universal-mcp-servers
A comprehensive collection of MCP (Model Context Protocol) servers including GitHub, Vercel, and PyMuPDF4LLM implementations with setup guides and utilities
Documentation
๐ Universal MCP Servers Collection
> A comprehensive, production-ready collection of Model Context Protocol (MCP) servers with performance optimizations, bug fixes, and universal configurations.
๐ What's Included
๐ฏ 5 Production-Ready MCP Servers
| Server | Tools | Status | Key Features |
|---|---|---|---|
| ๐ GitHub MCP | 50+ | โ Fully Functional | Repository management, Issues, PRs, Search, Security alerts |
| ๐ Vercel MCP | 11+ | โ Production Ready | Deployments, Projects, Environment variables, Teams |
| ๐ PyMuPDF4LLM MCP | 1 | โ Performance Fixed | PDF processing 50% faster than original |
| ๐ Browser Tools MCP | 15+ | โ Ready | Web navigation, Screenshots, Form automation |
| ๐ Excel MCP | 6+ | โ Ready | Excel manipulation, Formulas, Screen capture |
๐ง Performance Fixes Applied
๐จ CRITICAL PyMuPDF4LLM Performance Bug Fixed
- Issue: PDF processed twice when using `save_path` (2x processing time)
- Solution: Eliminated duplicate `pymupdf4llm.to_markdown()` calls
- Result: 50% performance improvement for file operations
- Additional Fix: Server startup log level validation error resolved
๐ Quick Comparison
Before vs After Performance
| Operation | Original | Fixed Version | Improvement |
|---|---|---|---|
| PDF Direct Content | โก Fast | โก Fast | No change |
| PDF File Save | ๐ 2x processing | โก 1x processing | 50% faster |
| Large PDFs (100+ pages) | ๐ Very slow | โก Optimized | Significant |
| Server Startup | โ Failed | โ Reliable | Fixed |
๐ ๏ธ Quick Start
1. Clone & Install
git clone https://github.com/Mgabr90/universal-mcp-servers.git
cd universal-mcp-servers
# Quick installation (Windows)
.\scripts\setup\quick-install.ps12. Configure Your Client
๐ฏ Cursor IDE:
{
"mcpServers": {
"github-universal": {
"command": "github-mcp-server",
"env": { "GITHUB_PERSONAL_ACCESS_TOKEN": "your_token" }
},
"pymupdf4llm-fixed": {
"command": "python",
"args": ["C:/path/to/universal-mcp-servers/servers/pymupdf4llm-mcp/app.py"]
}
}
}๐ฌ Claude Desktop:
- Copy `configs/claude-desktop/claude_desktop_config.json`
- Update paths and add your API tokens
- Restart Claude Desktop
3. Test Your Setup
"List my GitHub repositories"
"Convert this PDF to markdown"
"List my Vercel deployments"
"Take a screenshot of this webpage"
"Read data from this Excel file"๐ฏ Real-World Usage Examples
GitHub Workflow Automation
๐ค "Search for Python repositories with machine learning topics"
๐ค "Fork the most popular one to my account"
๐ค "Create a new branch called 'feature-documentation'"
๐ค "Create an issue about improving the README"
๐ค "Create a pull request with documentation improvements"PDF Processing Pipeline
๐ค "Convert this aviation regulation PDF to markdown"
๐ค "Extract all images and save them to a separate folder"
๐ค "Process this 200-page technical manual efficiently"
๐ Result: 50% faster processing than original implementationVercel Deployment Management
๐ค "List my recent deployments and their status"
๐ค "Create a new project called 'universal-dashboard'"
๐ค "Set up environment variables for production"
๐ค "Deploy the latest version with custom domains"Excel Data Analysis
๐ค "Read the sales data from the Q4 spreadsheet"
๐ค "Create a summary table with totals and percentages"
๐ค "Add formulas to calculate year-over-year growth"
๐ค "Take a screenshot of the final dashboard"๐ Repository Structure
๐ฆ universal-mcp-servers/
โโโ ๐ฅ๏ธ servers/
โ โโโ ๐ pymupdf4llm-mcp/ # Performance-fixed PDF processing
โ โ โโโ app.py # โก 50% faster implementation
โ โ โโโ pyproject.toml # Dependencies and configuration
โ โ โโโ README.md # Performance fix documentation
โโโ โ๏ธ configs/
โ โโโ ๐ฏ cursor/ # Cursor IDE configurations
โ โโโ ๐ฌ claude-desktop/ # Claude Desktop configurations
โโโ ๐ scripts/
โ โโโ ๐ ๏ธ setup/ # Installation and setup scripts
โ โโโ ๐ช windows/ # Windows-specific utilities
โโโ ๐ docs/
โ โโโ ๐ setup-guides/ # Detailed setup instructions
โ โโโ ๐ troubleshooting/ # Problem resolution guides
โ โโโ ๐ api-documentation/ # Complete API reference
