Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →当 cook-web 执行 npm run build 出现 SyntaxError Unexpected token '?' 或 next/dist/compiled 报错时使用本技能，优先排查 Node 运行时版本不一致问题。
.claude/skills/ai-shifu-next-build-node-runtime/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -56% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -13% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -69% | 0% |
npm run build 在 next/dist/compiled 内部文件报语法错误。SyntaxError: Unexpected token '?'、Unexpected token '.' 等现代语法不兼容信息。node -v、npm -v、which node、which npm。package.json 的 engines.node，确认是否匹配。npm run build。Error 作为阻断点，不把 Warning 当作失败原因。Error，做最小改动修复；保留调试日志，不删除 console.log。| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-10 | pass→pass | 15,659 | 4,295 | -73% | 1 | 1 | 0% | 2,468 | 862 | -65% | 0 | 0 | — |
case-04 | fail→pass | 9,245 | 3,003 | -68% | 1 | 1 | 0% | 1,489 | 653 | -56% | 0 | 0 | — |
case-01 | fail→fail | 19,121 | 3,705 | -81% | 1 | 1 | 0% | 3,087 | 850 | -72% | 0 | 0 | — |
case-02 | fail→fail | 17,595 | 10,057 | -43% | 1 | 1 | 0% | 2,607 | 1,898 | -27% | 0 | 0 | — |
case-03 | fail→pass | 13,092 | 10,192 | -22% | 1 | 1 | 0% | 2,208 | 1,929 | -13% | 0 | 0 | — |
case-05 | pass→pass | 7,745 | 2,486 | -68% | 1 | 1 | 0% | 1,250 | 567 | -55% | 0 | 0 | — |
case-06 | pass→pass | 10,568 | 3,167 | -70% | 1 | 1 | 0% | 1,722 | 726 | -58% | 0 | 0 | — |
case-07 | pass→pass | 12,978 | 6,006 | -54% | 1 | 1 | 0% | 1,904 | 1,066 | -44% | 0 | 0 | — |
case-08 | pass→pass | 15,678 | 8,214 | -48% | 1 | 1 | 0% | 2,307 | 1,479 | -36% | 0 | 0 | — |
case-09 | fail→pass | 13,780 | 5,200 | -62% | 1 | 1 | 0% | 2,167 | 987 | -54% | 0 | 0 | — |
case-11 | pass→pass | 15,657 | 5,614 | -64% | 1 | 1 | 0% | 2,429 | 1,153 | -53% | 0 | 0 | — |
case-12 | pass→pass | 8,191 | 1,852 | -77% | 1 | 1 | 0% | 1,411 | 468 | -67% | 0 | 0 | — |
case-13 | fail→pass | 10,057 | 5,857 | -42% | 1 | 1 | 0% | 1,645 | 1,240 | -25% | 0 | 0 | — |
case-14 | pass→pass | 18,680 | 4,992 | -73% | 1 | 1 | 0% | 2,691 | 936 | -65% | 0 | 0 | — |
case-15 | fail→pass | 11,477 | 2,465 | -79% | 1 | 1 | 0% | 1,793 | 557 | -69% | 0 | 0 | — |
case-16 | pass→pass | 15,426 | 2,843 | -82% | 1 | 1 | 0% | 2,260 | 635 | -72% | 0 | 0 | — |
case-17 | fail→pass | 11,944 | 3,270 | -73% | 1 | 1 | 0% | 1,837 | 771 | -58% | 0 | 0 | — |
case-18 | pass→pass | 14,441 | 5,834 | -60% | 1 | 1 | 0% | 2,268 | 1,141 | -50% | 0 | 0 | — |
case-19 | fail→pass | 15,932 | 11,564 | -27% | 1 | 1 | 0% | 2,308 | 1,879 | -19% | 0 | 0 | — |
case-20 | pass→pass | 13,887 | 9,329 | -33% | 1 | 1 | 0% | 2,222 | 1,676 | -25% | 0 | 0 | — |
case-21 | pass→pass | 14,982 | 10,641 | -29% | 1 | 1 | 0% | 2,699 | 2,106 | -22% | 0 | 0 | — |
case-22 | pass→pass | 15,277 | 10,964 | -28% | 1 | 1 | 0% | 2,653 | 2,163 | -18% | 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 +32 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.