lc-test-coverage

name	lc:test-coverage
description	Analyze which models, controllers, and actions have tests and which don't.
argument-hint	[analyze | directory]
user-invocable	true
allowed-tools	Read Grep Bash Edit Write Glob

Test Coverage Analysis

Analyze which application classes have corresponding tests and which are untested. Provides a coverage map without requiring actual test execution.

Subcommands

Subcommand	Description
(no argument)	Analyze the full project for test coverage gaps. Read-only report.
[directory]	Analyze a specific directory (e.g., app/Models, app/Actions).

This is an analysis-only skill. It does not modify files. Use lc:generate-test to create missing tests.

Process

Step 1: Scan Application Classes

Use Glob to find all testable classes:

1. Models: app/Models/*.php
2. Controllers: app/Http/Controllers/**/*.php
3. Actions: app/Actions/**/*.php
4. Services: app/Services/*.php
5. Jobs: app/Jobs/*.php
6. Observers: app/Observers/*.php
7. Livewire/Volt components: resources/views/livewire/**/*.blade.php (only those with PHP class blocks)

If a [directory] argument is provided, limit scanning to that directory.

Step 2: Scan Test Files

Use Glob to find all test files:

1. Feature tests: tests/Feature/**/*.php
2. Unit tests: tests/Unit/**/*.php

Step 3: Match Classes to Tests

For each application class, check if a corresponding test exists:

1. By naming convention: App\Models\Property -> tests/Feature/Models/PropertyTest.php or tests/Unit/Models/PropertyTest.php
2. By class reference: Use Grep to search test files for references to the class (e.g., use App\Models\Property, Property::factory(), Property::create()
3. By method reference: Search test files for method names from the class being tested

Classify each class as:

• TESTED: A dedicated test file exists with meaningful test methods
• PARTIALLY_TESTED: The class is referenced in tests but has no dedicated test file (tested indirectly)
• UNTESTED: No test file and no references in any test

Step 4: Estimate Complexity

For untested classes, estimate testing priority based on:

1. Number of public methods: More methods = more to test = higher priority
2. Database interactions: Classes that create/update/delete data are higher priority
3. External service calls: Classes making API calls or sending emails
4. Business logic density: Actions and services with complex logic
5. User-facing: Controllers and Livewire components that users interact with directly

Score each untested class:

• High priority: Database mutations, user-facing, business logic, external services
• Medium priority: Read-only queries, internal helpers, observers
• Low priority: Simple data containers, enums, simple accessors

Step 5: Report

TEST COVERAGE ANALYSIS
=======================

SUMMARY
--------
Total application classes:  85
With dedicated tests:       23 (27%)
Partially tested:           12 (14%)
Untested:                   50 (59%)

BY CATEGORY
------------
Category         | Total | Tested | Partial | Untested | Coverage
-----------------|-------|--------|---------|----------|--------
Models           | 18    | 5      | 8       | 5        | 72%
Controllers      | 22    | 8      | 2       | 12       | 45%
Actions          | 15    | 3      | 1       | 11       | 27%
Services         | 8     | 2      | 0       | 6        | 25%
Jobs             | 12    | 3      | 0       | 9        | 25%
Observers        | 5     | 1      | 1       | 3        | 40%
Volt Components  | 5     | 1      | 0       | 4        | 20%

HIGH PRIORITY UNTESTED CLASSES
-------------------------------
These classes have the most impact and should be tested first:

1. App\Http\Controllers\PropertyController (8 public methods, DB mutations)
2. App\Actions\Property\ValidateToPublish (complex business logic)
3. App\Services\IdealistaService (external API integration)
4. App\Jobs\SyncPropertyToFotocasa (external service, queued)
5. App\Http\Controllers\OrganizationStripeWebhookController (payment handling)

FULLY TESTED CLASSES
---------------------
App\Models\User                      -> tests/Unit/Models/UserTest.php (12 tests)
App\Http\Controllers\AuthController  -> tests/Feature/AuthTest.php (8 tests)
...

UNTESTED CLASSES (by category)
-------------------------------
Models:
  - App\Models\Showcase (5 public methods)
  - App\Models\ShowcaseLogger (3 public methods)
  ...

Controllers:
  - App\Http\Controllers\DashboardController (4 methods)
  ...

Step 6: Recommendations

End with actionable recommendations:

RECOMMENDATIONS
================
1. Start with high-priority untested classes (listed above)
2. Use lc:generate-test to scaffold test files quickly
3. Focus on controllers with validation rules -- they catch the most bugs
4. Actions with business logic are the most valuable unit tests
5. Consider adding tests before refactoring any existing code

Notes

• This is static analysis. It counts test file existence and class references, not actual code coverage.
• For actual line-by-line coverage, run php artisan test --coverage (requires Xdebug or PCOV).
• Indirect testing (a class tested through another test) is common and valid, but dedicated tests are preferred for isolation.
• Volt components with only template code (no PHP class block) are excluded from the analysis.