What problem does it solve? Performance profiles from V8 and Chrome DevTools are massive, dense JSON files that are nearly impossible to interpret by hand. This Skill turns raw .cpuprofile and Trace-*.json data into concrete answers: which functions consume the most CPU time, where long tasks block the main thread, and why one code path is slower than another. ## Core Features & Use Cases - CPU Profile Analysis: Parse .cpuprofile files to compute self-time and total-time per function, identify activity regions, and measure timing between milestones. - DevTools Trace Analysis: Process Trace-*.json files to inspect user timing marks, long tasks, layout/paint events, GC pressure, input latency, and multi-process activity across Renderer, Browser, and GPU processes. - Huge File Handling: Buffer-based parsing strategies for profiles exceeding V8's ~512MB string limit, so even gigabyte-scale traces can be analyzed. - Use Case: A user captures a DevTools trace of a slow VS Code window load and asks why startup takes 4 seconds. The Skill extracts user timing marks like code/didResolveTextFileEditorModel, lists long tasks over 50ms, and ranks functions by duration with source locations. ## Quick Start Analyze the attached Trace-2024.json file and tell me which functions consumed the most CPU time on the renderer main thread.
