# 🔧 `mcp-viz configure` Interactive Demo
## What happens when you run `mcp-viz configure`
Here's the complete interactive experience users get:
```bash
$ mcp-viz configure
```
## 🎨 Interactive Setup Flow
### Step 1: Welcome & Status Check
```
🎯 MCP Data Visualization Server
Transform natural language into beautiful charts with Claude Desktop
🔧 Setting up Claude Desktop integration...
📋 Current Configuration Status:
┌──────────────────┬────────────────────────────────────────┐
│ Setting │ Value │
├──────────────────┼────────────────────────────────────────┤
│ Platform │ Windows │
│ Config Path │ C:\Users\YourName\AppData\Roaming\... │
│ Config Exists │ ✅ Yes │
│ Config Valid │ ✅ Yes │
│ Existing Servers │ None │
└──────────────────┴────────────────────────────────────────┘
```
### Step 2: Database Location Selection
```
📁 Configuration Options:
💾 Database Location Options:
1. C:\Users\YourName\.mcp-visualization\data.duckdb ✅ (recommended)
2. C:\Users\YourName\Documents\mcp-viz.duckdb
3. C:\Users\YourName\Desktop\data.duckdb
4. C:\Github\mcp-visualization-duckdb\data.duckdb
5. Custom path
Choose database location [1]: 1
```
### Step 3: Python Path Detection
```
Python executable [C:\Github\mcp-visualization-duckdb\.venv\Scripts\python.exe]:
(Press Enter to use detected path, or type custom path)
```
### Step 4: Configuration Preview
```
📝 Configuration Preview:
┌──────────────┬──────────────────────────────────────────────────────┐
│ Setting │ Value │
├──────────────┼──────────────────────────────────────────────────────┤
│ Server Name │ data-viz-server │
│ Database │ C:\Users\YourName\.mcp-visualization\data.duckdb │
│ Python Path │ C:\Github\mcp-visualization-duckdb\.venv\Scripts\... │
│ Config File │ C:\Users\YourName\AppData\Roaming\Claude\... │
└──────────────┴──────────────────────────────────────────────────────┘
Proceed with configuration? [Y/n]: Y
```
### Step 5: Automatic Setup
```
🚀 Applying configuration...
✅ Created directory: C:\Users\YourName\.mcp-visualization
✅ Created backup: claude_desktop_config.json.backup
✅ Successfully configured 'data-viz-server' for Claude Desktop
✅ Created sample database with 3 tables
🔄 Next steps:
1. Restart Claude Desktop completely
2. Open a new conversation
3. Try: "What MCP servers are available?"
4. Try: "List available tables"
ℹ️ Your MCP Data Visualization Server is ready! 🎉
```
## 🎯 What This Replaces
### Before (Manual Setup)
Users had to manually:
1. Find Claude Desktop config file location
2. Edit JSON configuration by hand
3. Figure out correct paths for their system
4. Set environment variables
5. Create database manually
### After (Interactive Setup)
The CLI automatically:
1. **Detects platform** (Windows/Mac/Linux)
2. **Finds Claude config** file location
3. **Suggests smart paths** based on platform
4. **Creates directories** as needed
5. **Backs up existing** configuration
6. **Validates setup** before finishing
7. **Creates sample data** for immediate use
## 🔧 CLI Options
### Quick Setup (No Prompts)
```bash
mcp-viz configure --auto
# Uses all defaults, perfect for scripts
```
### Force Reconfiguration
```bash
mcp-viz configure --force
# Updates existing configuration
```
### Custom Parameters
```bash
mcp-viz configure --database-path ./my-data.duckdb --python-path /usr/bin/python3
# Pre-specify options
```
## 🛠️ Other CLI Commands
### Check Status
```bash
$ mcp-viz status
📊 MCP Data Visualization Server Status
🖥️ Platform Information:
┌──────────────────┬────────────────────────────────────────┐
│ Setting │ Value │
├──────────────────┼────────────────────────────────────────┤
│ Operating System │ Windows │
│ Config Path │ C:\Users\YourName\AppData\Roaming\... │
│ Config Directory │ ✅ Exists │
│ Config File │ ✅ Exists │
└──────────────────┴────────────────────────────────────────┘
🔍 Configuration Validation:
✅ Configuration valid with 1 server(s)
🔧 Configured MCP Servers:
1. data-viz-server
🧪 Server Import Test:
✅ Server code can be imported successfully
```
### Test Functionality
```bash
$ mcp-viz test
🧪 Testing MCP Data Visualization Server
Running: Import server module... ✅ PASS
Running: Claude Desktop config detection... ✅ PASS
Running: Database path creation... ✅ PASS
✅ All 3 tests passed! 🎉
```
### Create Sample Database
```bash
$ mcp-viz create-db
🗃️ Creating new database with sample data...
✅ Database created: C:\Users\YourName\.mcp-visualization\data.duckdb
ℹ️ Database includes sample tables: sales, customers, products
```
### Remove Configuration
```bash
$ mcp-viz remove
🗑️ Removing MCP server configuration...
Remove server 'data-viz-server' from Claude Desktop configuration? (y/N): y
✅ Successfully removed 'data-viz-server' from Claude Desktop
ℹ️ Please restart Claude Desktop to apply changes
```
## 🎨 Rich Terminal UI
The CLI uses the **Rich** library for beautiful terminal output:
- **Colored text** and emojis for better UX
- **Tables** for organized information display
- **Progress indicators** during setup
- **Interactive prompts** with smart defaults
- **Error handling** with helpful suggestions
## 🔄 Complete User Flow
```bash
# 1. Install package
pip install mcp-visualization-duckdb
# 2. Run interactive setup
mcp-viz configure
# (Beautiful interactive prompts guide user through setup)
# 3. Restart Claude Desktop
# 4. Start chatting with Claude:
# "What tables are available?"
# "Create a bar chart of sales by region"
# "Show me customer demographics"
```
**That's it!** No manual configuration, no environment variables, no complex setup. The CLI handles everything automatically with a beautiful, user-friendly interface. 🎉
