Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when setting up a new project's production observability stack, reviewing incident response readiness, or investigating why errors are going undetected.
.claude/skills/thedaviddias-error-monitoring/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-21 | ✓→✗ | ▼ Worse | -9% | 0% |
| case-20 | ✓→✓ | ＝ Same ✓ | -8% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 9% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -3% | 0% |
Production errors are invisible without monitoring. Users rarely file detailed bug reports — they simply leave. Error monitoring gives you actionable stack traces, breadcrumbs, and user context within seconds of an issue occurring, cutting mean time to detect (MTTD) from days to minutes and dramatically reducing the cost of incidents.
Check whether an error monitoring service is configured and capturing unhandled errors, promise rejections, React/framework-level errors, failed network requests, and production Core Web Vitals signals.
Integrate an error monitoring SDK, configure environment and release tagging, capture failed network requests and RUM metrics, and set up alerts for critical error thresholds and regressions.
Explain how error monitoring services work, what signals they capture, and how error grouping reduces alert noise.
Review the monitoring initialisation code. Flag missing user context, absent release tagging, overly broad event filters, missing network instrumentation, and any patterns that would swallow errors before they reach the monitoring service.
For full implementation details, code examples, and framework-specific guidance, see references/rule.md.
Rule page: https://frontendchecklist.io/en/rules/testing/error-monitoring
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 12,298 | 10,151 | -17% | 1 | 1 | 0% | 2,578 | 2,377 | -8% | 0 | 0 | — |
case-21 | pass→fail | 11,108 | 7,643 | -31% | 1 | 1 | 0% | 2,181 | 1,975 | -9% | 0 | 0 | — |
case-01 | pass→pass | 9,285 | 8,280 | -11% | 1 | 1 | 0% | 1,901 | 2,066 | +9% | 0 | 0 | — |
case-02 | pass→pass | 5,369 | 3,422 | -36% | 1 | 1 | 0% | 1,090 | 1,053 | -3% | 0 | 0 | — |
case-03 | pass→pass | 8,361 | 8,313 | -1% | 1 | 1 | 0% | 1,879 | 1,903 | +1% | 0 | 0 | — |
case-04 | pass→pass | 12,055 | 6,267 | -48% | 1 | 1 | 0% | 2,140 | 1,615 | -25% | 0 | 0 | — |
case-05 | pass→pass | 5,377 | 3,358 | -38% | 1 | 1 | 0% | 1,230 | 1,039 | -16% | 0 | 0 | — |
case-06 | pass→pass | 4,067 | 2,525 | -38% | 1 | 1 | 0% | 800 | 797 | -0% | 0 | 0 | — |
case-07 | pass→pass | 10,012 | 8,933 | -11% | 1 | 1 | 0% | 1,818 | 2,222 | +22% | 0 | 0 | — |
case-08 | pass→pass | 5,145 | 4,927 | -4% | 1 | 1 | 0% | 1,050 | 1,332 | +27% | 0 | 0 | — |
case-09 | pass→pass | 10,286 | 8,690 | -16% | 1 | 1 | 0% | 1,884 | 1,964 | +4% | 0 | 0 | — |
case-10 | pass→pass | 8,998 | 6,355 | -29% | 1 | 1 | 0% | 1,827 | 1,519 | -17% | 0 | 0 | — |
case-11 | pass→pass | 6,893 | 4,575 | -34% | 1 | 1 | 0% | 1,409 | 1,340 | -5% | 0 | 0 | — |
case-12 | pass→pass | 6,490 | 4,827 | -26% | 1 | 1 | 0% | 1,250 | 1,340 | +7% | 0 | 0 | — |
case-13 | pass→pass | 6,618 | 4,214 | -36% | 1 | 1 | 0% | 1,203 | 1,141 | -5% | 0 | 0 | — |
case-14 | pass→pass | 7,429 | 4,409 | -41% | 1 | 1 | 0% | 1,326 | 1,197 | -10% | 0 | 0 | — |
case-15 | pass→pass | 9,847 | 6,953 | -29% | 1 | 1 | 0% | 1,960 | 1,568 | -20% | 0 | 0 | — |
case-16 | pass→pass | 13,072 | 11,005 | -16% | 1 | 1 | 0% | 2,333 | 2,602 | +12% | 0 | 0 | — |
case-17 | fail→pass | 5,164 | 3,896 | -25% | 1 | 1 | 0% | 981 | 1,010 | +3% | 0 | 0 | — |
case-18 | pass→pass | 3,944 | 4,977 | +26% | 1 | 1 | 0% | 757 | 1,374 | +82% | 0 | 0 | — |
case-19 | pass→pass | 7,823 | 7,191 | -8% | 1 | 1 | 0% | 1,419 | 1,622 | +14% | 0 | 0 | — |
case-22 | pass→pass | 13,156 | 13,416 | +2% | 1 | 1 | 0% | 2,941 | 3,362 | +14% | 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.