Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →技能创建模板。复制此目录并修改 SKILL.md 即可创建新技能。请将所有 <占位符> 替换为实际内容。
.claude/skills/anbeime-skill-template/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 14% | 0% |
<描述用户需要提供什么输入>
| 参数 | 类型 | 必填 | 说明 | |------|------|------|------| | input_text | string | 是 | 输入文本 | | mode | string | 否 | 处理模式 (默认: auto) |
<描述技能的输出格式>
json{ "result": "处理结果", "metadata": { "processing_time": "0.5s", "status": "success" } }
<详细说明此步骤的逻辑>
<详细说明此步骤的逻辑>
<详细说明此步骤的逻辑>
输入:
<示例输入>输出:
<示例输出>输入:
<示例输入>输出:
<示例输出>| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 14,550 | 17,409 | +20% | 1 | 1 | 0% | 3,493 | 3,967 | +14% | 0 | 0 | — |
case-02 | pass→pass | 20,609 | 18,065 | -12% | 1 | 1 | 0% | 3,695 | 3,881 | +5% | 0 | 0 | — |
case-03 | pass→pass | 19,238 | 16,740 | -13% | 1 | 1 | 0% | 3,038 | 3,554 | +17% | 0 | 0 | — |
case-04 | fail→fail | 16,248 | 21,637 | +33% | 1 | 1 | 0% | 3,228 | 3,848 | +19% | 0 | 0 | — |
case-05 | pass→pass | 14,507 | 13,317 | -8% | 1 | 1 | 0% | 2,681 | 3,041 | +13% | 0 | 0 | — |
case-06 | pass→pass | 15,162 | 14,218 | -6% | 1 | 1 | 0% | 3,097 | 3,327 | +7% | 0 | 0 | — |
case-07 | fail→pass | 19,991 | 18,385 | -8% | 1 | 1 | 0% | 3,289 | 4,077 | +24% | 0 | 0 | — |
case-08 | fail→pass | 18,393 | 19,191 | +4% | 1 | 1 | 0% | 3,662 | 4,114 | +12% | 0 | 0 | — |
case-09 | fail→pass | 17,185 | 17,977 | +5% | 1 | 1 | 0% | 2,782 | 3,013 | +8% | 0 | 0 | — |
case-10 | pass→pass | 22,022 | 14,853 | -33% | 1 | 1 | 0% | 3,723 | 3,318 | -11% | 0 | 0 | — |
case-11 | fail→fail | 18,455 | 19,472 | +6% | 1 | 1 | 0% | 3,773 | 3,352 | -11% | 0 | 0 | — |
case-12 | pass→pass | 16,724 | 21,286 | +27% | 1 | 1 | 0% | 2,820 | 3,876 | +37% | 0 | 0 | — |
case-13 | fail→fail | 15,982 | 16,942 | +6% | 1 | 1 | 0% | 3,086 | 3,086 | 0% | 0 | 0 | — |
case-14 | fail→pass | 17,855 | 15,028 | -16% | 1 | 1 | 0% | 3,807 | 3,216 | -16% | 0 | 0 | — |
case-15 | pass→pass | 15,443 | 11,426 | -26% | 1 | 1 | 0% | 2,805 | 2,880 | +3% | 0 | 0 | — |
case-16 | pass→pass | 17,156 | 15,162 | -12% | 1 | 1 | 0% | 3,412 | 3,198 | -6% | 0 | 0 | — |
case-17 | fail→pass | 21,723 | 19,268 | -11% | 1 | 1 | 0% | 3,549 | 4,057 | +14% | 0 | 0 | — |
case-18 | fail→fail | 13,554 | 11,016 | -19% | 1 | 1 | 0% | 2,890 | 2,585 | -11% | 0 | 0 | — |
case-19 | fail→pass | 20,045 | 17,146 | -14% | 1 | 1 | 0% | 4,031 | 3,734 | -7% | 0 | 0 | — |
case-20 | pass→pass | 6,421 | 7,803 | +22% | 1 | 1 | 0% | 1,192 | 1,932 | +62% | 0 | 0 | — |
case-21 | pass→pass | 4,967 | 19,408 | +291% | 1 | 1 | 0% | 878 | 3,435 | +291% | 0 | 0 | — |
case-22 | pass→pass | 5,396 | 8,040 | +49% | 1 | 1 | 0% | 995 | 2,056 | +107% | 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 +27 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.