Self-hosted alternatives to python-performance-optimization

P

python-performance-optimization

Profile and optimize Python code

python-performance-optimization is an agent skill from the wshobson/agents collection for profiling and optimizing Python code. Its rule is to measure before changing anything, so effort goes to the real bottleneck rather than to guesses. The skill covers the profiling types and the tools for each: cProfile for CPU profiling of whole programs, line_profiler for line-by-line timing inside a function, memory_profiler for tracking memory use, and py-spy for profiling production applications without changing their code. It explains how to find and read the hot spots in a profile and how to reduce latency and response times, optimize CPU-intensive operations, cut memory consumption and find leaks, improve database query performance and I/O, speed up data processing pipelines, and choose high-performance algorithms. The skill applies when debugging slow Python code, identifying bottlenecks, reducing latency, optimizing CPU-heavy work, reducing memory use, and profiling a live application. Optimization changes can make code harder to read, so the skill's guidance favors changes that the measurements justify and that keep behavior the same. The agent uses the profiles to pick targets and checks that the change speeds up the measured path. The skill is installed with the skills CLI from the wshobson/agents repository. Its skills.sh page showed about 34.9K installs and passing results from three security audits at the time of the scan.