Basic Usage
Learn how to use SumoData Toolbox effectively in your daily workflow.
Three Ways to Use Tools
Method 1: Sidebar (Recommended)
The sidebar provides organized access to all tools.
- Click the SumoData icon in VS Code sidebar
- Expand the category you need:
- 📂 Sumo Pipes (Data Engineering)
- 📊 Sumo Lens (Data Analysis)
- 🤖 Sumo Core (DS/ML)
- Click the tool button
Pros:
- Visual organization
- See all available tools
- Quick model switching
Method 2: Context Menu (Fastest)
Right-click on selected code for quick access.
- Select code in your editor
- Right-click to open context menu
- Choose "🛡 SumoData"
- Select the tool
Pros:
- Fastest for frequent tasks
- Context-aware (only shows relevant tools)
- No need to switch to sidebar
Method 3: Command Palette
Use keyboard shortcuts for power users.
- Press
Ctrl+Shift+P(Windows/Linux) orCmd+Shift+P(Mac) - Type "SumoData"
- Select the tool from the list
Pros:
- Keyboard-driven workflow
- No mouse needed
- Searchable tool names
Working with Results
Viewing Results
Results open in a new editor tab beside your current file.
- Syntax highlighting based on detected language
- Side-by-side view for easy comparison
- Unsaved document so you can review before applying
Applying Results
After reviewing, you have three options:
1. Copy to Clipboard
Click "Copy to Clipboard" button to copy the entire result.
Result ready
[Copy to Clipboard] [Insert at Cursor]2. Insert at Cursor
Click "Insert at Cursor" to add result at your current cursor position.
3. Manual Selection
Select specific parts of the result and copy manually.
Common Workflows
Workflow 1: SQL Optimization
-- 1. Write your query
SELECT * FROM users WHERE name LIKE '%john%'
-- 2. Select the query
-- 3. Right-click → SumoData → SQL Optimizer
-- 4. Review optimized version
-- 5. Replace original with optimized codeWorkflow 2: Adding Documentation
# 1. Write your function
def calculate_metrics(data, threshold):
return [x for x in data if x > threshold]
# 2. Select the function
# 3. Sidebar → Sumo Core → Generate Docstring
# 4. Insert docstring above functionWorkflow 3: Understanding Legacy Code
-- 1. Find complex query you don't understand
-- 2. Select the entire query
-- 3. Right-click → SumoData → SQL Logic Explainer
-- 4. Read explanation in new tab
-- 5. Add comments to original codeSelection Tips
What to Select
✅ Good Selections:
- Complete functions or queries
- Logical code blocks
- Self-contained snippets
❌ Avoid:
- Partial statements
- Mixed languages
- Extremely large files (>10,000 chars)
No Selection?
If you don't select anything, the extension will prompt:
No code selected. Use entire document?
[Yes] [Cancel]This is useful for:
- Small files
- Single-function files
- Complete scripts
Model Selection
Choose the right model for your task:
Quick Tasks (Use Fast Models)
- ⚡ Sumo Lightning
- 💨 Sumo Flash
Examples:
- SQL Optimizer
- Cron Generator
- Type Hints
Complex Tasks (Use Powerful Models)
- 🔥 Sumo Titan Pro
- ⚡ Sumo Thunder
Examples:
- ML Boilerplate
- Complex SQL Explainer
- Advanced Pandas Cleaning
Change model:
- Open sidebar
- Use "AI Model" dropdown
- Select preferred model
Keyboard Shortcuts
While SumoData doesn't define custom shortcuts, you can create your own:
- Open Keyboard Shortcuts (
Ctrl+K Ctrl+S) - Search for "SumoData"
- Assign shortcuts to frequently used tools
Suggested shortcuts:
Ctrl+Alt+O- SQL OptimizerCtrl+Alt+D- Generate DocstringCtrl+Alt+T- Type Hinting
Best Practices
1. Review Before Applying
Always review AI-generated code before using it in production.
2. Start Small
Test tools on small code snippets before using on large files.
3. Use Appropriate Models
Fast models for simple tasks, powerful models for complex ones.
4. Keep Code Focused
Select only the relevant code, not entire files.
5. Iterate
If results aren't perfect, try:
- Different model
- Better code selection
- More context in the code
Troubleshooting Common Issues
"No active editor"
Solution: Open a file before using tools.
"Please select code"
Solution: Select code or choose to use entire document.
"Request timeout"
Solution:
- Check internet connection
- Increase timeout in settings
- Try a faster model
Results not as expected
Solution:
- Try a more powerful model
- Provide more context in code
- Select more complete code blocks