Skip to content

Commit 1cec5da

Browse files
AI Testing MCPclaude
AI Testing MCP
and
claude
committed
Initial AI Testing MCP Server implementation
- Core testing tools: analyze_codebase, generate_unit_tests, run_tests - Multi-language support: JavaScript/TypeScript, Python, Go - Framework auto-detection: Jest, Pytest, Mocha, etc. - AI-powered test generation with smart test cases - Complete Railway deployment configuration - Ready for production hosting 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
0 parents  commit 1cec5da

File tree

8 files changed

+981
-0
lines changed

8 files changed

+981
-0
lines changed

‎.gitignore‎

Lines changed: 11 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,11 @@
1+
node_modules/
2+
dist/
3+
*.log
4+
.env
5+
.env.local
6+
.env.production
7+
.DS_Store
8+
coverage/
9+
.nyc_output/
10+
*.tgz
11+
*.tar.gz

‎Dockerfile‎

Lines changed: 40 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,40 @@
1+
# Multi-stage build for smaller image
2+
FROM node:18-alpine AS builder
3+
4+
WORKDIR /app
5+
6+
# Copy package files
7+
COPY package*.json ./
8+
COPY tsconfig.json ./
9+
10+
# Install dependencies
11+
RUN npm ci --only=production
12+
13+
# Copy source code
14+
COPY src/ ./src/
15+
16+
# Build the application
17+
RUN npm run build
18+
19+
# Production stage
20+
FROM node:18-alpine AS production
21+
22+
WORKDIR /app
23+
24+
# Copy package files and install production dependencies
25+
COPY package*.json ./
26+
RUN npm ci --only=production && npm cache clean --force
27+
28+
# Copy built application
29+
COPY --from=builder /app/dist ./dist
30+
31+
# Create non-root user
32+
RUN addgroup -g 1001 -S nodejs
33+
RUN adduser -S mcp -u 1001
34+
35+
# Change to non-root user
36+
USER mcp
37+
38+
EXPOSE 3000
39+
40+
CMD ["npm", "start"]

