Skip to main content
skill CX Codex

xcode-compilation-analyzer — Claude Code skill: what it does, install

A skill that explains why an Xcode build drags: it reads build timings and Swift frontend diagnostics, finds the expensive spots in your code, and returns a fix plan.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add AvdLee/Xcode-Build-Optimization-Agent-Skill --skill xcode-compilation-analyzer -a claude-code
Open on GitHub

What xcode-compilation-analyzer does

The skill takes over when compile time, rather than project setup, is what hurts. It walks the Build Timing Summary from clean and incremental builds, checks long CompileSwiftSources tasks along with SwiftEmitModule and Planning Swift module time, then runs scripts/diagnose_compilation.py for a ranked list of functions and expressions that type-check slowly.

You need Python 3 and an Xcode build that actually runs: the script takes --project, --scheme, --configuration, --destination, a millisecond --threshold, and writes its output into .build-benchmark. Hand it no recent timing report and the analysis turns into guesswork, so save one first.

What comes back is a plan, not a patch. Each item lists the evidence, the file or module it touches, the expected effect on wall-clock wait, and how confident the skill is about it. Nothing in your source or build settings changes without your approval. When the trail leads to project configuration instead, it hands off to xcode-project-analyzer.

What it helps with

  • find the files that dominate clean-build compile time
  • rank functions with slow type checking
  • explain why SwiftEmitModule balloons on incremental builds
  • separate source hotspots from project configuration problems
  • get an approval-gated fix list instead of silent edits

Example prompts for the agent

  • Why does the clean build of MyApp take so long? Pull the timing summary and show me the slowest files.
  • Run the compilation diagnostics on App.xcodeproj with scheme MyApp and rank the type-checking hotspots.
  • Read .build-benchmark and tell me which of these fixes would actually cut build wait time

How to install xcode-compilation-analyzer

1Skills CLI (Vercel)

npx skills add AvdLee/Xcode-Build-Optimization-Agent-Skill --skill xcode-compilation-analyzer -a claude-code

2Manually: put the files in place

~/.claude/skills/xcode-compilation-analyzer/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/xcode-compilation-analyzer/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about xcode-compilation-analyzer

How do I install xcode-compilation-analyzer in Claude Code?

Run npx skills add AvdLee/Xcode-Build-Optimization-Agent-Skill --skill xcode-compilation-analyzer -a claude-code. The skill file then sits at ~/.claude/skills/xcode-compilation-analyzer/SKILL.md.

What does xcode-compilation-analyzer do?

It hunts down the causes of slow Swift compilation: reads build timings, runs frontend diagnostics against a millisecond threshold, and produces a fix list sorted by effect on build wait time.

What do I need for the skill to work?

Python 3 for scripts/diagnose_compilation.py, plus an Xcode setup where you can build the given scheme and destination. The script writes results to the directory passed in --output-dir.

Source text

SKILL.md · MIT
---
name: xcode-compilation-analyzer
description: Analyze Swift and mixed-language compile hotspots using build timing summaries and Swift frontend diagnostics, then produce a recommend-first source-level optimization plan. Use when a developer reports slow compilation, type-checking warnings, expensive clean-build compile phases, long CompileSwiftSources tasks, warn-long-function-bodies output, or wants to speed up Swift type checking.
---

# Xcode Compilation Analyzer

Use this skill when compile time, not just general project configuration, looks like the bottleneck.

## Core Rules

- Start from evidence, ideally a recent `.build-benchmark/` artifact or raw timing-summary output.
- Prefer analysis-only compiler flags over persistent project edits during investigation.
- Rank findings by expected **wall-clock** impact, not cumulative compile-time impact. When compile tasks are heavily parallelized (sum of compile categories >> wall-clock median), note that fixing individual hotspots may improve parallel efficiency without reducing build wait time.
- When the evidence points to parallelized work rather than serial bottlenecks, label recommendations as "Reduces compiler workload (parallel)" rather than "Reduces build time."
- Do not edit source or build settings without explicit developer approval.

