Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when debugging any bug. Enforces a 5-phase protocol where code edits are blocked until root cause is confirmed. Prevents premature fix attempts.
.claude/skills/sickn33-phase-gated-debugging/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -13% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -20% | 0% |
AI coding agents see an error and immediately edit code. They guess at fixes, get it wrong, and spiral. This skill enforces a strict 5-phase protocol where you CANNOT edit source code until the root cause is identified and confirmed.
Based on claude-debug (full plugin with PreToolUse hook enforcement).
Use this skill when:
Run the failing command/test. Capture the exact error. Run 2-3 times for consistency.
Read code. Add diagnostic logging marked // DEBUG. Re-run with diagnostics. Binary search to narrow down.
// DEBUG marked logging is allowedAnalyze WHY at the isolated location. Use "5 Whys" technique. Remove debug logging.
State: "This is my root cause analysis: explanation]. Do you agree, or should I investigate further?"
WAIT for user confirmation. Do NOT proceed without it.
Remove all // DEBUG lines. Apply minimal change addressing confirmed root cause.
Run original failing test — must pass. Run related tests. For intermittent bugs, run 5+ times. If verification fails: root cause was wrong, go back to Phase 2.
| Type | Technique | |------|-----------| | Crash/Panic | Stack trace backward — trace the bad value to its source | | Wrong Output | Binary search — log midpoint, halve search space each iteration | | Intermittent | Compare passing vs failing run logs — find ordering divergence | | Regression | git bisect — find the offending commit | | Performance | Timing at stage boundaries — find the bottleneck |
// DEBUG in phase 2)User request:
> a bug keeps getting "fixed" without resolving the underlying issue; diagnose it and return the concrete next action.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-06 | fail→pass | 12,267 | 9,003 | -27% | 1 | 1 | 0% | 1,816 | 1,378 | -24% | 0 | 0 | — |
case-01 | fail→fail | 1,868 | 2,695 | +44% | 1 | 1 | 0% | 293 | 1,081 | +269% | 0 | 0 | — |
case-02 | fail→fail | 2,165 | 3,069 | +42% | 1 | 1 | 0% | 254 | 1,087 | +328% | 0 | 0 | — |
case-03 | pass→pass | 16,470 | 10,869 | -34% | 1 | 1 | 0% | 3,509 | 3,003 | -14% | 0 | 0 | — |
case-04 | pass→pass | 12,213 | 10,052 | -18% | 1 | 1 | 0% | 2,211 | 2,303 | +4% | 0 | 0 | — |
case-05 | pass→pass | 18,024 | 12,903 | -28% | 1 | 1 | 0% | 3,935 | 3,295 | -16% | 0 | 0 | — |
case-07 | fail→pass | 10,615 | 4,739 | -55% | 1 | 1 | 0% | 1,751 | 1,499 | -14% | 0 | 0 | — |
case-08 | pass→pass | 9,659 | 3,073 | -68% | 1 | 1 | 0% | 1,555 | 1,197 | -23% | 0 | 0 | — |
case-09 | pass→pass | 10,749 | 6,056 | -44% | 1 | 1 | 0% | 1,768 | 1,682 | -5% | 0 | 0 | — |
case-10 | pass→pass | 8,006 | 3,287 | -59% | 1 | 1 | 0% | 1,483 | 1,267 | -15% | 0 | 0 | — |
case-11 | fail→pass | 14,183 | 9,119 | -36% | 1 | 1 | 0% | 2,632 | 2,283 | -13% | 0 | 0 | — |
case-12 | pass→pass | 15,048 | 9,727 | -35% | 1 | 1 | 0% | 2,465 | 2,251 | -9% | 0 | 0 | — |
case-13 | pass→pass | 15,591 | 6,729 | -57% | 1 | 1 | 0% | 2,532 | 1,883 | -26% | 0 | 0 | — |
case-14 | fail→pass | 15,528 | 7,722 | -50% | 1 | 1 | 0% | 2,445 | 1,992 | -19% | 0 | 0 | — |
case-15 | fail→pass | 11,747 | 5,549 | -53% | 1 | 1 | 0% | 1,941 | 1,558 | -20% | 0 | 0 | — |
case-16 | fail→pass | 10,510 | 4,701 | -55% | 1 | 1 | 0% | 1,716 | 1,518 | -12% | 0 | 0 | — |
case-17 | pass→pass | 6,328 | 2,270 | -64% | 1 | 1 | 0% | 1,036 | 1,012 | -2% | 0 | 0 | — |
case-18 | pass→pass | 5,098 | 2,513 | -51% | 1 | 1 | 0% | 837 | 1,038 | +24% | 0 | 0 | — |
case-19 | pass→pass | 8,432 | 3,186 | -62% | 1 | 1 | 0% | 1,358 | 1,210 | -11% | 0 | 0 | — |
case-20 | fail→pass | 8,874 | 4,674 | -47% | 1 | 1 | 0% | 1,303 | 1,489 | +14% | 0 | 0 | — |
case-21 | fail→pass | 10,365 | 8,189 | -21% | 1 | 1 | 0% | 1,595 | 1,976 | +24% | 0 | 0 | — |
case-22 | fail→pass | 9,263 | 7,966 | -14% | 1 | 1 | 0% | 1,510 | 1,926 | +28% | 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 +41 percentage points is the difference between those two pass rates over the 22 comparable cases.
The publisher has shipped newer versions since this run, so these numbers describe v1, not the version currently listed.
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.