‎README.md‎

Lines changed: 191 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,191 @@
1+
# AI Testing MCP Server
2+
3+
An AI-powered testing automation MCP server for Claude Code that provides TestSprite-like functionality.
4+
5+
## Features
6+
7+
### Core Testing Tools
8+
- **Code Analysis**: Analyze project structure and identify testable components
9+
- **Test Generation**: AI-powered unit, integration, and E2E test generation
10+
- **Test Execution**: Run tests with framework auto-detection
11+
- **Results Analysis**: Intelligent test result analysis with fix suggestions
12+
- **Framework Setup**: Initialize testing frameworks in projects
13+
14+
### Supported Languages & Frameworks
15+
- **JavaScript/TypeScript**: Jest, Mocha, Vitest
16+
- **Python**: Pytest
17+
- **Go**: Built-in testing
18+
- **Frameworks**: React, Vue, Express, FastAPI, Flask
19+
20+
## Installation
21+
22+
1. Clone or create the project:
23+
```bash
24+
git clone <repo> ai-testing-mcp
25+
cd ai-testing-mcp
26+
```
27+
28+
2. Install dependencies:
29+
```bash
30+
npm install
31+
```
32+
33+
3. Build the project:
34+
```bash
35+
npm run build
36+
```
37+
38+
## Configuration
39+
40+
### For Claude Code
41+
Add to your Claude Code MCP configuration:
42+
43+
```json
44+
{
45+
"mcpServers": {
46+
"ai-testing": {
47+
"command": "node",
48+
"args": ["dist/index.js"],
49+
"cwd": "/absolute/path/to/ai-testing-mcp"
50+
}
51+
}
52+
}
53+
```
54+
55+
## Usage
56+
57+
Once configured, you can use these tools in Claude Code:
58+
59+
### 1. Analyze Codebase
60+
```
61+
Use the analyze_codebase tool on my project at /path/to/project
62+
```
63+
64+
### 2. Generate Unit Tests
65+
```
66+
Generate unit tests for src/utils.js using Jest framework
67+
```
68+
69+
### 3. Run Tests
70+
```
71+
Run all tests in my project with coverage
72+
```
73+
74+
### 4. Analyze Test Results
75+
```
76+
Analyze these test failures and suggest fixes: [paste test output]
77+
```
78+
79+
## Available Tools
80+
81+
### `analyze_codebase`
82+
- **Purpose**: Scan project structure, detect languages/frameworks, find API endpoints
83+
- **Input**: `projectPath` - Path to project directory
84+
- **Output**: JSON with language, framework, API endpoints, testable components
85+
86+
### `generate_unit_tests`
87+
- **Purpose**: Create unit tests for functions/components
88+
- **Input**:
89+
- `filePath` - File to test
90+
- `functionName` - Specific function (optional)
91+
- `testFramework` - Framework to use
92+
- **Output**: Generated test code
93+
94+
### `generate_integration_tests`
95+
- **Purpose**: Create API endpoint tests
96+
- **Input**:
97+
- `projectPath` - Project directory
98+
- `apiSpec` - API documentation (optional)
99+
- **Output**: Integration test code
100+
101+
### `run_tests`
102+
- **Purpose**: Execute tests with auto-detection
103+
- **Input**:
104+
- `projectPath` - Project directory
105+
- `testPattern` - Specific tests to run (optional)
106+
- `framework` - Force specific framework (optional)
107+
- **Output**: Test results with coverage
108+
109+
### `analyze_test_results`
110+
- **Purpose**: Analyze test output and provide insights
111+
- **Input**: `testOutput` - Raw test results
112+
- **Output**: Analysis with suggestions
113+
114+
### `suggest_fixes`
115+
- **Purpose**: Generate fix suggestions for failing tests
116+
- **Input**:
117+
- `failureOutput` - Test failure output
118+
- `sourceCode` - Source code (optional)
119+
- `testCode` - Test code (optional)
120+
- **Output**: Suggested fixes
121+
122+
### `setup_testing_framework`
123+
- **Purpose**: Initialize testing in a project
124+
- **Input**:
125+
- `projectPath` - Project directory
126+
- `framework` - Testing framework
127+
- `language` - Programming language
128+
- **Output**: Setup instructions/files
129+
130+
## Example Workflow
131+
132+
1. **Analyze your project**:
133+
```
134+
Use analyze_codebase on ./my-app
135+
```
136+
137+
2. **Generate tests for a component**:
138+
```
139+
Generate unit tests for src/components/Button.jsx using Jest
140+
```
141+
142+
3. **Run the generated tests**:
143+
```
144+
Run tests in ./my-app matching *Button*
145+
```
146+
147+
4. **Fix any failures**:
148+
```
149+
Analyze these test failures and suggest fixes: [paste output]
150+
```
151+
152+
## Benefits Over Manual Testing
153+
154+
- **Speed**: Generate comprehensive test suites in seconds
155+
- **Coverage**: Automatically identifies all testable components
156+
- **Intelligence**: AI-powered test case generation with edge cases
157+
- **Integration**: Works with existing test frameworks and CI/CD
158+
- **Analysis**: Intelligent failure analysis with fix suggestions
159+
160+
## Development
161+
162+
### Build
163+
```bash
164+
npm run build
165+
```
166+
167+
### Development Mode
168+
```bash
169+
npm run dev
170+
```
171+
172+
### Adding New Features
173+
The server is modular - add new tools by:
174+
1. Adding tool definition to `ListToolsRequestSchema` handler
175+
2. Adding tool implementation to `CallToolRequestSchema` handler
176+
3. Creating helper methods as needed
177+
178+
## Comparison to TestSprite
179+
180+
| Feature | TestSprite | AI Testing MCP |
181+
|---------|------------|----------------|
182+
| Test Generation | ✅ | ✅ |
183+
| Multi-language | ✅ | ✅ |
184+
| Cloud Execution | ✅ | Local |
185+
| IDE Integration | ✅ | ✅ (via Claude Code) |
186+
| Cost | Subscription | Free |
187+
| Customization | Limited | Full source access |
188+
189+
## License
190+
191+
MIT

‎claude-code-config.json‎

Lines changed: 9 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,9 @@
1+
{
2+
"mcpServers": {
3+
"ai-testing": {
4+
"command": "node",
5+
"args": ["dist/index.js"],
6+
"cwd": "/path/to/ai-testing-mcp"
7+
}
8+
}
9+
}

‎package.json‎

Lines changed: 23 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,23 @@
1+
{
2+
"name": "ai-testing-mcp",
3+
"version": "1.0.0",
4+
"description": "AI-powered testing MCP server for Claude Code",
5+
"main": "dist/index.js",
6+
"type": "module",
7+
"scripts": {
8+
"build": "tsc",
9+
"start": "node dist/index.js",
10+
"dev": "tsc && node dist/index.js",
11+
"postinstall": "npm run build"
12+
},
13+
"dependencies": {
14+
"@modelcontextprotocol/sdk": "^0.5.0",
15+
"node:fs": "latest",
16+
"node:path": "latest",
17+
"node:child_process": "latest"
18+
},
19+
"devDependencies": {
20+
"@types/node": "^20.0.0",
21+
"typescript": "^5.0.0"
22+
}
23+
}

‎railway.toml‎

Lines changed: 10 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,10 @@
1+
[build]
2+
builder = "nixpacks"
3+
4+
[deploy]
5+
startCommand = "npm start"
6+
restartPolicyType = "on-failure"
7+
restartPolicyMaxRetries = 10
8+
9+
[variables]
10+
NODE_ENV = "production"

0 commit comments

Comments
 (0)