This configuration includes a collection of Model Context Protocol (MCP) servers that extend the capabilities of AI assistants like Claude Desktop and code editors like Cursor by providing access to external tools, data sources, and specialized functionalities.
MCP defines a client-server architecture where hosts (applications like Claude Desktop, Cursor IDE, or custom agents) connect to servers that expose Tools, Resources, and Prompts via a standard API to AI models. The protocol enables tools (model-controlled functions), resources (application-controlled data sources), and prompts (user-controlled templates) to enhance AI capabilities.
A Model Context Protocol server that enhances Claude's ability to solve complex programming tasks through structured, step-by-step thinking.
Features:
- Uses the HYBRID_DESIGN prompt which demonstrated the highest average solution quality (89%) and most consistent performance across programming scenarios
- Provides ready-to-go prompts that trigger code reasoning through Claude Desktop's "+" icon or Claude Code's /help commands
- Enhanced problem-solving through sequential thinking methodology
Usage: Append "Use sequential thinking to reason about this" to your chat messages to trigger the MCP
A local Go binary that provides package version management capabilities. This appears to be a custom server located at /Users/eboney/go/bin/mcp-package-version.
Provides structured reasoning framework with predefined stages such as Problem Definition, Research, Analysis, Synthesis, and Conclusion.
Features:
- Thought tracking and management with metadata for precise navigation and retrospective analysis
- Export and integration capabilities for use in other tools and integration with platforms and code editors
- Session management for complex reasoning workflows
A Model Context Protocol server that provides browser automation capabilities using Playwright, enabling LLMs to interact with web pages through structured accessibility snapshots.
Key Advantages:
- Fast and lightweight - uses Playwright's accessibility tree, not pixel-based input
- LLM-friendly - no vision models needed, operates purely on structured data
- Deterministic tool application - avoids ambiguity common with screenshot-based approaches
Use Cases:
- Web navigation and form-filling, data extraction from structured content, automated testing driven by LLMs
- Browser automation for testing and scraping
- Automated test generation by exploring web applications like a real user
Context7 MCP pulls up-to-date, version-specific documentation and code examples straight from the source and places them directly into your prompt.
Features:
- Real-time documentation access - get the most recent official docs delivered directly into your prompt
- Version-specific code examples that are accurate for the exact version of the library you're using
- Universal compatibility with major MCP-compatible clients including Claude Desktop, Cursor, Windsurf, and more
Usage: Just say "use context7" in Cursor and you're done - no more copying links or switching tabs for documentation.
A Model Context Protocol server that provides web search capabilities through DuckDuckGo, with additional features for content fetching and parsing.
Features:
- Web search with advanced rate limiting and result formatting
- Content fetching to retrieve and parse webpage content with intelligent text extraction
- LLM-friendly output with results formatted specifically for large language model consumption
- Privacy-conscious search functionality without compromising user data, with specialized search types for news, videos, and images
A web scraping and crawling MCP server that provides powerful web content extraction capabilities. Requires a Firecrawl API key (set in the FIRECRAWL_API_KEY environment variable).
Features:
- Advanced web scraping capabilities
- Content extraction and parsing
- Integration with Firecrawl's crawling infrastructure
Provides mind mapping capabilities for visual organization and structuring of information and ideas through the MCP protocol.
Provides MCP multi-cluster Docker management and operations, featuring management interface and built-in tools covering common DevOps scenarios.
