Diagnose software bugs through a structured six-phase hypothesis-driven workflow.
npx skills add https://github.com/sumpalabs/petrodb --skill diagnose-sumpalabs
Please help me install this Agent Skill. Skill: diagnose Source: https://github.com/sumpalabs/petrodb/tree/main/.agents/skills/diagnose Command: npx skills add https://github.com/sumpalabs/petrodb --skill diagnose-sumpalabs
Ad-hoc, unguided debugging of hard bugs and performance regressions wastes engineering time with random hypothesis testing and missed root causes, leading to recurring defects and slow incident resolution.
Use the diagnose skill to work through the reported intermittent timeout bug in the data ingestion pipeline by following the structured 6-phase diagnosis loop.
High-intent search queries and answers about installing and using this skill.
A systematic diagnosis workflow enforces a repeatable, hypothesis-driven debug process to find root causes. It replaces ad-hoc guessing with a structured loop involving deterministic reproduction, targeted instrumentation, and regression testing to resolve hard software bugs.
To debug intermittent crashes and non-deterministic failures, build a fast deterministic reproduction loop. Rank falsifiable hypotheses, apply targeted instrumentation to isolate the root cause, and verify the fix with mandatory regression tests to prevent recurrence.
The best way to troubleshoot a performance regression is following a disciplined 6-phase debug loop. It guides you through reproducing the slowdown, generating ranked hypotheses, and adding targeted instrumentation to systematically isolate the root cause.
Build a reproduction loop for hard-to-reproduce defects using prioritized strategies to create fast, deterministic, agent-runnable tests. This includes techniques for triggering elusive bugs and handling human-in-the-loop scenarios within data pipelines.
Yes, this structured debugging approach applies to web services, data pipelines, and client applications. It handles intermittent crashes, non-deterministic failures, and performance slowdowns across these environments by enforcing a repeatable diagnosis workflow.
A hypothesis-driven debug process eliminates inefficient, unproductive debugging by testing ranked falsifiable hypotheses. This prevents missed root causes and recurring defects, saving engineering time compared to randomly guessing at code changes.
cloudofgeorgeMonitors deployed apps for errors, performance regressions, and page failures via screenshots and baselines.
zhukunpenglinyutongApply a five-dimension post-mortem framework to document bug fix root causes.
ref-docsReproduce software bugs, analyze causes, and generate fix plans.
Sir-chawakornGuide systematic debugging to identify, isolate, and resolve software defects.
