Test Expansion

Analyze existing test patterns and create comprehensive, idiomatic tests for: $ARGUMENTS

Workflow

1. Detect Test Framework and Conventions

Search for test infrastructure and capture:

• Framework (see references/framework-templates.md for detection table)
• Directory structure and naming patterns
• Import style, assertion library, mocking patterns
• Setup/teardown conventions

2. Locate Existing Tests for Target Code

Find tests via import search and name matching. Analyze:

• Which functions/methods have tests?
• What scenarios are covered (happy path, errors, edge cases)?
• What's missing?

If no tests exist, note conventions from similar files and plan a comprehensive initial suite.

3. Analyze Target Code

Read and understand:

• Public API surface (exports, public methods)
• Input parameters and return types
• Side effects (I/O, state mutations, external calls)
• Error conditions and edge cases
• Dependencies that need mocking

4. Design Test Cases

For each function/method, consider:

Happy Path: Typical valid inputs, expected output, normal flow

Boundary Conditions: Empty inputs, single element, max/min values, Unicode

Error Cases: Invalid types, out-of-range values, missing params, external failures

Edge Cases: Concurrency, timing, resource exhaustion, partial failures

Naming: should <behavior> when <condition>

5. Write Tests Idiomatically

Match the project's existing style exactly — same imports, describe/it nesting,
assertion methods, setup/teardown, and mocking approach.

See references/framework-templates.md for per-framework templates.

6. Verify Tests

Run new tests and verify they:

• Pass when code is correct
• Fail when code has the bug they're testing
• Don't create false positives
• Run quickly (mock slow dependencies)

Run full suite to confirm no conflicts with existing tests.

7. Report Coverage Improvement

## Test Coverage Expansion

### Target: <file/module>
### Existing Coverage (before)
- Tests found: <count>
- Functions covered: <list>
- Gaps identified: <list>

### New Tests Added
| Test Name | Scenario | Type |
|-----------|----------|------|

### Coverage Improvement
- New tests: <count>
- Functions now covered: <list>
- Remaining gaps: <list or "none">

### Run Command
<command to run new tests>

Guidelines

Quality over quantity: One thing per test, independent, deterministic, fast.

Idiomatic is critical: New tests should look like existing tests.

Test behavior, not implementation: Tests should survive refactoring.

Handle missing infrastructure: If no tests exist, ask which framework,
set up minimal config, create foundational patterns, document how to run.

Error Handling

• No target specified: Ask user to specify file/function, or suggest analyzing recent changes.
• No framework detected: List options for the language, ask user to choose.
• Tests fail unexpectedly: Investigate if test is wrong or code has bug; report findings.