## What To Inspect

- `Build Timing Summary` output from clean and incremental builds
- long-running `CompileSwiftSources` or per-file compilation tasks
- `SwiftEmitModule` time -- can reach 60s+ after a single-line change in large modules; if it dominates incremental builds, the module is likely too large or macro-heavy
- `Planning Swift module` time -- if this category is disproportionately large in incremental builds (up to 30s per module), it signals unexpected input invalidation or macro-related rebuild cascading
- ad hoc runs with:
  - `-Xfrontend -warn-long-expression-type-checking=<ms>`
  - `-Xfrontend -warn-long-function-bodies=<ms>`
- deeper diagnostic flags for thorough investigation:
  - `-Xfrontend -debug-time-compilation` -- per-file compile times to rank the slowest files
  - `-Xfrontend -debug-time-function-bodies` -- per-function compile times (unfiltered, complements the threshold-based warning flags)
  - `-Xswiftc -driver-time-compilation` -- driver-level timing to isolate driver overhead
  - `-Xfrontend -stats-output-dir <path>` -- detailed compiler statistics (JSON) per compilation unit for root-cause analysis
- mixed Swift and Objective-C surfaces that increase bridging work

## Analysis Workflow

1. Identify whether the main issue is broad compilation volume or a few extreme hotspots.
2. Parse timing-summary categories and rank the biggest compile contributors.
3. Run the diagnostics script to surface type-checking hotspots:
   ```bash
   python3 scripts/diagnose_compilation.py \
     --project App.xcodeproj \
     --scheme MyApp \
     --configuration Debug \
     --destination "platform=iOS Simulator,name=iPhone 16" \
     --threshold 100 \
     --output-dir .build-benchmark
   ```
   This produces a ranked list of functions and expressions that exceed the millisecond threshold. Use the diagnostics artifact alongside source inspection to focus on the most expensive files first.
4. Map the evidence to a concrete recommendation list.
5. Separate code-level suggestions from project-level or module-level suggestions.

## Apple-Derived Checks

Look for these patterns first:

- missing explicit type information in expensive expressions
- complex chained or nested expressions that are hard to type-check
- delegate properties typed as `AnyObject` instead of a concrete protocol
- oversized Objective-C bridging headers or generated Swift-to-Objective-C surfaces
- header imports that skip framework qualification and miss module-cache reuse
- classes missing `final` that are never subclassed
- overly broad access control (`public`/`open`) on internal-only symbols
- monolithic SwiftUI `body` properties that should be decomposed into subviews
- long method chains or closures without intermediate type annotations

## Reporting Format

For each recommendation, include:

- observed evidence
- likely affected file or module
- expected wait-time impact (e.g. "Expected to reduce your clean build by ~2s" or "Reduces parallel compile work but unlikely to reduce build wait time")
- confidence
- whether approval is required before applying it

If the evidence points to project configuration instead of source, hand off to [`xcode-project-analyzer`](../xcode-project-analyzer/SKILL.md) by reading its SKILL.md and applying its workflow to the same project context.

## Preferred Tactics

- Suggest ad hoc flag injection through the build command before recommending persistent build-setting changes.
- Prefer narrowing giant view builders, closures, or result-builder expressions into smaller typed units.
- Recommend explicit imports and protocol typing when they reduce compiler search space.
- Call out when mixed-language boundaries are the real issue rather than Swift syntax alone.

## Additional Resources

- For the detailed audit checklist, see [references/code-compilation-checks.md](references/code-compilation-checks.md)
- For the shared recommendation structure, see [references/recommendation-format.md](references/recommendation-format.md)
- For source citations, see [references/build-optimization-sources.md](references/build-optimization-sources.md)

© AvdLee · MIT · AvdLee/Xcode-Build-Optimization-Agent-Skill

Similar skills

Everything for Mobile →

From the same set6

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor