Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use this skill when debugging runtime failures in Node.js or Bun: stack traces, async errors, startup crashes, port conflicts, and memory leaks.
.claude/skills/amariahak-node-js-debugger/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-20 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-22 | ✓→✗ | ▼ Worse | 42% | 0% |
| case-18 | ✓→✓ | = Same ✓ | 0% | 0% |
| case-06 | ✓→✓ | = Same ✓ | 33% | 0% |
| case-01 | ✓→✓ | = Same ✓ | 22% | 0% |
Use this skill when debugging runtime failures in Node.js or Bun: stack traces, async errors, startup crashes, port conflicts, and memory leaks.
Rule: the top frames are often internal wrappers. Find the first frame in your project path.
Checklist:
Common causes:
awaitPromise.all where one rejects and the rejection isn’t handled upstreamFix patterns:
await async calls in request handlersprocess.on("unhandledRejection") logging in services (carefully)Prefer explicit boundaries:
ENOENT: missing file → check path + cwd + packagingEADDRINUSE: port in use → pick a new port or kill existing processECONNREFUSED: service not reachable → check host/port, local service runningMODULE_NOT_FOUND: missing dependency/build output → reinstall or rebuildFrequent culprits:
Fix patterns:
--inspect)When reproduction is hard, use a debugger:
--inspect or --inspect-brkPrefer:
Example (structured):
tslogger.error("db.connect_failed", { host, port, cause: String(err) });
Before changing code:
EADDRINUSE)Workflow:
:3000, :5173, etc.)Avoid “random port hopping” without understanding why the port is busy.
grep to locate error strings and catch blocks.When reporting issues:
catch without rethrowing or returning failuretry/catch everywhere without understanding the root cause| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-18 | pass→pass | 10,902 | 5,750 | -47% | 1 | 1 | 0% | 1,809 | 1,805 | -0% | 0 | 0 | — |
case-06 | pass→pass | 8,966 | 6,576 | -27% | 1 | 1 | 0% | 1,551 | 2,070 | +33% | 0 | 0 | — |
case-01 | pass→pass | 7,136 | 3,876 | -46% | 1 | 1 | 0% | 1,160 | 1,412 | +22% | 0 | 0 | — |
case-12 | pass→pass | 11,975 | 7,828 | -35% | 1 | 1 | 0% | 2,174 | 2,273 | +5% | 0 | 0 | — |
case-02 | pass→pass | 5,238 | 5,397 | +3% | 1 | 1 | 0% | 930 | 1,833 | +97% | 0 | 0 | — |
case-03 | pass→pass | 14,813 | 11,696 | -21% | 1 | 1 | 0% | 2,566 | 3,111 | +21% | 0 | 0 | — |
case-04 | pass→pass | 10,618 | 8,909 | -16% | 1 | 1 | 0% | 1,877 | 2,216 | +18% | 0 | 0 | — |
case-05 | pass→pass | 12,226 | 7,575 | -38% | 1 | 1 | 0% | 2,404 | 2,326 | -3% | 0 | 0 | — |
case-07 | pass→pass | 4,034 | 3,657 | -9% | 1 | 1 | 0% | 682 | 1,478 | +117% | 0 | 0 | — |
case-08 | pass→pass | 7,546 | 8,719 | +16% | 1 | 1 | 0% | 1,400 | 2,121 | +52% | 0 | 0 | — |
case-09 | pass→pass | 11,790 | 7,646 | -35% | 1 | 1 | 0% | 2,053 | 2,111 | +3% | 0 | 0 | — |
case-10 | pass→pass | 10,267 | 8,803 | -14% | 1 | 1 | 0% | 1,913 | 2,334 | +22% | 0 | 0 | — |
case-11 | pass→pass | 3,333 | 3,167 | -5% | 1 | 1 | 0% | 625 | 1,380 | +121% | 0 | 0 | — |
case-13 | pass→pass | 7,804 | 4,263 | -45% | 1 | 1 | 0% | 1,341 | 1,493 | +11% | 0 | 0 | — |
case-14 | pass→pass | 12,427 | 10,336 | -17% | 1 | 1 | 0% | 2,113 | 2,302 | +9% | 0 | 0 | — |
case-15 | pass→pass | 13,282 | 16,780 | +26% | 1 | 1 | 0% | 2,349 | 2,925 | +25% | 0 | 0 | — |
case-16 | pass→pass | 8,375 | 3,678 | -56% | 1 | 1 | 0% | 1,413 | 1,380 | -2% | 0 | 0 | — |
case-17 | pass→pass | 11,773 | 8,833 | -25% | 1 | 1 | 0% | 1,798 | 2,196 | +22% | 0 | 0 | — |
case-19 | pass→pass | 5,545 | 4,476 | -19% | 1 | 1 | 0% | 991 | 1,652 | +67% | 0 | 0 | — |
case-20 | fail→pass | 14,610 | 8,720 | -40% | 1 | 1 | 0% | 2,665 | 2,342 | -12% | 0 | 0 | — |
case-21 | pass→pass | 9,000 | 7,093 | -21% | 1 | 1 | 0% | 1,591 | 2,125 | +34% | 0 | 0 | — |
case-22 | pass→fail | 11,854 | 10,925 | -8% | 1 | 1 | 0% | 2,126 | 3,009 | +42% | 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 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.