performance-profiler-tools

Python tools for analyzing Instruments traces and performance data

$ Instalar

git clone https://github.com/paleoterra/PaleoRose /tmp/PaleoRose && cp -r /tmp/PaleoRose/.claude/skills/performance-profiler-tools ~/.claude/skills/PaleoRose

// tip: Run this command in your terminal to install the skill


name: performance-profiler-tools description: Python tools for analyzing Instruments traces and performance data type: skill language: python

Performance Profiler Tools

Analyze Instruments trace files and performance data.

Capabilities

  • Parse Instruments .trace files
  • Extract Time Profiler data
  • Analyze Allocations data
  • Generate performance reports
  • Compare trace files
  • Identify performance regressions
  • Calculate time percentages
  • Find memory growth patterns

Tools

instruments_analyzer.py - Parse Instruments traces

Commands

# Analyze trace
./instruments_analyzer.py analyze app.trace

# Compare traces
./instruments_analyzer.py compare --baseline baseline.trace --current current.trace

# Find hot paths
./instruments_analyzer.py hotpaths app.trace --threshold 5%

# Memory analysis
./instruments_analyzer.py memory app.trace

Output

Performance Analysis
===================
Total Time: 2.45s
Top Methods:

1. drawRect: (456ms, 18.6%)
2. calculateStatistics (234ms, 9.5%)
3. readFromStore (198ms, 8.1%)

Recommendations:
- Cache drawRect results
- Optimize calculateStatistics algorithm
- Add index to database query

Complements

performance-profiler agent (launches Instruments)