โโโ ๐ README.md # This file๐ง Server Details
๐ GitHub MCP Server (50+ Tools)
๐ฏ Repository Management:
- `create_repository`, `fork_repository`, `search_repositories`
- `get_file_contents`, `create_or_update_file`, `delete_file`, `push_files`
๐ Issue & PR Workflows:
- `create_issue`, `update_issue`, `add_issue_comment`
- `create_pull_request`, `merge_pull_request`, `request_copilot_review`
๐ Search & Discovery:
- `search_code`, `search_issues`, `search_users`, `search_repositories`
๐ Notifications:
- `list_notifications`, `mark_all_notifications_read`, `dismiss_notification`
๐ Security:
- `list_code_scanning_alerts`, `list_secret_scanning_alerts`
๐ Vercel MCP Server (11+ Tools)
๐ฆ Deployment Management:
- `vercel-list-all-deployments`, `vercel-get-deployment`
- `vercel-create-deployment`, `vercel-list-deployment-files`
๐๏ธ Project Configuration:
- `vercel-list-projects`, `vercel-create-project`
- `vercel-get-environments`, `vercel-create-environment-variables`
๐ฅ Team Management:
- `vercel-list-all-teams`, `vercel-create-team`
๐ PyMuPDF4LLM MCP (Performance Fixed)
โก PDF Processing:
- `convert_pdf_to_markdown` - 50% faster than original
- Image extraction and referencing
- Large file support with chunking
- Reliable server startup
๐ง Technical Improvements:
# โ Original (slow - duplicate processing)
content = pymupdf4llm.to_markdown(file_path)
if save_path:
content = pymupdf4llm.to_markdown(file_path) # DUPLICATE!
# โ
Fixed (fast - single processing)
content = pymupdf4llm.to_markdown(file_path)
if save_path:
# Reuse existing content - 50% faster!
with open(save_path, "w") as f:
f.write(content)๐ Browser Tools MCP (15+ Tools)
- Web navigation and interaction
- Screenshot and content extraction
- Form automation and submission
๐ Excel MCP Server (6+ Tools)
- Sheet reading and writing
- Formula and table creation
- Screen capture (Windows)
๐จ Benefits of This Collection
๐ Universal Compatibility
- โ Works everywhere - any workspace, any directory
- โ No per-project setup required
- โ Consistent tool availability across all projects
โก Performance Optimized
- โ 50% faster PDF processing (fixed duplicate bug)
- โ Reliable server startup (fixed log level validation)
- โ Efficient API usage across all servers
๐๏ธ Production Ready
- โ Based on popular packages with proven track records
- โ Comprehensive error handling and logging
- โ Security best practices for token management
๐ Developer Friendly
- โ Complete documentation and setup guides
- โ Troubleshooting assistance for common issues
- โ Example configurations for popular MCP clients
๐ Prerequisites
- ๐ช Windows 10/11 (PowerShell 7+)
- ๐ Python 3.10+ (for PyMuPDF4LLM MCP)
- ๐ฆ Node.js 18+ (for Vercel, Browser, Excel MCPs)
- ๐ API Tokens:
- GitHub Personal Access Token (repo, read:user, notifications scopes)
- Vercel API Token (full access)
๐ง Installation Options
๐ Option 1: Quick Install (Recommended)
# Clone repository
git clone https://github.com/Mgabr90/universal-mcp-servers.git
cd universal-mcp-servers
# Run quick installer
.\scripts\setup\quick-install.ps1๐ Option 2: Manual Setup
1. Follow individual guides in `docs/setup-guides/`
2. Copy configurations from `configs/`
3. Update paths and tokens
4. Test each server individually
๐ฏ Option 3: Custom Configuration
1. Use universal templates in `configs/`
2. Select only the servers you need
3. Customize settings for your workflow
๐ Troubleshooting
Common Issues & Quick Fixes
| Issue | Quick Fix |
|---|---|
| ๐จ PyMuPDF4LLM slow processing | โ Use our fixed version |
| ๐ GitHub 403 permission error | Update token scopes (repo, notifications) |
| ๐ซ Server not starting | Check PATH and dependencies |
| โ๏ธ Config not loading | Validate JSON, restart client |
| ๐ Path errors | Use absolute paths, escape backslashes |
| ๐ฆ NPX not found | Install Node.js |
๐ Detailed Solutions: See `docs/troubleshooting/COMMON_ISSUES.md`
๐ค Contributing
1. ๐ด Fork the repository
2. ๐ฟ Create a feature branch (`git checkout -b feature/amazing-feature`)
3. ๐พ Commit your changes (`git commit -m 'Add amazing feature'`)
4. ๐ค Push to the branch (`git push origin feature/amazing-feature`)
5. ๐ Open a Pull Request
๐ License
MIT License - see LICENSE file for details.
