#What it does

The /test-gen skill analyzes existing code and generates a comprehensive test suite covering happy paths, edge cases, error conditions, and boundary values. Unlike /tdd which writes tests before implementation, /test-gen works with code that already exists -- perfect for adding coverage to untested legacy code or filling gaps in an existing test suite.

#How to use

/test-gen src/lib/pricing.ts

/test-gen src/api/orders.ts --coverage 90

/test-gen src/components/CheckoutForm.tsx --type integration

#What it generates

#Test categories

• Happy path -- Standard inputs that exercise the primary code path
• Edge cases -- Empty arrays, zero values, single-character strings, maximum lengths
• Boundary values -- Off-by-one conditions, pagination limits, timeout thresholds
• Error conditions -- Invalid inputs, network failures, missing permissions, null references
• Concurrency -- Race conditions, duplicate submissions, stale data scenarios

#Test quality rules

• Tests describe behavior, not implementation: "should return discounted price for gold tier customers"
• Each test asserts one thing clearly
• No mocking of the unit under test -- only external dependencies
• Test data is explicit and readable, never hidden behind factory abstractions
• Cleanup and setup are minimal and obvious

#Workflow

1. Analyze -- Read the source file, extract all exported functions, identify code paths, branching logic, and external dependencies
2. Map Coverage -- Determine which paths need tests: conditionals, early returns, catch blocks, default cases
3. Generate -- Write tests organized by function and scenario, using the project's existing test framework and conventions
4. Deduplicate -- Check for existing tests that already cover these scenarios. Skip duplicates, fill gaps only
5. Verify -- Run the generated tests to confirm they pass against the current implementation

#Example

> /test-gen src/lib/discount.ts
 
## Generated 8 tests for calculateDiscount()
 
describe("calculateDiscount", () => {
  // Happy path
  it("should apply 10% discount for silver tier", ...);
  it("should apply 20% discount for gold tier", ...);
  it("should apply 0% discount for bronze tier", ...);
 
  // Edge cases
  it("should return 0 for zero price", ...);
  it("should handle fractional cents correctly", ...);
 
  // Boundary values
  it("should cap discount at max discount amount", ...);
 
  // Error conditions
  it("should throw for negative price", ...);
  it("should throw for unknown tier", ...);
});
 
## Coverage: 4 branches covered, 0 uncovered

#Configuration

• Detects your test framework automatically (Jest, Vitest, Mocha, Playwright, pytest)
• Follows your existing test file naming convention (*.test.ts, *.spec.ts, __tests__/)
• Respects your project's mocking patterns and test utilities