Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when you need to analyze Java profiling data collected during the detection phase — including interpreting flamegraphs, memory allocation patterns, CPU hotspots, threading issues, systematic problem categorization, evidence documentation with profiling-problem-analysis and profiling-solutions markdown files, or prioritizing fixes using Impact/Effort scoring. This should trigger for requests such as Analyze JFR profile; Analyze the profile; Analyze the performance; Analyze the memory; Analyze
.claude/skills/jabrena-162-java-profiling-analyze/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -7% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -38% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 15% | 0% |
| case-07 | ✓→✓ | ＝ Same ✓ | 10% | 0% |
Analyze profiling results systematically: inventory results (flamegraphs, JFR, GC logs, thread dumps), identify problems (memory leaks, CPU hotspots, threading issues), document findings using standardized templates (profiling-problem-analysis-YYYYMMDD.md, profiling-solutions-YYYYMMDD.md), prioritize using Impact/Effort scores, and correlate multiple profiling files for validation.
What is covered in this Skill?
Scope: Validate profiling results represent realistic load scenarios. Cross-reference multiple files. Include quantitative metrics.
Validate profiling results represent realistic load before analysis. Document assumptions and limitations. Cross-reference multiple files.
Read references/162-java-profiling-analyze.md and inventory profiling artifacts in profiler/results/.
Confirm datasets represent realistic load conditions and record assumptions/limitations before drawing conclusions.
Analyze memory/CPU/threading findings, cross-reference multiple files, and prioritize issues by Impact/Effort.
Create docs/profiling-problem-analysis-YYYYMMDD.md and docs/profiling-solutions-YYYYMMDD.md with quantitative evidence.
For detailed guidance, examples, and constraints, see references/162-java-profiling-analyze.md.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 31,567 | 5,484 | -83% | 1 | 1 | 0% | 4,278 | 966 | -77% | 0 | 0 | — |
case-02 | fail→fail | 5,568 | 6,606 | +19% | 1 | 1 | 0% | 197 | 1,088 | +452% | 0 | 0 | — |
case-03 | fail→fail | 29,168 | 6,748 | -77% | 1 | 1 | 0% | 4,547 | 1,056 | -77% | 0 | 0 | — |
case-04 | fail→pass | 18,163 | 13,867 | -24% | 1 | 1 | 0% | 2,530 | 2,834 | +12% | 0 | 0 | — |
case-05 | pass→pass | 10,212 | 6,057 | -41% | 1 | 1 | 0% | 1,342 | 1,537 | +15% | 0 | 0 | — |
case-06 | fail→pass | 8,441 | 3,085 | -63% | 1 | 1 | 0% | 1,238 | 1,150 | -7% | 0 | 0 | — |
case-07 | pass→pass | 10,697 | 6,314 | -41% | 1 | 1 | 0% | 1,510 | 1,655 | +10% | 0 | 0 | — |
case-08 | pass→pass | 16,241 | 9,444 | -42% | 1 | 1 | 0% | 2,294 | 2,111 | -8% | 0 | 0 | — |
case-09 | pass→pass | 16,396 | 10,316 | -37% | 1 | 1 | 0% | 2,393 | 2,202 | -8% | 0 | 0 | — |
case-10 | fail→pass | 11,519 | 2,645 | -77% | 1 | 1 | 0% | 1,755 | 1,087 | -38% | 0 | 0 | — |
case-11 | pass→pass | 5,409 | 4,332 | -20% | 1 | 1 | 0% | 778 | 1,296 | +67% | 0 | 0 | — |
case-12 | fail→fail | 9,307 | 1,871 | -80% | 1 | 1 | 0% | 1,350 | 930 | -31% | 0 | 0 | — |
case-13 | pass→pass | 10,724 | 6,820 | -36% | 1 | 1 | 0% | 1,657 | 1,803 | +9% | 0 | 0 | — |
case-14 | pass→pass | 15,444 | 11,714 | -24% | 1 | 1 | 0% | 2,328 | 2,470 | +6% | 0 | 0 | — |
case-15 | pass→pass | 10,510 | 5,426 | -48% | 1 | 1 | 0% | 1,374 | 1,537 | +12% | 0 | 0 | — |
case-16 | pass→pass | 8,024 | 6,527 | -19% | 1 | 1 | 0% | 1,093 | 1,562 | +43% | 0 | 0 | — |
case-17 | pass→pass | 11,302 | 5,038 | -55% | 1 | 1 | 0% | 1,732 | 1,357 | -22% | 0 | 0 | — |
case-18 | pass→pass | 8,490 | 5,436 | -36% | 1 | 1 | 0% | 1,220 | 1,456 | +19% | 0 | 0 | — |
case-19 | fail→fail | 14,175 | 2,097 | -85% | 1 | 1 | 0% | 1,950 | 975 | -50% | 0 | 0 | — |
case-20 | pass→pass | 5,507 | 6,997 | +27% | 1 | 1 | 0% | 881 | 1,871 | +112% | 0 | 0 | — |
case-21 | pass→pass | 9,801 | 12,541 | +28% | 1 | 1 | 0% | 1,606 | 2,758 | +72% | 0 | 0 | — |
case-22 | fail→fail | 7,806 | 6,670 | -15% | 1 | 1 | 0% | 1,531 | 1,818 | +19% | 0 | 0 | — |
case-23 | pass→pass | 14,086 | 6,555 | -53% | 1 | 1 | 0% | 2,336 | 1,672 | -28% | 0 | 0 | — |
case-24 | pass→pass | 9,717 | 3,499 | -64% | 1 | 1 | 0% | 1,381 | 1,163 | -16% | 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. 24 cases were attempted, and 21 counted toward the lift figure. The other 3 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 +13 percentage points is the difference between those two pass rates over the 21 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.