Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when you need to generate a checklist document with embedded agents inventory, following the embedded template exactly and producing INVENTORY-AGENTS-JAVA.md in the project root. This should trigger for requests such as Create embedded agents inventory checklist; Generate INVENTORY-AGENTS-JAVA.md; Use @002-agents-inventory; Inventory embedded Java agent definitions; List generated agent roles for Java development. Part of Plinth Toolkit
.claude/skills/jabrena-002-agents-inventory/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -55% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -71% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -62% | 0% |
Create a comprehensive checklist document for embedded agents inventory by following the embedded template exactly.
What is covered in this Skill?
INVENTORY-AGENTS-JAVA.mdFollow the template exactly without adding or removing sections, rows, or details.
Read references/002-agents-inventory.md before generating output and use it as the authoritative template.
Step constraints:
Create INVENTORY-AGENTS-JAVA.md in the project root using the exact wording and ordering from the reference template.
Confirm no extra agent rows were introduced and all required reference sections are present in the generated file.
For detailed guidance, examples, and constraints, see references/002-agents-inventory.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 | 17,100 | 4,561 | -73% | 1 | 1 | 0% | 2,316 | 664 | -71% | 0 | 0 | — |
case-02 | fail→fail | 22,590 | 4,867 | -78% | 1 | 1 | 0% | 3,689 | 676 | -82% | 0 | 0 | — |
case-03 | fail→fail | 18,239 | 5,183 | -72% | 1 | 1 | 0% | 3,003 | 719 | -76% | 0 | 0 | — |
case-04 | pass→pass | 18,440 | 18,894 | +2% | 1 | 1 | 0% | 3,242 | 4,270 | +32% | 0 | 0 | — |
case-05 | pass→pass | 11,596 | 13,350 | +15% | 1 | 1 | 0% | 2,187 | 2,940 | +34% | 0 | 0 | — |
case-06 | pass→fail | 15,441 | 4,775 | -69% | 1 | 1 | 0% | 2,661 | 1,128 | -58% | 0 | 0 | — |
case-07 | fail→pass | 19,537 | 7,496 | -62% | 1 | 1 | 0% | 3,147 | 1,765 | -44% | 0 | 0 | — |
case-08 | pass→pass | 7,430 | 7,175 | -3% | 1 | 1 | 0% | 1,105 | 972 | -12% | 0 | 0 | — |
case-09 | fail→pass | 17,251 | 5,174 | -70% | 1 | 1 | 0% | 3,021 | 1,364 | -55% | 0 | 0 | — |
case-10 | fail→fail | 12,084 | 4,972 | -59% | 1 | 1 | 0% | 1,807 | 729 | -60% | 0 | 0 | — |
case-11 | fail→pass | 7,966 | 5,582 | -30% | 1 | 1 | 0% | 1,197 | 1,324 | +11% | 0 | 0 | — |
case-12 | fail→fail | 18,472 | 4,703 | -75% | 1 | 1 | 0% | 2,990 | 628 | -79% | 0 | 0 | — |
case-13 | fail→pass | 21,178 | 3,892 | -82% | 1 | 1 | 0% | 3,800 | 1,085 | -71% | 0 | 0 | — |
case-14 | fail→pass | 26,493 | 6,493 | -75% | 1 | 1 | 0% | 4,161 | 1,584 | -62% | 0 | 0 | — |
case-15 | pass→pass | 12,082 | 5,807 | -52% | 1 | 1 | 0% | 2,140 | 1,402 | -34% | 0 | 0 | — |
case-16 | fail→pass | 7,333 | 7,299 | -0% | 1 | 1 | 0% | 1,188 | 1,634 | +38% | 0 | 0 | — |
case-17 | fail→pass | 9,349 | 1,769 | -81% | 1 | 1 | 0% | 1,648 | 709 | -57% | 0 | 0 | — |
case-18 | pass→fail | 8,007 | 4,880 | -39% | 1 | 1 | 0% | 1,233 | 672 | -45% | 0 | 0 | — |
case-19 | pass→fail | 3,157 | 2,711 | -14% | 1 | 1 | 0% | 480 | 743 | +55% | 0 | 0 | — |
case-20 | fail→pass | 17,470 | 1,919 | -89% | 1 | 1 | 0% | 710 | 695 | -2% | 0 | 0 | — |
case-21 | fail→fail | 20,934 | 3,697 | -82% | 1 | 1 | 0% | 3,102 | 1,060 | -66% | 0 | 0 | — |
case-22 | fail→fail | 20,028 | 3,250 | -84% | 1 | 1 | 0% | 3,125 | 860 | -72% | 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 15 counted toward the lift figure. The other 7 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 15 comparable cases. 4 cases got worse with the skill loaded, and they are 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.