Features:
- Docker container management through MCP
- DevOps workflow integration
- Container orchestration capabilities
- Node.js version 16 or higher (for npm-based servers)
- Python 3.10+ with uv (for Python-based servers)
- Go (for the package-version server)
- Claude Desktop or Cursor IDE with MCP support enabled
Configure Claude Desktop by editing ~/Library/Application Support/Claude/claude_desktop_config.json on macOS or the equivalent location on other platforms:
{
"mcpServers": {
"code-reasoning": {
"command": "npx",
"args": ["-y", "@mettamatt/code-reasoning"]
},
"package-version": {
"command": "/Users/eboney/go/bin/mcp-package-version"
},
"sequential-thinking": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-sequential-thinking"]
},
"playwright": {
"command": "npx",
"args": ["@playwright/mcp@latest"]
},
"Context7": {
"command": "npx",
"args": ["-y", "@upstash/context7-mcp"]
},
"ddg-search": {
"command": "uvx",
"args": ["duckduckgo-mcp-server"]
},
"firecrawl": {
"command": "npx",
"args": ["-y", "firecrawl-mcp"],
"env": {
"FIRECRAWL_API_KEY": "fc-YOUR_FREE_KEY_HERE"
}
},
"mindmap": {
"command": "npx",
"args": ["-y", "mindmap-mcp-server"]
},
"mcp-server-docker": {
"command": "uvx",
"args": ["mcp-server-docker"]
}
}
}You can also configure MCP servers on a per-project basis by creating .cursor/mcp.json in your project root. This is useful for project-specific tools like Git integration:
{
"mcpServers": {
"git": {
"command": "uvx",
"args": [
"mcp-server-git",
"--repository",
"/put/your/path/here/to/your/repo"
]
}
}
}Note: Replace /put/your/path/here/to/your/repo with the actual path to your Git repository. This Git MCP server provides Git operations and repository management capabilities specific to your project.
A project-specific MCP server that provides Git repository management and operations.
Features:
- Git command execution through MCP
- Repository status and history access
- Branch management and operations
- Commit and staging operations
Configuration: This server is typically configured at the project level in .cursor/mcp.json with the --repository flag pointing to your specific Git repository path.
Create or edit .cursor/mcp.json globally in ~/.cursor/mcp.json:
{
"mcpServers": {
"code-reasoning": {
"command": "npx",
"args": ["-y", "@mettamatt/code-reasoning"]
},
"package-version": {
"command": "/Users/eboney/go/bin/mcp-package-version"
},
"sequential-thinking": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-sequential-thinking"]
},
"playwright": {
"command": "npx",
"args": ["@playwright/mcp@latest"]
},
"Context7": {
"command": "npx",
"args": ["-y", "@upstash/context7-mcp"]
},
"ddg-search": {
"command": "uvx",
"args": ["duckduckgo-mcp-server"]
},
"firecrawl": {
"command": "npx",
"args": ["-y", "firecrawl-mcp"],
"env": {
"FIRECRAWL_API_KEY": "fc-YOUR_FREE_KEY_HERE"
}
},
"mindmap": {
"command": "npx",
"args": ["-y", "mindmap-mcp-server"]
},
"mcp-server-docker": {
"command": "uvx",
"args": ["mcp-server-docker"]
}
}
}-
Restart Required: Save the configuration file and restart Claude Desktop or Cursor for changes to take effect
-
Tool Discovery: Once configured, you should see MCP tools available in the interface, indicated by a hammer icon or similar MCP indicator
-
Environment Variables: Make sure to replace
"fc-YOUR_FREE_KEY_HERE"with your actual Firecrawl API key -
Path Adjustments: Update the package-version server path (
/Users/eboney/go/bin/mcp-package-version) to match your actual installation location -
Tool Invocation: For code-reasoning, append phrases like "Use sequential thinking to reason about this" to trigger the enhanced reasoning capabilities
- Server Not Found: Ensure all required dependencies (Node.js, Python, uv, etc.) are installed and accessible in your PATH
- Permission Issues: Check that executable files have proper permissions
- API Keys: Verify that environment variables for services like Firecrawl are properly set
- Network Issues: Some servers require internet access for functionality
- Official MCP Documentation: https://modelcontextprotocol.io/
- Awesome MCP Servers Collection: https://github.com/punkpeye/awesome-mcp-servers
- MCP Configuration Guide: https://docs.cursor.com/context/model-context-protocol
This collection of MCP servers significantly extends the capabilities of AI assistants, enabling everything from enhanced reasoning and web automation to real-time documentation access and advanced search functionality.