benchmark

Performance regression detection in CI workflows requires recording baseline metrics and comparing them against current code changes. This skill stores baselines in a centralized repository and uses compare commands to identify drift across page, API, and build performance.

Build performance baseline measurement captures current project metrics to establish a standard reference point. You need it when performance predictability is critical, allowing you to quantify the impact of feature changes and prevent drift across multiple code stacks.

To compare before and after performance metrics across multiple stacks, run a baseline benchmark first, then apply changes and use the compare command. This quantifies the impact of modifications on page, API, and build performance before delivery.

Yes, custom instrumentation to record metrics is required for benchmarking. You must instrument your code to emit the necessary performance data so the skill can capture baselines and execute accurate regression comparisons.

Performance baselines for regression detection are stored in the .ecc/benchmarks/ directory. This centralized store preserves your captured metrics, enabling consistent before and after comparisons throughout your development cycle.

The best way to automate regression detection for API performance is integrating baseline and compare commands into your CI workflows. This automates metric collection and comparison, instantly flagging regressions before changes are merged.