Test Code Generator
Generates Vitest test code from JSON specifications using boundary value analysis and equivalence partitioning.
Test Code Generator MCP Server
An MCP server that generates Vitest test code based on boundary value analysis and equivalence partitioning from JSON specifications.
Features
- Boundary Value Analysis: Automatically generates boundary values and their adjacent values for continuous ranges
- Multiple Range Support: Handles discontinuous valid ranges (e.g., 0-10, 20-30, 40-50)
- Equivalence Partitioning: Includes all valid and invalid values from categorical data
- Pairwise Testing: Generates efficient test combinations using pairwise method
- Vitest Format: Generates clean test code using
test.eachpattern - Specification Guide: Provides prompts to help create JSON specifications
- Language-Optimized: All processing and output in English for optimal LLM performance
Quick Start
Using npx (No installation required)
# Run directly with npx
npx testing-mcp-vitest
# Configure Claude Desktop as shown below
See QUICKSTART.md for detailed setup instructions.
Installation
Using npx (Recommended)
npx testing-mcp-vitest
Local Installation
# Clone repository
git clone <repository-url>
cd test-code-generator-mcp
# Install dependencies
npm install
# Build
npm run build
Usage
Claude Desktop Configuration
Add to your claude_desktop_config.json:
{
"mcpServers": {
"test-generator": {
"command": "npx",
"args": ["testing-mcp-vitest"]
}
}
}
Config file locations:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
Available Tools
1. generate-test-code
Generates test code from JSON specification.
2. Prompts
create-test-spec: Explains how to create test specificationsboundary-value-example: Provides boundary value analysis examplesequivalence-class-example: Provides equivalence partitioning examples
Specification Format
Basic structure with single range:
{
"type": "model",
"name": "UserRegistration",
"columns": [
{
"name": "age",
"type": "number",
"valueType": "continues",
"values": {
"min": 18,
"max": 120
}
},
{
"name": "userType",
"type": "string",
"valueType": "category",
"values": {
"valid": ["admin", "user", "guest"],
"invalid": ["superuser", "root", ""]
}
}
]
}
Multiple ranges example:
{
"name": "workHours",
"type": "number",
"valueType": "continues",
"values": {
"ranges": [
{ "min": 9, "max": 12 }, // Morning shift
{ "min": 14, "max": 18 } // Afternoon shift
]
}
}
Generated Test Code Example
import { describe, test, expect } from 'vitest';
describe('Tests for UserRegistration', () => {
// Valid cases
test.each([
{ inputs: { age: 18, userType: "admin" }, description: 'Combination 1: Valid case' },
{ inputs: { age: 69, userType: "user" }, description: 'Combination 2: Valid case' },
{ inputs: { age: 120, userType: "guest" }, description: 'Combination 3: Valid case' },
// ... more test cases
])('Valid case: $description', ({ inputs }) => {
// TODO: Import the function or model to test
// import { UserRegistration } from './path/to/UserRegistration';
// TODO: Execute the test subject
// const result = UserRegistration(inputs);
// TODO: Check expected values
// expect(result).toBeDefined();
// expect(result.error).toBeUndefined();
});
// Invalid cases
test.each([
{ inputs: { age: 17, userType: "admin" }, description: 'Combination 4: Invalid case' },
{ inputs: { age: 121, userType: "user" }, description: 'Combination 5: Invalid case' },
{ inputs: { age: 50, userType: "superuser" }, description: 'Combination 6: Invalid case' },
{ inputs: { age: 50, userType: "root" }, description: 'Combination 7: Invalid case' },
{ inputs: { age: 50, userType: "" }, description: 'Combination 8: Invalid case' },
// ... more test cases
])('Invalid case: $description', ({ inputs }) => {
// TODO: Import the function or model to test
// import { UserRegistration } from './path/to/UserRegistration';
// TODO: Verify that an error occurs
// expect(() => UserRegistration(inputs)).toThrow();
// or
// const result = UserRegistration(inputs);
// expect(result.error).toBeDefined();
});
});
Key Features
Full Coverage
- All valid and invalid equivalence class values are included in test cases
- Boundary values include: min-1, min, min+1, middle, max-1, max, max+1
- For multiple ranges, includes gap values between ranges
Efficient Combinations
- 3 or fewer parameters: generates all combinations
- 4 or more parameters: uses pairwise method for efficiency
Multiple Range Support
Useful for:
- Working hours (9-12, 14-18)
- Valid scores (0-40 passing, 60-100 honors)
- Temperature ranges (different valid ranges for different conditions)
- Time slots (morning, afternoon, evening sessions)
Examples
See the examples/ directory for more specification examples:
user-registration.json- Basic model validationcalculate-discount.json- Business logic with multiple parametersmulti-range-validation.json- Complex scheduling with multiple rangesjapanese-postal-code.json- Regional validation patterns
Development
# Development mode
npm run dev
# Build
npm run build
# Production mode
npm start
# Debug mode
DEBUG=1 npm start
Publishing to npm
# Build and publish
npm publish
Contributing
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add some amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
License
MIT - see LICENSE for details
Support
- GitHub Issues: [Report bugs or request features](/issues)
- Discussions: [Ask questions or share ideas](/discussions)
相关服务器
Alpha Vantage MCP Server
赞助Access financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
BlenderMCP
Connects Blender to Claude AI via the Model Context Protocol (MCP), enabling direct interaction and control for prompt-assisted 3D modeling, scene creation, and manipulation.
Wopee MCP
AI testing agents for web apps — dispatch test runs, analysis crawls, and AI agent tests, fetch artifacts and project status
Vercel AI SDK MCP Server Project
An MCP server for the Vercel AI SDK, enabling integrations with Figma and 21st.dev Magic.
Authless Remote MCP Server
An authentication-free, remote MCP server designed for deployment on Cloudflare Workers.
QGIS
connects QGIS Desktop to Claude AI through the MCP. This integration enables prompt-assisted project creation, layer loading, code execution, and more.
PageBolt
Take screenshots, generate PDFs, and create OG images from your AI assistant
Terry-Form MCP
Execute Terraform commands locally in a secure, containerized environment. Features LSP integration for intelligent Terraform development.
GitHub Workflow Debugger MCP
Diagnose and fix GitHub Actions workflow failures using the GitHub API.
GrowthBook
Create and read feature flags, review experiments, generate flag types, search docs, and interact with GrowthBook's feature flagging and experimentation platform.
Bedrock Server Manager
Manage your Bedrock server with natural language commands using the Bedrock Server Manager API.