Third-Party Acknowledgments
- PyMuPDF4LLM: Original by pymupdf/pymupdf4llm-mcp (modified for performance)
- Vercel MCP: nganiet/mcp-vercel
- Browser Tools: @agentdeskai/browser-tools-mcp
- Excel MCP: @negokaz/excel-mcp-server
๐ What You Get
โ 70+ Professional Tools ready for immediate use
โ Performance Optimizations with measurable improvements
โ Universal Compatibility across development environments
โ Production-Grade Reliability for mission-critical workflows
โ Comprehensive Documentation for quick onboarding
โ Active Maintenance with ongoing improvements
๐ฏ Ready to Transform Your Workflow?
git clone https://github.com/Mgabr90/universal-mcp-servers.git
cd universal-mcp-servers
.\scripts\setup\quick-install.ps1๐ Your AI assistant will have universal access to:
- Complete GitHub workflow automation
- Lightning-fast PDF processing
- Seamless Vercel deployment management
- Advanced web browser control
- Powerful Excel data manipulation
Frequently asked questions
What is universal-mcp-servers?
universal-mcp-servers is A comprehensive collection of MCP (Model Context Protocol) servers including GitHub, Vercel, and PyMuPDF4LLM implementations with setup guides and utilities
How do I install universal-mcp-servers?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is universal-mcp-servers open source?
Yes โ it is hosted on GitHub at https://github.com/Mgabr90/universal-mcp-servers and has 1 stars.
Related MCP tools
๐ OpenHands: Code Less, Make More for the Model Context Protocol. Enhance AI assistants with powerful integrations. Python-based implementation.
The all-in-one Desktop & Docker AI application with built-in RAG, AI agents, No-code agent builder, MCP compatibility, and more.
Universal memory layer for AI Agents; Announcing OpenMemory MCP - local and secure memory management. Python-based implementation.
ๅบไบๅคงๆจกๅๆญๅปบ็่ๅคฉๆบๅจไบบ๏ผๅๆถๆฏๆ ๅพฎไฟกๅ ฌไผๅทใไผไธๅพฎไฟกๅบ็จใ้ฃไนฆใ้้ ็ญๆฅๅ ฅ๏ผๅฏ้ๆฉChatGPT/Claude/DeepSeek/ๆๅฟไธ่จ/่ฎฏ้ฃๆ็ซ/้ไนๅ้ฎ/ Gemini/GLM-4/Kimi/LinkAI๏ผ่ฝๅค็ๆๆฌใ่ฏญ้ณๅๅพ็๏ผ่ฎฟ้ฎๆไฝ็ณป็ปๅไบ่็ฝ๏ผๆฏๆๅบไบ่ชๆ็ฅ่ฏๅบ่ฟ่กๅฎๅถไผไธๆบ่ฝๅฎขๆใ
:robot: The free, Open Source alternative to OpenAI, Claude and others. Self-hosted and local-first. Drop-in replacement for OpenAI, running on consumer-gra...
๐ Cherry Studio is a desktop client that supports for multiple LLM providers. Built for the Model Context Protocol to enhance AI capabilities.
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP