Skill: Mutation Testing

Overview

This skill runs mutation testing to measure test suite quality:

• Injects mutations (bugs) into code
• Runs tests against mutations
• Measures mutation score (killed vs survived)
• Identifies weak tests

Purpose

Verify that tests actually catch bugs. Use this skill:

• After writing new tests
• Before major releases
• To identify test gaps
• For test quality improvement

Execution Flow

1. ANALYZE codebase structure
2. CONFIGURE mutation testing
3. GENERATE mutations
4. RUN tests against each mutation
5. CALCULATE mutation score
6. REPORT surviving mutants
7. SUGGEST test improvements

Input Parameters

Parameter	Type	Description	Required
files	string[]	Files to mutate	No
threshold	number	Minimum mutation score	No (default: 80)
reporters	string[]	Report formats	No
concurrency	number	Parallel processes	No

Stryker Configuration

JavaScript/TypeScript

// stryker.config.js
module.exports = {
    packageManager: 'npm',
    reporters: ['html', 'clear-text', 'progress', 'dashboard'],
    testRunner: 'jest',
    coverageAnalysis: 'perTest',
    mutate: ['src/**/*.ts', '!src/**/*.spec.ts'],
    thresholds: {
        high: 80,
        low: 60,
        break: 70,
    },
    jest: {
        projectType: 'custom',
        config: require('./jest.config.js'),
    },
};

Python

# pyproject.toml
[tool.mutmut]
paths_to_mutate = ["src/"]
runner = "pytest -x"
tests_dir = ["tests/"]

Java

<!-- pom.xml -->
<plugin>
    <groupId>io.stryker-mutator</groupId>
    <artifactId>stryker4s-maven-plugin</artifactId>
    <configuration>
        <mutate>
            <element>src/main/java/**/*.java</element>
        </mutate>
        <thresholds>
            <high>80</high>
            <low>60</low>
            <break>70</break>
        </thresholds>
    </configuration>
</plugin>

Mutation Operators

Operator	Description	Example
Arithmetic	Change operators	+ → -
Conditional	Change conditions	> → >=
Equality	Change operators	== → !=
Boolean	Invert values	true → false
String	Change strings	"hello" → ""
Array	Change lengths	.length → .length - 1
Object	Remove calls	obj.method() → ""

Output Format

# Mutation Testing Report

**Date:** [timestamp]
**Framework:** Stryker
**Duration:** 12m 34s

## Mutation Score

| Metric | Value | Target |
|--------|-------|--------|
| **Mutation Score** | **85%** | 80% ✅ |
| Killed | 170 | - |
| Survived | 20 | - |
| Timeout | 5 | - |
| No Coverage | 5 | - |
| **Total Mutants** | **200** | - |

## File Scores

| File | Score | Killed | Survived |
|------|-------|--------|----------|
| auth.service.ts | 95% | 57 | 3 |
| user.service.ts | 82% | 45 | 10 |
| payment.service.ts | 78% | 39 | 11 |
| email.service.ts | 88% | 35 | 5 |

## Surviving Mutants (Need Tests)

### 1. auth.service.ts:45

**Original:**
```typescript
if (user.isActive) {
    return generateToken(user);
}

Mutated:

if (false) {  // Changed: user.isActive → false
    return generateToken(user);
}

Test Needed: Verify that inactive users cannot get tokens

Suggested Test:

it('should reject inactive users', async () => {
    user.isActive = false;
    await expect(authService.login(user))
        .rejects.toThrow('User is inactive');
});

2. payment.service.ts:78

Original:

const fee = amount * 0.03;

Mutated:

const fee = amount * 0;  // Changed: 0.03 → 0

Test Needed: Verify fee calculation

Suggested Test:

it('should calculate correct fee', () => {
    const fee = paymentService.calculateFee(100);
    expect(fee).toBe(3);  // 3% of 100
});

Ignored Mutants

Reason	Count
Equivalent mutation	5
Not worth testing	3

Recommendations

1. Add tests for payment.service.ts fee calculation
2. Add edge case tests for user.service.ts
3. Increase coverage for error paths

## Example Usage

### Example 1: Run Full Mutation Testing

USER: /mutation-test

SKILL OUTPUT:

Mutation Testing Report

Duration: 12m 34s

Mutation Score: 85% ✅

Category	Count
Killed	170
Survived	20
Timeout	5
No Coverage	5
Total	200

20 Surviving Mutants

Top Priority Fixes

1. auth.service.ts:45 - Inactive user check bypassed
2. payment.service.ts:78 - Fee calculation zeroed
3. user.service.ts:102 - Email validation removed

View Report: reports/mutation/html/index.html

### Example 2: Test Specific File

USER: /mutation-test auth.service.ts

SKILL INPUT: { "files": ["src/services/auth.service.ts"] }

SKILL OUTPUT:

Mutation Testing Report

File: auth.service.ts Duration: 2m 15s

Mutation Score: 95%

Mutant Type	Killed	Survived
Conditional	25	2
Equality	18	1
Boolean	14	0
Total	57	3

3 Surviving Mutants

1. Line 45: Inactive User Check

// Original
if (user.isActive) {

// Mutated
if (false) {

Add Test:

it('should reject inactive users', async () => {
    const inactiveUser = { ...user, isActive: false };
    await expect(service.login(inactiveUser))
        .rejects.toThrow('User is inactive');
});

### Example 3: Set Threshold

USER: /mutation-test --threshold 90

SKILL INPUT: { "threshold": 90 }

SKILL OUTPUT:

Mutation Testing Report

Threshold: 90% Actual Score: 85%

❌ Build Failed: Mutation score below threshold

Missing 5% (10 mutants)

Priority fixes needed:

1. Add fee calculation test
2. Add inactive user test
3. Add email validation test

Run npx stryker run to see full report.

## Running Mutation Tests

### JavaScript/TypeScript
```bash
npx stryker run

Python

mutmut run

Java

mvn stryker4s:run

.NET

dotnet stryker

CI/CD Integration

# .github/workflows/mutation.yml
name: Mutation Testing

on: [push, pull_request]

jobs:
    mutation:
        runs-on: ubuntu-latest
        steps:
            - uses: actions/checkout@v4
            - uses: actions/setup-node@v4
            - run: npm ci
            - run: npx stryker run
              env:
                  STRYKER_DASHBOARD_API_KEY: ${{ secrets.STRYKER_TOKEN }}

Best Practices

1. Start Small - Run on changed files first
2. Set Thresholds - Enforce minimum scores
3. Fix Survivors - Add tests for surviving mutants
4. Ignore Equivalents - Some mutations are equivalent
5. Parallelize - Use concurrency for speed

Integration

This skill integrates with:

• coverage-analysis: Combined quality metrics
• test-generation: Generate missing tests
• CI/CD for automated quality gates