# Filament Desktop Testing

> Filament Desktop Testing is an Agent Skill for testing & quality published in the google/filament repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Execute unit tests and performance benchmarks for Filament on desktop platforms. Use this skill to run and filter tests or benchmarks locally.

- Skill name: `filament-desktop-testing`
- Category: [Testing & Quality](https://www.openparable.com/skill/testing-quality/1)
- Source: https://github.com/google/filament/blob/HEAD/skills/filament_desktop_testing/SKILL.md
- Repository: google/filament (20,258 GitHub stars when collected)

## SKILL.md

# Filament Desktop Testing and Benchmarking

This skill covers running tests and performance benchmarks on desktop platforms.

## 1. Executing a Unit Test

Run compiled unit test binaries directly from their respective build output directories.

### Example: Running utils unit tests (Debug)
```bash
./out/cmake-debug/libs/utils/test_utils
```

### Example: Running filament unit tests (Debug)
```bash
./out/cmake-debug/filament/test/test_filament
```

### Filtering Tests
Use standard Google Test (`gtest`) command-line parameters to run a subset of tests:
```bash
./out/cmake-debug/filament/test/test_filament --gtest_filter=Suite.TestName
```

---

## 2. Executing a Benchmark

Run compiled performance benchmarks from the release build output directories.

### Example: Running filament benchmarks (Release)
```bash
./out/cmake-release/filament/benchmark/benchmark_filament
```

### Filtering Benchmarks
Use standard Google Benchmark parameters to filter specific benchmarks:
```bash
./out/cmake-release/filament/benchmark/benchmark_filament --benchmark_filter=lutGeneration
```

---

Source: https://www.openparable.com/skill/testing-quality/filament-desktop-testing (OpenParable). Cite the canonical URL above.
