lovable
Unofficial MCP server for analyzing Lovable-generated projects with Claude Desktop. Community-built tool for enhanced development workflow with React, TypeScript, Supabase integration. Not affiliated with Lovable.
claude mcp add --transport stdio hiromima-lovable-mcp-server lovable-mcp-server --project-path /path/to/your/lovable/project \ --env NODE_ENV="production"
How to use
This MCP server provides real-time analysis for Lovable-generated projects and exposes a set of tools through Claude Desktop. It includes eight analysis tools (analyze_project, get_components, get_routing_structure, analyze_dependencies, get_tailwind_usage, get_hooks_usage, analyze_api_calls, analyze_database_schema) that you can invoke via Claude prompts to inspect project structure, code patterns, routing, dependencies, styling usage, hooks, API usage, and database schemas. Live resources such as project://structure and project://components give you immediate visibility into the file tree, dependencies, components, and routes. Use the smart prompts like code_review, refactor_suggest, and performance_audit to receive targeted recommendations and optimization ideas. The server emphasizes security with path validation, authentication, rate limiting, and sanitization of error messages. To start, ensure Claude Desktop is configured to connect to the Lovable MCP server, then query Claude Desktop with prompts like “Analyze the structure of this Lovable project” or “Show me the API integration patterns.”
How to install
Prerequisites:\n- Node.js 18+ (for the CLI and development workflow)\n- Claude Desktop app (to connect to the MCP server)\n\nQuick Setup:\n1. Install the MCP server globally:\nbash\nnpm install -g lovable-mcp-server\n\n2. Add to Claude Desktop configuration (claude_desktop_config.json):\njson\n{\n "mcpServers": {\n "lovable-mcp": {\n "command": "lovable-mcp-server",\n "args": ["--project-path", "/path/to/your/lovable/project"],\n "env": {\n "NODE_ENV": "production"\n }\n }\n }\n}\n\n3. (Optional) Development setup: clone the repo, install, and run locally:\nbash\ngit clone https://github.com/yourusername/lovable-mcp-server.git\ncd lovable-mcp-server\nnpm install\n\n4. For development usage in Claude config:\njson\n{\n "mcpServers": {\n "lovable-mcp-dev": {\n "command": "node",\n "args": ["/path/to/lovable-mcp-server/enhanced-simple-server.js", "--project-path=/path/to/your/lovable/project"],\n "env": {\n "NODE_ENV": "development"\n }\n }\n }\n}\n
Additional notes
Tips and notes:\n- Use absolute paths for project locations in both production and development configurations to avoid path resolution issues.\n- If Claude Desktop cannot connect, verify the claude_desktop_config.json syntax and ensure the JSON structure is valid.\n- Monitor logs at the location indicated in your environment (e.g., Claude Desktop logs) for troubleshooting.\n- Ensure your Lovable project is accessible and follows Lovable project structure (package.json, src/ directory).\n- When upgrading, re-run npm install and validate that the CLI command still maps to the same server binary.\n- If you change the project path, remember to restart the MCP server so Claude Desktop re-establishes the connection.
Related MCP Servers
plate
Rich-text editor with AI, MCP, and shadcn/ui
cui
A web UI for Claude Code agents
vrchat
This project is a Model Context Protocol (MCP) server for interacting with the VRChat API.
docmole
Dig through any documentation with AI - MCP server for Claude, Cursor, and other AI assistants
vscode-context
MCP Server to Connect with VS Code IDE
civitai
A Model Context Protocol server for browsing and discovering AI models on Civitai