Cursor Vibe Coding Project Template

A comprehensive project template for vibe coding in Cursor with pre-configured AI tooling, MCP servers, and intelligent development workflows.

Table of Contents

• Quick Start
• Features
• What's Included
• Vibe Coding Workflow
• Configuration
• Documentation
• Contributing
• License

🚀 Quick Start

1. Create Your Project

Recommended: Use as GitHub Template

1. Click "Use this template" button on GitHub
2. Create your new repository from the template
3. Clone your new repository:

   git clone git@github.com:yourusername/your-project-name.git
   cd your-project-name
   cursor .
   

4. Prompt the AI agent: "initialize this project as <my project name>"

Alternative: Clone Template Directly

# If you cloned the template repository directly
git clone git@github.com:your-org/project_template.git my-new-project
cd my-new-project
cursor .

# Then prompt the AI agent:
# "Initialize the repo for my new project"

The AI agent will automatically handle template setup including:

• Setting up your MCP configuration
• Customizing files for your project
• Initializing git repository and GitHub repo (only needed for cloned templates)

Manual Setup (If Preferred)

# After cloning the template to your project folder
cd my-new-project

# REQUIRED: Set up MCP configuration
cp .cursor/.mcp.json.example .cursor/mcp.json
# Edit .cursor/mcp.json with your API keys

# Optional: Copy .env only if you plan to use TaskMaster CLI
# (Not needed for the recommended MCP workflow)
cp .env.example .env  # Optional
# Add your API keys to .env  # Only if using CLI

# IMPORTANT: Customize for your project
# - Replace this README.md with your project's README
# - Update SECURITY.md to remove template references
# - Remove or update any hardcoded paths in configuration files

2. Start Vibe Coding

cursor .
# For GitHub template usage: "initialize this project as <my project name>"
# For cloned templates: "Initialize the repo for my new project"

✨ Features

This template provides a ready-to-use development environment with:

• 🎯 TaskMaster AI Integration - Intelligent task management with AI-powered PRD parsing, complexity analysis, and task breakdown
• 🧠 Memory MCP Server - Persistent knowledge graph for maintaining context across sessions
• 🔗 Context7 MCP - Enhanced codebase understanding and contextual assistance
• 🐙 GitHub MCP Integration - Seamless GitHub repository and issue management
• 📝 Pre-configured Cursor Rules - Optimized rules for frontend, testing, Python, and workflow automation
• 🚀 Automated Git Workflows - Intelligent repository initialization with scoped permissions
• 🤖 Agent-Driven Setup - AI assistant handles template initialization automatically

🛠️ What's Included

MCP Servers (Pre-configured)

• TaskMaster AI - AI-powered project management and task orchestration
• Memory - Persistent knowledge graph for context retention
• Context7 - Enhanced codebase analysis and suggestions
• GitHub - Repository management and issue tracking

Cursor Rules

• development.mdc - Frontend, testing, and Python development best practices
• rule_management.mdc - Rule structure and formatting guidelines
• self_improve.mdc - Self-improving rule management patterns
• workflow.mdc - Git, Taskmaster, and Memory MCP workflows
• taskmaster/ - Comprehensive TaskMaster workflow rules

Setup & Configuration

• .cursor/.mcp.json.example - Template MCP server configuration
• .env.example - Optional environment variables for CLI usage

Project Structure

project_template/
├── .cursor/
│   ├── mcp.json              # MCP server configuration
│   ├── .mcp.json.example     # Template for MCP setup
│   └── rules/                # Cursor coding rules
├── .taskmaster/              # TaskMaster configuration
│   ├── config.json           # AI model settings
│   ├── docs/                 # Documentation directory
│   ├── tasks/                # Task files
│   └── templates/            # PRD templates
├── .github/
│   └── instructions/         # GitHub workflow instructions
├── AGENTS.md                 # TaskMaster integration guide
├── CLAUDE.md                 # Claude Code integration guide
└── .env.example              # Optional: Environment variables for CLI usage only

🎯 Vibe Coding Workflow

This template is optimized for vibe coding - a development style that balances structure with creative flow using controlled MCP workflows.

🚀 Initialize Git Repository

Note: Only needed when cloning the template directly. When using as a GitHub template (recommended), your repository is already initialized.

