Database Tools

MyDeskBot Desktop provides powerful database tools that allow you to interact with databases directly from your AI assistant.

Features

Connect to Databases

MyDeskBot supports multiple database connections:

• PostgreSQL
• MySQL
• SQLite
• Microsoft SQL Server

Database Operations

The AI assistant can perform the following database operations:

• Query data with natural language
• Insert, update, and delete records
• Create and modify tables
• Run custom SQL queries
• Export query results
• Visualize data

Integration with AI

The database tools are deeply integrated with the AI assistant:

User: "Show me all users who registered in the last week"

AI: [Automatically generates SQL query]
SELECT * FROM users WHERE created_at >= NOW() - INTERVAL '7 days';

Setup

1. Configure Database Connections

1. Open MyDeskBot Desktop
2. Go to Settings → Database
3. Click "Add Connection"
4. Fill in the connection details:
   • Database type
   • Host address
   • Port number
   • Database name
   • Username and password

2. Test Connection

After configuring, click "Test Connection" to verify the database is accessible.

Usage Examples

Natural Language Queries

User: "How many orders were placed this month?"

AI: [Queries the database]
"1,247 orders were placed this month."

Data Analysis

User: "Analyze sales trends over the past year"

AI: [Runs multiple queries and creates visualization]
"Here's a breakdown of your sales trends..."

Data Export

User: "Export all users who subscribed to the premium plan"

AI: [Queries and exports data]
"Exported 842 records to CSV file."

Security

Connection Security

• All database connections use SSL/TLS encryption
• Credentials are stored securely in your keychain
• Connections can be configured to require re-authentication

AI Safety

• The AI cannot execute destructive operations without confirmation
• Write operations (INSERT, UPDATE, DELETE) require explicit approval
• Destructive operations (DROP, TRUNCATE) require double confirmation

Best Practices

1. Use Descriptive Table Names

-- Good
CREATE TABLE customer_orders (
  id SERIAL PRIMARY KEY,
  customer_id INTEGER,
  order_date TIMESTAMP
);

-- Avoid
CREATE TABLE t1 (
  id INT,
  c_id INT,
  date DATETIME
);

2. Add Indexes for Performance

CREATE INDEX idx_orders_date ON customer_orders(order_date);

3. Use Prepared Statements

The AI assistant automatically uses prepared statements to prevent SQL injection.

4. Limit Query Results

User: "Show me the first 100 users"

AI: [Executes SELECT * FROM users LIMIT 100]

Troubleshooting

Connection Failed

• Verify database credentials
• Check if the database is running
• Ensure firewall allows the connection
• Check SSL/TLS settings

Slow Queries

• Add appropriate indexes
• Optimize WHERE clauses
• Consider query caching
• Use EXPLAIN to analyze query plans

Permission Denied

• Verify user has necessary permissions
• Check database role assignments
• Ensure user has access to the specific database/schema

Advanced Features

Query Templates

Save frequently used queries as templates:

-- Template: Get active users
SELECT COUNT(*) FROM users WHERE last_active > NOW() - INTERVAL '7 days';

Query History

All queries executed through the AI are logged and can be reviewed in Settings → Database → History.

Custom AI Rules

Create rules for how the AI interacts with your database:

{
  "databaseRules": {
    "allowDestructive": false,
    "requireConfirmationForWrites": true,
    "maxResultLimit": 10000,
    "allowedTables": ["users", "orders", "products"]
  }
}

See Also

• SSH/SFTP Tools
• Configuration Guide