Show HN: Claude skill for Apple Instruments performance traces (iOS/Mac)(github.com)

Show HN: Claude skill for Apple Instruments performance traces (iOS/Mac)

https://github.com/jlreyes/instruments-analyzer

I was running into scroll performance issues with a mac app I was building a few months back and I got frustrated with the manual instrument → trace → diagnose → fix loop, so I built a way for claude to do it automatically.

Instruments Analyzer exports .trace files captured programmatically or via Instruments (file → save) into DuckDB with Parquet backing, giving your agents full SQL access to CPU profiling, animation hitches, SwiftUI view updates, Core Animation frame lifetimes, RunLoop activity, signposts, and 30+ other tables.

Key features:

1. Natural Language Queries. Since it’s just a skill, you can ask “why did scroll lag in this trace” rather than dig through WWDC videos to understand which instruments to use and how to use them.

2. Derived views for analysis: Frame-level analysis, cascade analysis, and per-frame attribution are available via prepared views (prepare_analysis.py), making it easy to query performance metrics without manual preprocessing.

3. Scroll & animation jank workflow: Optional workflow for diagnosing scroll lag or animation stutter, only loads when relevant.

Install as a Claude Code skill or with whatever agent you’re building with.

https://github.com/jlreyes/instruments-analyzer

3 comments

HelloMCP3 mesi fa

This feels like a preview of where debugging is going. Dashboards are good for people but we also need to make raw telemetry directly queryable by agents.

Curious how well this holds up across different traces. Instruments data can be pretty noisy and context-dependent. @jlreyes - do the derived views generalize, or do you end up encoding a lot of assumptions about what “jank” looks like?

jlreyes3 mesi fa

Identifying bottlenecks is pretty generalizable. There is a distinction the skill tries to draw between targeting median FPS and P95, but from there the AI is quite good at narrowing to the relevant data.

Where the AI trips up is getting distracted by aggregate signals instead of digging deep into root causing specific frame drops, but I see humans and existing tooling getting distracted by that too.

Root causes are often context-dependent, but they tend to cluster into a handful of common issues. If you're able to enable the new swiftui instrument (from WWDC 2025), the entire attribute graph is encoded, and it can get you to the precise issue quite well.

silbercue3 mesi fa

been eyeing Instruments integration for my own iOS tooling for months and kept bouncing off it. xctrace output + parsing traces felt like a rabbit hole. DuckDB + Parquet is a way nicer angle than what I had in mind.

> Where the AI trips up is getting distracted by aggregate signals

this shows up everywhere in agent loops. anytime I hand Claude a wide slice of anything it starts reasoning in averages. the moment I narrow to ~10 rows it locks onto the actual root cause. so derived views sound like exactly the right shape. how big do real traces get in your duckdb format? does a 5min scroll-heavy trace stay manageable =)