Overview
LLMs can generate comprehensive test suites from source code. Provide the implementation, specify the test framework, and describe the expected behavior to get well-structured tests with good edge case coverage.
Test Generation Prompt Template
Prompt
Write tests for this function using [Vitest/Jest/pytest].
```typescript
[paste function here]
```
Cover:
1. Happy path with typical inputs
2. Edge cases: empty input, null/undefined, boundary values
3. Error cases: invalid input, network failures
4. Type safety: ensure TypeScript types are correct
Use describe/it blocks. One assertion per test.
Mock external dependencies with vi.mock().Test Categories
- •Happy path — typical inputs produce expected outputs
- •Edge cases — empty arrays, zero values, max integers, Unicode strings
- •Error handling — invalid inputs, missing data, network failures
- •Boundary values — off-by-one, min/max, pagination limits
- •Integration — API endpoints, database queries, external services
Coverage Strategy
Ask the LLM to identify untested code paths, generate tests for each branch, and suggest property-based tests for functions with large input spaces. For React components, ask for both unit tests (logic) and rendering tests (user interactions).