits available with API keys - Automatic retry with exponential backoff ### Legal Compliance - Respect copyright laws - Use downloads for research only - Follow platform terms of service - Sci-Hub usage may have legal implications ### Data Quality - Results vary by platform coverage - Crossref provides most comprehensive metadata - Preprints (arXiv) are not peer-reviewed - Always verify publication status ## 🔧 Troubleshooting ### Common Issues #### "No results found" - Try broader search terms - Check spelling - Try different platforms - Use English terms for better coverage #### "API key required" - Some platforms need API keys - Check your .env configuration - Verify key is valid - Check platform subscription status #### "Rate limit exceeded" - Wait a moment before retrying - Reduce search frequency - Consider upgrading API plan - Use multiple platforms alternately #### "Download failed" - Verify DOI is correct - Paper may not be openly available - Try different download platform - Check if paper exists in open access ### Debug Mode Enable debug logging: ```bash DEBUG=paper-search:* npm start ``` ### Platform Status Check platform health: ``` Show me the status of all platforms ``` ## 📚 Best Practices ### Search Tips 1. **Use specific terms**: "convolutional neural networks" vs "neural networks" 2. **Combine concepts**: "machine learning" AND "healthcare" 3. **Use quotes**: "exact phrase search" 4. **Try synonyms**: "COVID-19" OR "coronavirus" 5. **Filter by year**: Focus on recent research ### Download Tips 1. **Use DOI**: Most reliable identifier 2. **Check multiple sources**: Different platforms may have access 3. **Respect rate limits**: Don't overwhelm services 4. **Verify files**: Ensure downloads are complete 5. **Cite properly**: Give credit to authors ### Research Workflow 1. Start broad search on Crossref 2. Refine with platform-specific search 3. Check citations for related work 4. Download relevant papers 5. Organize findings ## 🔗 Integration Examples ### Zotero Integration Export results and import to Zotero for reference management. ### LaTeX Integration Generate BibTeX entries from search results. ### Jupyter Notebook ```python # Example Python integration import subprocess import json result = subprocess.run([ 'node', 'dist/server.js', 'search_papers', '--query', 'machine learning', '--platform', 'arxiv' ], capture_output=True, text=True) papers = json.loads(result.stdout) ``` ## 📞 Support ### Getting Help 1. Check this documentation 2. Review error messages carefully 3. Check platform status 4. Search existing issues 5. Create detailed bug reports ### Reporting Issues Include: - Platform being used - Search query - Error message - Expected vs actual behavior - Environment details ### Contributing 1. Read contributing guidelines 2. Fork the repository 3. Create feature branch 4. Add tests for new features 5. Submit pull request --- *This user guide covers the essential features of AIPaper-assisant. For advanced usage and development, see the [Developer Documentation](./DEVELOPER_GUIDE.md).*