| name | kumo-assistant |
| description | Kumo development assistant for MySQL database management tool. Use when working on Kumo features, understanding architecture, writing tests, or navigating the codebase. Helps with React components, API endpoints, database features, and Electron app development. |
Kumo Development Assistant
Expert assistant for developing Kumo - a comprehensive cross-platform MySQL/MongoDB database management tool built with TypeScript, React, and Electron.
When to Use This Skill
Use this skill when:
- Adding new features to Kumo
- Understanding the feature-based architecture
- Working with database connections (MySQL/MongoDB)
- Creating or modifying UI components
- Writing API endpoints
- Adding tests (Vitest unit tests or Playwright E2E tests)
- Working with the Electron desktop app
- Understanding project conventions and patterns
Quick Start Guide
Project Overview
Kumo is a cross-platform database management application similar to Navicat, featuring:
- Database Support: MySQL and MongoDB with multi-connection management
- Query Tools: Monaco editor with SQL syntax highlighting, visual query builder (React Flow)
- Data Management: Advanced data viewer with inline editing, import/export (CSV, JSON, SQL, Excel)
- Development Tools: API tester, terminal, Playwright integration, Git browser
- AI Features: Anthropic Claude integration for query assistance
Tech Stack:
- Frontend: React 18, TypeScript, TanStack Query, Tailwind CSS, Monaco Editor
- Backend: Node.js, Express, mysql2, mongodb driver, socket.io
- Desktop: Electron with secure IPC
- Testing: Vitest (unit), Playwright (E2E)
- State: Zustand (client), TanStack Query (server)
Architecture Patterns
Three-Tier Architecture:
React Frontend → Node.js API Server → MySQL/MongoDB Database
Key Principles:
- Feature-based folder structure in
src/features/ - Server-side connection pooling (never direct database access from browser)
- RESTful API with WebSocket for streaming operations
- Separation of server state (TanStack Query) and client state (Zustand)
- TypeScript path aliases:
@/*forsrc/*,@server/*forserver/*
For detailed architecture information, see architecture.md.
Common Development Tasks
1. Adding a New Feature
Location: src/features/<feature-name>/
Pattern: Each feature is self-contained with:
src/features/myFeature/
├── MyFeatureComponent.tsx # Main component
├── components/ # Sub-components
│ ├── SubComponent1.tsx
│ └── SubComponent2.tsx
└── utils/ # Feature-specific utilities
└── helpers.ts
Steps:
- Create feature folder in
src/features/ - Build React components using functional components + hooks
- Use Zustand for local feature state if needed
- Use TanStack Query for server data fetching
- Follow Tailwind CSS for styling
- Add tests in
src/test/ortests/
See features.md for detailed feature development guide.
2. Creating API Endpoints
Location: server/routes/
Pattern:
// server/routes/myRoute.ts
import { Router } from 'express';
const router = Router();
router.get('/api/my-endpoint', async (req, res) => {
try {
// Business logic here
res.json({ success: true, data: result });
} catch (error) {
res.status(500).json({ error: error.message });
}
});
export default router;
Register in server/index.ts:
import myRoute from './routes/myRoute';
app.use(myRoute);
3. Database Operations
MySQL Example:
import { query } from '@server/utils/database';
const results = await query(connectionId, 'SELECT * FROM users WHERE id = ?', [userId]);
MongoDB Example:
import { getMongoClient } from '@server/services/mongodb';
const client = await getMongoClient(connectionId);
const db = client.db(databaseName);
const collection = db.collection(collectionName);
const docs = await collection.find({ status: 'active' }).toArray();
Important: Always use parameterized queries to prevent SQL injection.
4. Adding Tests
Unit Tests (Vitest):
npm run test:unit # Run all unit tests
npm run test:unit:ui # Run with UI
Location: src/test/components/
E2E Tests (Playwright):
npm run test # Run all E2E tests
npm run test:ui # Run with Playwright UI
npm run test:headed # Run in headed mode
Location: tests/
See testing.md for comprehensive testing guide.
