xcode-build-benchmark

The xcode-build-benchmark skill provides a structured workflow for benchmarking Xcode build performance using repeatable inputs and consistent build conditions. It is designed to establish reliable baseline measurements before any build optimization work begins. The skill focuses on collecting clean, cached clean, zero-change, and optional incremental build timings while preserving reproducibility across runs. By enforcing consistent build commands, destinations, configurations, caching behavior, and warm-up rules, it helps developers accurately compare build performance before and after project changes. All benchmark results are stored as timestamped JSON artifacts in a dedicated `.build-benchmark/` directory for future analysis and comparison.

The skill includes a defined benchmarking workflow that captures multiple build scenarios and summarizes performance using medians, minimums, maximums, and timing spread rather than relying on a single run. It supports both helper-script-based execution and equivalent manual `xcodebuild` workflows with timing summaries preserved. Additional capabilities include detection of compilation cache behavior, measurement of fixed per-build overhead through zero-change builds, optional incremental rebuild testing via touched source files, and preservation of raw build output for analysis. The workflow also includes safeguards for git worktree environments and explicitly avoids modifying project files during benchmarking.

This skill is intended for iOS and macOS developers, build engineers, and performance-focused teams who need accurate Xcode build timing baselines. Common use cases include evaluating whether build times improved after refactoring, diagnosing slow incremental builds, comparing branch performance, validating caching effectiveness, and generating reproducible performance artifacts for optimization specialists. It is especially useful in projects where build duration impacts developer productivity or continuous integration workflows.