Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Systematic root cause analysis for errors, bugs, and unexpected behaviors using 5-Why methodology. Use when user reports errors, build failures, test failures, performance issues, integration problems, or any "it's not working" scenarios.
.claude/skills/umputun-root-cause-investigator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 145% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 12% | 0% |
Apply systematic 5-Why methodology to identify fundamental root causes of issues rather than treating symptoms. Guide thorough investigation through structured evidence gathering and analysis.
Systematically ask "why" five times to drill down to root cause:
Collect information about the issue:
## Issue Summary
[brief description of reported problem]
## Initial Symptoms
- what user is experiencing
- error messages or logs
- observable behavior
## Context Gathering
- environment details
- recent changes
- related components
- steps to reproduceStructure the investigation with progressive depth:
## 5-Why Analysis
### Why #1: [surface cause]
Evidence: [logs, errors, behavior]
Impact: [what this affects]
### Why #2: [deeper cause]
Evidence: [code, configuration]
Impact: [cascading effects]
### Why #3: [system cause]
Evidence: [architecture, dependencies]
Impact: [broader implications]
### Why #4: [process/design cause]
Evidence: [patterns, decisions]
Impact: [long-term effects]
### Why #5: [root cause]
Evidence: [fundamental issue]
Impact: [core problem]Document the fundamental issue requiring attention:
## Root Cause Identified
[the fundamental issue that needs addressing]
## Recommended Investigation Areas
- specific files to examine
- components to test
- systems to verifyLoad reference files as needed during investigation:
The goal is to find the fundamental cause, not just fix symptoms.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 20,463 | 20,577 | +1% | 1 | 1 | 0% | 3,064 | 3,931 | +28% | 0 | 0 | — |
case-02 | fail→fail | 17,385 | 18,391 | +6% | 1 | 1 | 0% | 2,676 | 3,335 | +25% | 0 | 0 | — |
case-03 | pass→fail | 9,035 | 14,096 | +56% | 1 | 1 | 0% | 1,584 | 2,809 | +77% | 0 | 0 | — |
case-04 | pass→fail | 31,170 | 26,812 | -14% | 1 | 1 | 0% | 5,093 | 5,046 | -1% | 0 | 0 | — |
case-05 | pass→fail | 21,906 | 19,901 | -9% | 1 | 1 | 0% | 4,635 | 5,015 | +8% | 0 | 0 | — |
case-06 | fail→fail | 36,010 | 16,087 | -55% | 1 | 1 | 0% | 3,151 | 3,101 | -2% | 0 | 0 | — |
case-07 | fail→pass | 65,205 | 17,015 | -74% | 1 | 1 | 0% | 1,306 | 3,198 | +145% | 0 | 0 | — |
case-08 | fail→pass | 18,449 | 16,973 | -8% | 1 | 1 | 0% | 2,833 | 3,263 | +15% | 0 | 0 | — |
case-09 | fail→fail | 20,053 | 11,837 | -41% | 1 | 1 | 0% | 3,264 | 2,536 | -22% | 0 | 0 | — |
case-10 | fail→fail | 19,770 | 23,758 | +20% | 1 | 1 | 0% | 3,232 | 4,127 | +28% | 0 | 0 | — |
case-11 | fail→pass | 22,898 | 20,435 | -11% | 1 | 1 | 0% | 3,606 | 3,693 | +2% | 0 | 0 | — |
case-12 | fail→pass | 15,087 | 15,681 | +4% | 1 | 1 | 0% | 2,443 | 3,082 | +26% | 0 | 0 | — |
case-13 | fail→pass | 18,533 | 20,369 | +10% | 1 | 1 | 0% | 3,142 | 3,517 | +12% | 0 | 0 | — |
case-14 | fail→pass | 19,976 | 16,918 | -15% | 1 | 1 | 0% | 3,198 | 3,189 | -0% | 0 | 0 | — |
case-15 | fail→fail | 20,126 | 17,279 | -14% | 1 | 1 | 0% | 3,177 | 3,384 | +7% | 0 | 0 | — |
case-16 | fail→fail | 14,879 | 16,183 | +9% | 1 | 1 | 0% | 2,279 | 3,024 | +33% | 0 | 0 | — |
case-17 | fail→fail | 17,534 | 14,205 | -19% | 1 | 1 | 0% | 2,665 | 2,913 | +9% | 0 | 0 | — |
case-18 | fail→pass | 12,953 | 13,188 | +2% | 1 | 1 | 0% | 2,211 | 2,762 | +25% | 0 | 0 | — |
case-19 | fail→pass | 18,258 | 15,428 | -16% | 1 | 1 | 0% | 2,892 | 2,960 | +2% | 0 | 0 | — |
case-20 | fail→fail | 16,942 | 26,286 | +55% | 1 | 1 | 0% | 2,601 | 3,203 | +23% | 0 | 0 | — |
case-21 | fail→fail | 16,726 | 19,576 | +17% | 1 | 1 | 0% | 2,484 | 3,468 | +40% | 0 | 0 | — |
case-22 | fail→pass | 16,121 | 12,272 | -24% | 1 | 1 | 0% | 2,639 | 2,622 | -1% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +27 percentage points is the difference between those two pass rates over the 21 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.