Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Create new genre documentation files for the bitwize-music genre library. Use when the user wants to add a genre, says "/genre-creator", "neues Genre erstellen", "Genre hinzufuegen", "add genre", or asks to create genre documentation. Takes a genre name as argument.
.claude/skills/bitwize-music-studio-genre-creator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-16 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 123% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-10 | ✓→✗ | ▼ Worse | 4% | 0% |
Create a new genre README.md for the bitwize-music genre library at ${CLAUDE_PLUGIN_ROOT}/genres/.
Input: $ARGUMENTS (genre name, e.g. "Math Rock", "Nu-Metal", "City Pop")
math-rock)genres/{slug}/README.md exists → abort, inform usergenres/INDEX.md to confirm genre is not already listedgenres/hip-hop/README.md, genres/phonk/README.md)genres/{slug}/tools/mastering/genre-presets.yaml — Add YAML entry with target_lufs, cut_highmid, cut_highs values appropriate for the genre. Place in the correct category section or create a new one.skills/mastering-engineer/genre-presets.md — Add a new ### Genre Name section under ## Genre Presets with LUFS target, dynamics, EQ focus, MCP command, and characteristics.artists/ subdirectory — those are created separately when artist deep-dives are writtenThe file starts directly with # Genre Name — no YAML frontmatter.
ALWAYS use this exact section order:
# {Genre Name}
## Genre Overview
[3 paragraphs — see rules below]
## Characteristics
[6 bullet fields — see rules below]
## Lyric Conventions
[6 bullet fields — see rules below]
## Subgenres & Styles
[Table — see rules below]
## Artists
[Table — see rules below]
## Suno Prompt Keywords
[Code block — see rules below]
## Reference Tracks
[List — see rules below]## Genre Overview — 3 paragraphs of prose (no bullets):
## Characteristics — Bullet list, exactly these 6 fields:
## Lyric Conventions — Bullet list, exactly these 6 fields:
Default **X lines/verse** at Y BPM. [Context]. Topics: Z/verse.## Subgenres & Styles — Markdown table:
| Style | Description | Reference Artists | |-------|-------------|-------------------|
## Artists — Markdown table:
| Artist | Key Albums | Era | Style Focus | |--------|-----------|-----|-------------|
Deep Dive link to the artist file in Style Focus## Suno Prompt Keywords — Fenced code block with comma-separated keywords organized in thematic lines:
## Reference Tracks — 10-15 entries:
- **Artist - "Track Title"** — [Description]| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | pass→pass | 17,787 | 18,995 | +7% | 1 | 1 | 0% | 3,586 | 5,555 | +55% | 0 | 0 | — |
case-06 | pass→pass | 13,280 | 13,572 | +2% | 1 | 1 | 0% | 2,137 | 3,563 | +67% | 0 | 0 | — |
case-16 | fail→pass | 18,264 | 18,056 | -1% | 1 | 1 | 0% | 3,162 | 4,517 | +43% | 0 | 0 | — |
case-21 | pass→pass | 13,688 | 9,127 | -33% | 1 | 1 | 0% | 1,519 | 2,942 | +94% | 0 | 0 | — |
case-04 | pass→pass | 23,697 | 43,719 | +84% | 1 | 1 | 0% | 3,873 | 7,549 | +95% | 0 | 0 | — |
case-01 | fail→fail | 34,619 | 6,532 | -81% | 1 | 1 | 0% | 5,883 | 1,792 | -70% | 0 | 0 | — |
case-02 | fail→fail | 36,916 | 6,475 | -82% | 1 | 1 | 0% | 6,212 | 1,847 | -70% | 0 | 0 | — |
case-03 | fail→fail | 32,492 | 5,539 | -83% | 1 | 1 | 0% | 5,713 | 1,662 | -71% | 0 | 0 | — |
case-07 | fail→fail | 16,887 | 4,942 | -71% | 1 | 1 | 0% | 3,044 | 1,682 | -45% | 0 | 0 | — |
case-08 | fail→fail | 15,389 | 5,769 | -63% | 1 | 1 | 0% | 2,897 | 1,656 | -43% | 0 | 0 | — |
case-09 | fail→fail | 16,615 | 5,259 | -68% | 1 | 1 | 0% | 2,754 | 1,581 | -43% | 0 | 0 | — |
case-10 | pass→fail | 11,650 | 5,805 | -50% | 1 | 1 | 0% | 1,704 | 1,769 | +4% | 0 | 0 | — |
case-11 | fail→fail | 11,436 | 6,142 | -46% | 1 | 1 | 0% | 1,691 | 1,677 | -1% | 0 | 0 | — |
case-12 | fail→pass | 24,520 | 41,274 | +68% | 1 | 1 | 0% | 3,388 | 7,539 | +123% | 0 | 0 | — |
case-13 | fail→fail | 19,393 | 5,026 | -74% | 1 | 1 | 0% | 3,295 | 1,554 | -53% | 0 | 0 | — |
case-14 | fail→fail | 22,503 | 4,974 | -78% | 1 | 1 | 0% | 3,845 | 1,571 | -59% | 0 | 0 | — |
case-15 | fail→pass | 13,921 | 6,713 | -52% | 1 | 1 | 0% | 2,212 | 2,520 | +14% | 0 | 0 | — |
case-17 | fail→fail | 8,053 | 2,898 | -64% | 1 | 1 | 0% | 1,283 | 1,837 | +43% | 0 | 0 | — |
case-18 | fail→fail | 16,852 | 6,109 | -64% | 1 | 1 | 0% | 2,779 | 1,692 | -39% | 0 | 0 | — |
case-19 | fail→fail | 12,275 | 5,204 | -58% | 1 | 1 | 0% | 2,312 | 1,507 | -35% | 0 | 0 | — |
case-20 | fail→pass | 9,256 | 2,654 | -71% | 1 | 1 | 0% | 1,487 | 1,759 | +18% | 0 | 0 | — |
case-22 | fail→fail | 11,132 | 2,385 | -79% | 1 | 1 | 0% | 1,823 | 1,751 | -4% | 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 10 counted toward the lift figure. The other 12 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 +14 percentage points is the difference between those two pass rates over the 10 comparable cases. 5 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.