Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Trace bugs backward to their origin, not just symptoms
.claude/skills/dicklesworthstone-root-cause-tracing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 25% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 10% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 3% | 0% |
| case-18 | ✓→✓ | = Same ✓ | -5% | 0% |
TRACE BACKWARD TO THE BUG'S ORIGIN
Don't patch symptoms - fix the diseaseSYMPTOM
↓ Where does this value come from?
IMMEDIATE CAUSE
↓ What set that to wrong value?
DEEPER CAUSE
↓ Why did that happen?
ROOT CAUSE ← Fix HEREQ: What exactly went wrong?
→ Document the exact symptom
→ Note file, line, error messageQ: Where does this value/state come from?
→ Find the direct source
→ Check: is THIS the bug or just a carrier?Q: What set THAT to the wrong value?
→ Follow the data flow backward
→ Each step: is this the origin or just passing it along?SIGNS YOU'VE FOUND ROOT:
- Logic error in requirements interpretation
- Initial state incorrectly set
- Wrong assumption in algorithm
- Missing validation at entry point
- Broken invariant at creationYou've found root cause when:
You're still at symptoms if:
| Bug Pattern | Root Usually At | |-------------|-----------------| | Wrong value displayed | Data calculation, not display code | | Null pointer | Where object should've been created | | Off-by-one | Loop bounds definition, not body | | Race condition | Shared state design, not specific access | | Invalid state | State transition logic, not current handler |
❌ Symptom Patching
// Bad: Fix where error happens
if (value === undefined) value = defaultValue
// Good: Find why value is undefined❌ Guard Stacking
// Bad: Add guards everywhere
if (x) if (y) if (z) doThing()
// Good: Ensure x, y, z correct at source❌ Quick Fix Pressure
"Just make it work" → Technical debt
Better: "5 more minutes tracing saves 5 hours debugging later"After identifying root cause:
Use with:
assumption-checker → Was root cause an assumption?dependency-validator → Was root cause a broken dependency?pre-action-verifier → Verify root cause fix before applying| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | pass→pass | 18,256 | 17,044 | -7% | 1 | 1 | 0% | 2,933 | 3,215 | +10% | 0 | 0 | — |
case-05 | pass→pass | 13,540 | 11,154 | -18% | 1 | 1 | 0% | 2,284 | 2,357 | +3% | 0 | 0 | — |
case-17 | fail→fail | 12,872 | 12,200 | -5% | 1 | 1 | 0% | 2,153 | 2,860 | +33% | 0 | 0 | — |
case-18 | pass→pass | 15,061 | 10,289 | -32% | 1 | 1 | 0% | 2,146 | 2,044 | -5% | 0 | 0 | — |
case-01 | pass→pass | 14,989 | 43,299 | +189% | 1 | 1 | 0% | 2,291 | 2,610 | +14% | 0 | 0 | — |
case-02 | fail→pass | 16,541 | 11,157 | -33% | 1 | 1 | 0% | 2,064 | 2,574 | +25% | 0 | 0 | — |
case-03 | pass→pass | 15,330 | 11,715 | -24% | 1 | 1 | 0% | 2,245 | 2,378 | +6% | 0 | 0 | — |
case-06 | pass→pass | 14,065 | 9,497 | -32% | 1 | 1 | 0% | 2,173 | 2,386 | +10% | 0 | 0 | — |
case-07 | pass→pass | 15,161 | 14,767 | -3% | 1 | 1 | 0% | 2,266 | 3,163 | +40% | 0 | 0 | — |
case-08 | pass→pass | 18,470 | 14,288 | -23% | 1 | 1 | 0% | 2,639 | 2,784 | +5% | 0 | 0 | — |
case-09 | pass→pass | 18,209 | 15,347 | -16% | 1 | 1 | 0% | 2,500 | 2,873 | +15% | 0 | 0 | — |
case-10 | pass→pass | 14,104 | 9,432 | -33% | 1 | 1 | 0% | 2,126 | 2,033 | -4% | 0 | 0 | — |
case-11 | pass→pass | 12,235 | 10,628 | -13% | 1 | 1 | 0% | 1,991 | 2,375 | +19% | 0 | 0 | — |
case-12 | pass→pass | 14,683 | 12,731 | -13% | 1 | 1 | 0% | 2,215 | 2,586 | +17% | 0 | 0 | — |
case-13 | pass→pass | 5,590 | 4,587 | -18% | 1 | 1 | 0% | 676 | 1,421 | +110% | 0 | 0 | — |
case-14 | pass→pass | 14,375 | 10,747 | -25% | 1 | 1 | 0% | 2,238 | 2,366 | +6% | 0 | 0 | — |
case-15 | pass→pass | 14,848 | 7,054 | -52% | 1 | 1 | 0% | 1,960 | 1,801 | -8% | 0 | 0 | — |
case-16 | fail→pass | 7,563 | 3,532 | -53% | 1 | 1 | 0% | 1,375 | 1,317 | -4% | 0 | 0 | — |
case-19 | fail→fail | 15,878 | 9,705 | -39% | 1 | 1 | 0% | 2,216 | 2,215 | -0% | 0 | 0 | — |
case-20 | pass→pass | 8,863 | 7,654 | -14% | 1 | 1 | 0% | 1,230 | 2,022 | +64% | 0 | 0 | — |
case-21 | pass→pass | 15,111 | 12,768 | -16% | 1 | 1 | 0% | 2,407 | 3,085 | +28% | 0 | 0 | — |
case-22 | pass→pass | 11,061 | 13,837 | +25% | 1 | 1 | 0% | 2,166 | 3,243 | +50% | 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. The headline lift of +9 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.