Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Deprecated — forwards to bmad-review
.claude/skills/bmad-code-org-bmad-editorial-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 56% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 558% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 921% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 2615% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 3407% | 0% |
Merged into bmad-review. Invoke the bmad-review skill on the same content with the structure and prose lenses — both, structure first, so prose runs on top of the structure findings — unless the caller asked for a structure-only or prose-only review, in which case pass only that lens. Pass through any also_consider areas, and forward this skill's resolved [workflow] fields as pre-resolved values — but only those that resolved to something, since an empty value here means no legacy override exists and bmad-review's own default should stand: reader_type, style_guide, review_guidance, output_preferences, persistent_facts, activation_steps_prepend, activation_steps_append, on_complete, and review_output_path as the report path. Present the findings in the legacy shape: the two-pass findings table | Pass | Original Text | Revised Text | Changes | with the purpose/audience read above it and, when the structure pass ran, the reduction summary below it — and no other lens's output.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 8,469 | 45,139 | +433% | 1 | 1 | 0% | 364 | 8,491 | +2233% | 0 | 0 | — |
case-02 | fail→pass | 5,865 | 19,932 | +240% | 1 | 1 | 0% | 696 | 1,089 | +56% | 0 | 0 | — |
case-03 | fail→pass | 5,333 | 26,601 | +399% | 1 | 1 | 0% | 758 | 4,988 | +558% | 0 | 0 | — |
case-04 | fail→pass | 2,608 | 73,476 | +2717% | 1 | 1 | 0% | 272 | 2,777 | +921% | 0 | 0 | — |
case-05 | fail→fail | 9,548 | 48,364 | +407% | 1 | 1 | 0% | 1,420 | 2,091 | +47% | 0 | 0 | — |
case-06 | fail→fail | 39,238 | 21,768 | -45% | 1 | 1 | 0% | 740 | 639 | -14% | 0 | 0 | — |
case-07 | fail→fail | 12,008 | 11,007 | -8% | 1 | 1 | 0% | 393 | 790 | +101% | 0 | 0 | — |
case-08 | pass→fail | 19,962 | 51,783 | +159% | 1 | 1 | 0% | 3,026 | 1,895 | -37% | 0 | 0 | — |
case-09 | pass→fail | 10,008 | 9,779 | -2% | 1 | 1 | 0% | 1,512 | 777 | -49% | 0 | 0 | — |
case-10 | fail→fail | 8,723 | 16,407 | +88% | 1 | 1 | 0% | 997 | 553 | -45% | 0 | 0 | — |
case-11 | pass→fail | 20,836 | 75,700 | +263% | 1 | 1 | 0% | 1,641 | 854 | -48% | 0 | 0 | — |
case-12 | fail→fail | 15,199 | 14,887 | -2% | 1 | 1 | 0% | 967 | 879 | -9% | 0 | 0 | — |
case-13 | pass→pass | 14,243 | 53,679 | +277% | 1 | 1 | 0% | 1,672 | 5,276 | +216% | 0 | 0 | — |
case-14 | fail→pass | 17,926 | 35,279 | +97% | 1 | 1 | 0% | 276 | 7,493 | +2615% | 0 | 0 | — |
case-15 | pass→fail | 3,239 | 18,244 | +463% | 1 | 1 | 0% | 394 | 634 | +61% | 0 | 0 | — |
case-16 | fail→fail | 16,986 | 12,087 | -29% | 1 | 1 | 0% | 2,661 | 623 | -77% | 0 | 0 | — |
case-17 | fail→fail | 12,426 | 63,941 | +415% | 1 | 1 | 0% | 1,505 | 874 | -42% | 0 | 0 | — |
case-18 | pass→pass | 8,307 | 14,251 | +72% | 1 | 1 | 0% | 1,144 | 2,429 | +112% | 0 | 0 | — |
case-19 | pass→pass | 55,539 | 11,894 | -79% | 1 | 1 | 0% | 1,746 | 2,283 | +31% | 0 | 0 | — |
case-20 | pass→pass | 8,555 | 13,551 | +58% | 1 | 1 | 0% | 1,622 | 1,676 | +3% | 0 | 0 | — |
case-21 | fail→pass | 20,525 | 37,845 | +84% | 1 | 1 | 0% | 206 | 7,224 | +3407% | 0 | 0 | — |
case-22 | fail→pass | 5,685 | 29,792 | +424% | 1 | 1 | 0% | 306 | 4,787 | +1464% | 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 11 counted toward the lift figure. The other 11 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 +9 percentage points is the difference between those two pass rates over the 11 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.