Analyze systing trace databases to diagnose system performance issues.
npx skills add https://github.com/josefbacik/systing --skill systing-analyze
Please help me install this Agent Skill. Skill: systing-analyze Source: https://github.com/josefbacik/systing/tree/main/skills/systing-analyze Command: npx skills add https://github.com/josefbacik/systing --skill systing-analyze
This Skill helps you analyze complex system performance traces stored in DuckDB databases, enabling you to diagnose performance bottlenecks, understand resource utilization, and identify root causes of system issues.
trace.duckdb file to pinpoint whether the issue is CPU-bound, I/O-bound, or related to network congestion.Use the systing-analyze skill to get information about the trace file located at /path/to/your/trace.duckdb.
High-intent search queries and answers about installing and using this skill.
To analyze a DuckDB trace database, you can use this Skill to orchestrate trace information retrieval, schema discovery, and generate flamegraphs to diagnose CPU hotspots, scheduling latency, and network behavior directly from the file.
Finding the root cause of CPU bottlenecks in a Linux trace file involves analyzing CPU utilization metrics, off-CPU time, and generating flamegraphs from the trace database to pinpoint exactly where the system resources are being consumed.
Yes, you can run ad-hoc SQL queries on the trace database to investigate TPU performance, allowing you to dive deep into TPU utilization, operational metrics, and identify specific performance gaps.
You can check scheduling latency and off-CPU time from a trace.duckdb file by utilizing the built-in scheduling statistics and CPU utilization analysis tools to understand thread blocking and context switching impacts.
Yes, this trace analysis approach works for identifying network congestion bottlenecks by analyzing the network behavior data captured within the DuckDB trace database to isolate I/O-bound system slowdowns.
To start diagnosing performance slowdowns using a trace database, you need a valid .duckdb trace file path; the Skill then automatically handles schema discovery, metric extraction, and visualization orchestration.
dawsonblockCapture Unreal Engine performance traces into .utrace files and export timing and counter data to CSV.
tamago0224Diagnose Go performance bottlenecks from pprof and fgprof profiles.
PovertyActionRun SQL queries and transform CSV, Parquet, JSON, and Excel files via the DuckDB CLI.
OpenCloudOSProfile Linux runtime events in memory for real-time system performance analysis.