Profile code execution and benchmark shell commands for performance diagnostics.
npx skills add https://github.com/alangeb/tau --skill performance-alangeb
Please help me install this Agent Skill. Skill: performance Source: https://github.com/alangeb/tau/tree/main/src/skills/performance Command: npx skills add https://github.com/alangeb/tau --skill performance-alangeb
This Skill addresses performance degradation by identifying bottlenecks, measuring execution latency, and providing actionable insights into token usage and tool call efficiency.
Use the performance skill to audit the current session log and identify the top ten slowest tool calls.
High-intent search queries and answers about installing and using this skill.
Profiling code execution uses standard library access to cProfile and pstats to pinpoint slow operations. It analyzes function execution times to provide deterministic performance diagnostics and actionable insights into resource usage.
Benchmarking shell commands requires the subprocess module to measure execution latency. It analyzes shell command performance to pinpoint slow operations and identify bottlenecks within script execution workflows.
Auditing tool call latency parses session logs to calculate tool call durations. It identifies high-latency components and highlights the top ten slowest tool calls to determine if delays are caused by excessive context size.
Optimizing AI agent context management and token counting is supported through execution profiling and latency auditing. It identifies inefficiencies in token usage and tool call efficiency to resolve sluggish agent responses.
Running a performance audit requires standard library access to cProfile, pstats, and subprocess modules. There are no external dependencies needed to perform deterministic performance diagnostics and execution analysis.
mokbhaiProfile Python CPU, memory, and I/O usage with cProfile and py-spy.
CallMeLuigiv2Profile Python CPU and memory usage with cProfile, memory_profiler, and line_profiler.
alangebParse AI agent audit logs to identify bottlenecks, errors, and anomalies.
infotechsrealmList enabled Unity profiler modules with their status flags.