Prompt: "Initialize git repo" or "Initialize the repo for my new project"

For cloned templates, the AI agent automatically follows this workflow:

1. Template Cleanup - Removes template-specific .git history
2. Local Git Setup - Initializes fresh repository with proper .gitignore
3. GitHub Repository - Creates private repo with issues enabled via GitHub CLI
4. Scoped Authentication - Generates repository-specific GitHub PAT with minimal permissions
5. MCP Configuration - Updates .cursor/mcp.json with the scoped token
6. Initial Commit - Pushes your new project to main branch using SSH authentication

📋 PRD-Driven Development (User-Controlled)

Step 1: Initialize TaskMaster

Prompt: "Initialize TaskMaster for this project"

• Uses the initialize_project MCP tool to set up project structure

Step 2: Create and Validate PRD

Prompt: "Help me create a Product Requirements Document for [describe your project idea]"

• Collaborate with AI to write a comprehensive PRD
• Refine requirements through iterative discussion
• PRD draft saved locally in .taskmaster/docs/prd.txt

Step 3: Create GitHub Issue from PRD

Prompt: "I accept this PRD. Please create a GitHub Issue with the full PRD content"

• AI creates GitHub Issue containing the complete PRD text
• GitHub Issue becomes the single source of truth for requirements
• Local PRD file is automatically deleted to prevent confusion

Step 4: Generate Tasks (Explicit Trigger Required)

Prompt: "Start working on the PRD" or "Generate tasks from the PRD Issue"

• Important: Tasks are only generated after this explicit instruction
• AI uses TaskMaster MCP to parse the GitHub Issue and create structured tasks
• Complex tasks are broken down with AI-powered complexity analysis

Step 5: Confirm Task Selection

Prompt: "I want to begin with task [ID]" or "Which task should I start with?"

• Required: AI must confirm which specific task to begin before executing
• AI will not automatically start task execution without your approval
• Uses next_task MCP tool to suggest optimal starting points based on dependencies

🧠 Memory & Context Management

Memory MCP Workflow:

• Before Memory Operations - AI asks: "May I save this information to the knowledge graph?"
• User Confirmation - "Yes, you can update memory" or "No, don't save this yet"
• Context Preservation - Understanding maintained across sessions with explicit consent

🔄 Development Loop with Explicit Control

• AI Suggests - Provides recommendations and analysis
• User Decides - Explicit approval required for significant operations
• Controlled Automation - MCP servers assist but don't override user decisions
• Progress Tracking - Development logged with user awareness

🎵 The Vibe

Intelligent assistance with human control:

1. Express your intent in natural language
2. AI structures and suggests work through MCP integration
3. You approve and direct with explicit confirmation points
4. Code in flow with full context awareness and controlled automation
5. Ship iteratively with transparent progress tracking

The MCP servers provide powerful assistance while keeping you in the driver's seat.

🔧 Configuration

Required API Keys

For MCP Workflow (Recommended): Add to .cursor/mcp.json:

{
  "mcpServers": {
    "taskmaster-ai": {
      "env": {
        "ANTHROPIC_API_KEY": "your_key_here",
        "PERPLEXITY_API_KEY": "your_key_here"
      }
    }
  }
}

For CLI Workflow (Optional): Add to .env file:

# AI Models
ANTHROPIC_API_KEY=your_key_here
PERPLEXITY_API_KEY=your_key_here  # For research features
OPENAI_API_KEY=your_key_here      # Optional
GOOGLE_API_KEY=your_key_here      # Optional

# GitHub (auto-configured during git init)
GITHUB_TOKEN=your_scoped_pat_here

Important Notes:

• The .env.example file is provided for users who want to use TaskMaster CLI commands directly. For the recommended MCP workflow in Cursor, only .cursor/mcp.json configuration is needed.
• Environment File Distinction: This template's .env.example contains TaskMaster CLI configuration. Your project may also need its own .env file for application configuration (database URLs, app API keys, etc.) - these are completely separate concerns.

MCP Server Setup

The template includes example configurations in .cursor/.mcp.json.example. Copy and customize:

{
  "mcpServers": {
    "taskmaster-ai": {
      "command": "npx",
      "args": ["-y", "--package=task-master-ai", "task-master-ai"],
      "env": {
        "ANTHROPIC_API_KEY": "your_key_here"
      }
    }
  }
}

📚 Documentation

• AGENTS.md - Comprehensive TaskMaster AI integration guide
• CLAUDE.md - Claude Code specific workflows and commands
• .cursor/rules/ - Detailed coding rules and patterns
• .taskmaster/templates/ - PRD templates and examples

📄 License

MIT License - see the LICENSE file for details.

🤝 Contributing

This template evolves with usage. The meta_rules.mdc enables self-improving patterns - rules automatically update based on emerging patterns in your codebase.

How to Contribute

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Make your changes
4. Commit your changes (git commit -m 'Add amazing feature')
5. Push to the branch (git push origin feature/amazing-feature)
6. Open a Pull Request

🔗 Related Projects

• TaskMaster AI - The core task management system
• Cursor - The AI-powered code editor this template is designed for

🙏 Acknowledgments

This template was inspired by several excellent resources in the AI development community:

• Viktor Farcic's YouTube Automation - Reference implementation of AI-powered development workflows
• My Workflow With AI: How I Code, Test, and Deploy Faster Than Ever - Comprehensive guide to AI-assisted development workflows by Viktor Farcic
• AI Labs Video - Insights on AI development practices and tooling

These resources provided valuable insights into structuring AI-powered development workflows, MCP server integration, and best practices for vibe coding with AI assistants.

---

Happy vibe coding! 🎵✨