Guide engineers through structured four-phase root-cause debugging from symptoms to verification.
npx skills add https://github.com/tropeks/Vitali --skill investigate-tropeks
Please help me install this Agent Skill. Skill: investigate Source: https://github.com/tropeks/Vitali/tree/main/.claude/skills/gstack-investigate Command: npx skills add https://github.com/tropeks/Vitali --skill investigate-tropeks
Root-cause debugging is hard and time-consuming. This Skill guides engineers through a repeatable, phase-driven approach to uncover the true cause of failures rather than patching symptoms.
Ask me to start the root-cause investigation flow with investigate, and I will guide you through collecting symptoms, forming hypotheses, testing, and implementing fixes.
High-intent search queries and answers about installing and using this skill.
Root-cause debugging uses a four-phase process to uncover true failure causes: collecting symptoms, reading code, generating and testing hypotheses, and verifying fixes. This prevents patching symptoms and ensures a repeatable investigation flow.
You trace bugs by collecting symptoms, reading code paths, and using guided experimentation with lightweight instrumentation. This approach gathers evidence from logs and traces to confirm root causes before proposing a minimal, verifiable fix.
The best way to verify a bug fix is through the final phase of root-cause debugging, which requires running tests and regression checks. This confirms the proposed minimal fix resolves the failure without introducing new issues.
Yes, root-cause debugging requires a stable environment to collect logs and reproduce errors. A stable setup ensures accurate evidence gathering and reliable hypothesis testing during the investigation phases.
You should use a phase-based investigation flow when a bug surfaces in a service or deployment and ad-hoc debugging is time-consuming. It provides a repeatable, structured process to move from symptoms to a verified fix.
Pryma-TechDiagnose OMC session or repository failures from logs and traces.
robkamEnforce a red-green-proof loop starting with a failing regression test.
huyhung9630Generate hypotheses and gather evidence to trace software issues to their root cause.
OneWave-AIReview code for security vulnerabilities, performance bottlenecks, and maintainability issues.