Project Structure
Kumo/
├── src/ # Frontend React application
│ ├── features/ # Feature modules (20+ features)
│ │ ├── apiTester/ # REST API testing
│ │ ├── connections/ # Database connection management
│ │ ├── dataViewer/ # Data grid with inline editing
│ │ ├── mongodb/ # MongoDB features
│ │ ├── query/ # SQL editor and execution
│ │ ├── queryBuilder/ # Visual query builder
│ │ ├── schema/ # Schema exploration
│ │ └── ... # 13+ more features
│ ├── components/ # Shared UI components
│ ├── hooks/ # Custom React hooks
│ ├── services/ # Frontend API clients
│ ├── store/ # Zustand state management
│ ├── types/ # TypeScript type definitions
│ └── utils/ # Utility functions
├── server/ # Node.js Express API
│ ├── routes/ # API route handlers
│ ├── services/ # Business logic
│ ├── types/ # Server types
│ └── utils/ # Server utilities
├── electron/ # Electron main process
│ ├── main.cjs # Entry point
│ └── preload.js # Preload script
├── tests/ # E2E tests (Playwright)
├── openspec/ # Spec-driven development
│ ├── project.md # Project conventions
│ ├── specs/ # Feature specifications
│ └── changes/ # Change proposals
└── .claude/ # Claude Code configuration
├── commands/ # Slash commands
└── skills/ # Agent skills
Code Conventions
TypeScript Style
- 2-space indentation
- Single quotes for strings
- Trailing commas in multi-line objects/arrays
- PascalCase for components, camelCase for functions/variables
- kebab-case for file names
React Patterns
- Functional components with hooks (no class components)
- Props destructuring in component signatures
- Custom hooks for reusable logic (prefix with
use) - Memoization with
useMemo/useCallbackfor expensive operations
State Management
- Server State: Use TanStack Query (
useQuery,useMutation) - Client State: Use Zustand stores for UI state
- Form State: Use controlled components with local
useState
API Patterns
- RESTful endpoints with proper HTTP methods (GET, POST, PUT, DELETE)
- Error handling with try-catch and appropriate status codes
- Parameterized queries to prevent SQL injection
- Connection pooling for database connections
Development Workflow
Running the Application
# Terminal 1: Frontend (Vite dev server)
npm run dev # http://localhost:5174
# Terminal 2: Backend API
npm run dev:server # http://localhost:3001
# Terminal 3 (optional): Electron app
npm run dev:electron
Building for Production
# Build frontend and server
npm run build
# Build Electron app for specific platform
npm run build:electron:win # Windows
npm run build:electron:mac # macOS
npm run build:electron:linux # Linux
Code Quality
npm run format # Format code with Prettier
npm run type-check # TypeScript type checking
npm run lint # ESLint code linting
npm run clean # Clean build artifacts
Key Features Overview
Database Connections (
src/features/connections/)- Multi-database connection management
- Secure credential storage
- SSH tunnel support
Query Editor (
src/features/query/)- Monaco editor with SQL syntax highlighting
- Query execution and results display
- Query history and saved queries
- AI-powered query suggestions
Visual Query Builder (
src/features/queryBuilder/)- Drag-and-drop query construction
- React Flow for visual representation
- Smart join suggestions based on foreign keys
Data Viewer (
src/features/dataViewer/)- TanStack Table for virtualized data grid
- Inline editing with validation
- Pagination, filtering, sorting
Schema Management (
src/features/schema/)- Visual schema browser
- Relationship diagrams
- Index management
API Tester (
src/features/apiTester/)- REST API testing
- cURL parsing and generation
- Request/response management
Git Integration (
src/features/git/)- Repository browser
- Pull request viewer
- Code diff viewer
Security Considerations
- Never expose credentials: Store in OS keychain (desktop) or encrypted (web)
- Parameterized queries: Always use placeholders for user input
- Connection limits: Max 10 connections per database
- Input validation: Validate all user input on both client and server
- Secure IPC: Electron preload script for secure communication
Performance Guidelines
- Large result sets: Warn at 10k rows, require confirmation for 100k+
- Virtual scrolling: Use TanStack Virtual for large lists
- Code splitting: Lazy load features with React.lazy()
- Streaming: Use streaming for large file imports/exports
- Connection pooling: Reuse database connections efficiently
OpenSpec Integration
Kumo uses OpenSpec for spec-driven development. See openspec/AGENTS.md for:
- Creating change proposals
- Writing spec deltas
- Implementing changes
- Archiving completed work
Getting Help
Documentation:
- architecture.md - Detailed architecture patterns
- features.md - Feature development guide
- testing.md - Testing strategies and examples
openspec/project.md- Project conventionsREADME.md- Setup and usage guide
Commands:
openspec list # View active changes
openspec list --specs # View feature specifications
npm run test:report # View test reports
Tips for AI Assistants
- Check existing features first: Similar functionality may already exist
- Follow feature structure: Keep features self-contained in
src/features/ - Use path aliases: Import with
@/*instead of relative paths - Write tests: Add tests when creating new features
- Update specs: Use OpenSpec for significant changes
- Security first: Validate input, use parameterized queries
- Performance aware: Consider virtual scrolling for large data sets
- Type safety: Leverage TypeScript for better DX
Next Steps
- Read architecture.md for in-depth technical patterns
- Review features.md before adding new features
- Check testing.md for testing best practices
- Explore existing features in
src/features/for examples - Review
openspec/project.mdfor project conventions