Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when you need to verify Java performance optimizations by comparing profiling results before and after refactoring — including baseline validation, post-refactoring report generation, quantitative before/after metrics comparison, side-by-side flamegraph analysis, regression detection, or creating profiling-comparison-analysis and profiling-final-results documentation. This should trigger for requests such as Verify performance fix; Verify the performance; Verify the memory; Verify the thread
.claude/skills/jabrena-164-java-profiling-verify/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -51% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 24% | 0% |
Verify performance optimizations through rigorous before/after comparison: ensure baseline and post-refactoring profiling data use identical test conditions, generate post-refactoring reports, compare metrics (memory, CPU, GC, threading), perform side-by-side flamegraph analysis, document findings in profiling-comparison-analysis-YYYYMMDD.md and profiling-final-results-YYYYMMDD.md, and validate success criteria.
What is covered in this Skill?
Scope: Identical test conditions are critical. Document test scenarios. Validate application runs with refactored code before generating new reports.
Use identical test conditions between baseline and post-refactoring. Verify both report sets are complete. Document test scenarios.
Read references/164-java-profiling-verify.md and verify baseline artifacts exist and are non-empty.
Run profiling with identical load/test conditions to produce comparable post-refactoring artifacts.
Perform quantitative comparisons for memory/CPU/GC/threading and side-by-side flamegraph analysis.
Create comparison and final results reports with regressions, gains, and reproducible scenario details.
For detailed guidance, examples, and constraints, see references/164-java-profiling-verify.md.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-09 | pass→pass | 13,519 | 4,842 | -64% | 1 | 1 | 0% | 2,095 | 1,332 | -36% | 0 | 0 | — |
case-01 | fail→fail | 10,446 | 3,796 | -64% | 1 | 1 | 0% | 1,535 | 1,218 | -21% | 0 | 0 | — |
case-02 | fail→fail | 26,955 | 3,504 | -87% | 1 | 1 | 0% | 4,516 | 1,066 | -76% | 0 | 0 | — |
case-03 | fail→fail | 18,988 | 5,479 | -71% | 1 | 1 | 0% | 3,041 | 937 | -69% | 0 | 0 | — |
case-04 | fail→pass | 16,918 | 6,058 | -64% | 1 | 1 | 0% | 2,376 | 1,505 | -37% | 0 | 0 | — |
case-05 | pass→pass | 10,843 | 5,579 | -49% | 1 | 1 | 0% | 1,648 | 1,506 | -9% | 0 | 0 | — |
case-06 | fail→pass | 14,671 | 10,572 | -28% | 1 | 1 | 0% | 2,056 | 2,088 | +2% | 0 | 0 | — |
case-07 | pass→pass | 12,417 | 5,255 | -58% | 1 | 1 | 0% | 1,730 | 1,450 | -16% | 0 | 0 | — |
case-08 | pass→pass | 11,093 | 4,608 | -58% | 1 | 1 | 0% | 1,689 | 1,417 | -16% | 0 | 0 | — |
case-10 | pass→pass | 19,794 | 9,047 | -54% | 1 | 1 | 0% | 2,985 | 1,921 | -36% | 0 | 0 | — |
case-11 | pass→pass | 13,625 | 6,924 | -49% | 1 | 1 | 0% | 1,996 | 1,579 | -21% | 0 | 0 | — |
case-12 | fail→pass | 15,549 | 9,049 | -42% | 1 | 1 | 0% | 2,303 | 2,175 | -6% | 0 | 0 | — |
case-13 | pass→pass | 10,536 | 6,987 | -34% | 1 | 1 | 0% | 1,510 | 1,609 | +7% | 0 | 0 | — |
case-14 | pass→pass | 11,981 | 6,416 | -46% | 1 | 1 | 0% | 1,914 | 1,726 | -10% | 0 | 0 | — |
case-15 | pass→pass | 19,157 | 8,490 | -56% | 1 | 1 | 0% | 2,673 | 1,939 | -27% | 0 | 0 | — |
case-16 | fail→pass | 15,240 | 2,347 | -85% | 1 | 1 | 0% | 2,044 | 1,005 | -51% | 0 | 0 | — |
case-17 | fail→pass | 6,500 | 3,255 | -50% | 1 | 1 | 0% | 957 | 1,184 | +24% | 0 | 0 | — |
case-18 | pass→pass | 12,735 | 6,825 | -46% | 1 | 1 | 0% | 1,859 | 1,617 | -13% | 0 | 0 | — |
case-19 | pass→pass | 2,958 | 2,859 | -3% | 1 | 1 | 0% | 498 | 1,181 | +137% | 0 | 0 | — |
case-20 | pass→pass | 12,602 | 9,362 | -26% | 1 | 1 | 0% | 2,040 | 2,221 | +9% | 0 | 0 | — |
case-21 | pass→pass | 16,334 | 11,919 | -27% | 1 | 1 | 0% | 2,644 | 2,621 | -1% | 0 | 0 | — |
case-22 | pass→pass | 22,294 | 15,683 | -30% | 1 | 1 | 0% | 3,395 | 3,209 | -5% | 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 +23 percentage points is the difference between those two pass rates over the 21 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.