Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →为古诗词提供配图与配乐的全流程创作指导；支持深度解析诗词意境、生成画面描述、提供配乐创作蓝图（Suno格式）；适用于诗词可视化、MV创作、文化传播等场景
.claude/skills/anbeime-poetry-music-visual/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 80% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 76% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 78% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 56% | 0% |
根据用户提供的信息（标题、作者、内容）进行多维度分析：
生成2-3组配图方案，每组包含：
生成规范：
生成纯音乐创作指导，严格遵循以下格式：
生成精准标签，用逗号分隔，需包含以下维度：
示例：
China-Chic, Cinematic, Ambient, Dreamy, Mysterious, Guzheng, Erhu, Pipa, Strings, Atmospheric Pads, Traditional Chinese Instruments, Modern Fusion, Slow Build, Rubato使用 Suno 可识别的结构标签，通过括号()添加器乐和氛围指令：
结构标签：
导演指令示例：
完整示例：
[Intro]
[Verse 1]
(Soft Guzheng plucking creates a mysterious atmosphere)
(Minimal strings provide a subtle backdrop)
[Chorus]
(Main theme emerges with fuller instrumentation)
(Harp and flutes join, creating an expansive cinematic feel)
[Instrumental Solo]
(Erhu takes center stage with a poignant melody)
(Strings and atmospheric pads provide support)
[Bridge]
(Music softens, becomes more reflective)
(Subtle wind chimes in the background)
[Final Chorus]
(Full orchestral texture returns)
(Guzheng and Erhu intertwine)
[Outro]
(Music gradually fades)
(Leaving only a soft wind effect)配图方案：
配乐蓝图：
Style of Music:
China-Chic, Ambient, Melancholic, Dreamy, Guzheng, Flute, Strings, Atmospheric Pads, Slow Build, Rubato
[Intro]
[Verse 1]
(Gentle Guzheng plucking, slow and contemplative)
(Soft atmospheric pads create a dreamy night atmosphere)
[Chorus]
(Flute melody enters with a nostalgic tone)
(Strings provide a subtle harmonic support)
[Bridge]
(Music becomes slightly more emotional)
(The full texture of strings and Guzheng)
[Outro]
(Guzheng melody gradually fades)
(Leaving only the sound of soft wind)配图方案：
配乐蓝图：
Style of Music:
China-Chic, Cinematic, Calm, Uplifting, Pipa, Guzheng, Dizi, Strings, Percussion, Mid-tempo, Rubato
[Intro]
[Verse 1]
(Gentle Pipa picking, mimicking flowing water)
(Soft Dizi melody, like birds singing in the forest)
[Chorus]
(Fuller instrumentation with Guzheng and strings)
(The feeling of autumn evening, peaceful and warm)
[Instrumental Solo]
(Dizi takes a brief melodic solo)
(Representing the flowing water and natural sounds)
[Bridge]
(Music becomes slightly more rhythmic)
(Light percussion like water drums or woodblocks)
[Final Chorus]
(Full ensemble returns with warm harmonies)
(Capturing the joy and tranquility of mountain life)
[Outro]
(Music gradually softens)
(Like the evening settling into night)| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→fail | 29,023 | 22,495 | -22% | 1 | 1 | 0% | 3,547 | 5,742 | +62% | 0 | 0 | — |
case-01 | fail→fail | 24,031 | 23,285 | -3% | 1 | 1 | 0% | 3,717 | 4,932 | +33% | 0 | 0 | — |
case-02 | fail→pass | 24,236 | 22,341 | -8% | 1 | 1 | 0% | 3,685 | 4,874 | +32% | 0 | 0 | — |
case-04 | pass→pass | 19,029 | 21,197 | +11% | 1 | 1 | 0% | 2,886 | 4,988 | +73% | 0 | 0 | — |
case-05 | pass→pass | 12,794 | 13,151 | +3% | 1 | 1 | 0% | 2,169 | 4,405 | +103% | 0 | 0 | — |
case-06 | pass→pass | 13,438 | 14,922 | +11% | 1 | 1 | 0% | 2,475 | 4,227 | +71% | 0 | 0 | — |
case-07 | pass→pass | 9,118 | 6,512 | -29% | 1 | 1 | 0% | 1,782 | 3,365 | +89% | 0 | 0 | — |
case-08 | pass→pass | 20,100 | 11,900 | -41% | 1 | 1 | 0% | 2,752 | 4,015 | +46% | 0 | 0 | — |
case-09 | pass→pass | 15,231 | 14,691 | -4% | 1 | 1 | 0% | 2,097 | 4,122 | +97% | 0 | 0 | — |
case-10 | fail→pass | 17,635 | 13,272 | -25% | 1 | 1 | 0% | 2,455 | 4,429 | +80% | 0 | 0 | — |
case-11 | pass→pass | 20,543 | 13,845 | -33% | 1 | 1 | 0% | 3,075 | 4,516 | +47% | 0 | 0 | — |
case-12 | pass→pass | 15,245 | 15,449 | +1% | 1 | 1 | 0% | 2,221 | 4,164 | +87% | 0 | 0 | — |
case-13 | pass→pass | 9,890 | 10,446 | +6% | 1 | 1 | 0% | 1,674 | 3,794 | +127% | 0 | 0 | — |
case-14 | pass→pass | 11,232 | 7,738 | -31% | 1 | 1 | 0% | 1,930 | 3,492 | +81% | 0 | 0 | — |
case-15 | fail→pass | 19,909 | 33,143 | +66% | 1 | 1 | 0% | 3,043 | 5,370 | +76% | 0 | 0 | — |
case-16 | pass→pass | 22,060 | 20,605 | -7% | 1 | 1 | 0% | 3,553 | 5,235 | +47% | 0 | 0 | — |
case-17 | pass→pass | 19,345 | 17,262 | -11% | 1 | 1 | 0% | 2,977 | 4,862 | +63% | 0 | 0 | — |
case-18 | fail→pass | 15,699 | 16,834 | +7% | 1 | 1 | 0% | 2,645 | 4,711 | +78% | 0 | 0 | — |
case-19 | fail→pass | 20,191 | 16,305 | -19% | 1 | 1 | 0% | 3,203 | 4,991 | +56% | 0 | 0 | — |
case-20 | fail→fail | 27,290 | 29,827 | +9% | 1 | 1 | 0% | 5,219 | 7,750 | +48% | 0 | 0 | — |
case-21 | fail→fail | 25,623 | 29,141 | +14% | 1 | 1 | 0% | 3,475 | 6,479 | +86% | 0 | 0 | — |
case-22 | fail→fail | 15,258 | 19,672 | +29% | 1 | 1 | 0% | 3,099 | 5,625 | +82% | 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. The headline lift of +23 percentage points is the difference between those two pass rates over the 22 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.