Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Content-Disposition: form-data; name="file"; filename="SKILL.md"
.claude/skills/go-install-zh/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -1% | 0% |
--------------------------80187b9f2f8945b5 Content-Disposition: form-data; name="file"; filename="SKILL.md" Content-Type: application/octet-stream
name: go-install-zh version: 1.0.0 description: 在 Linux 环境安装 Go 编译器，用于 Go 项目编译和测试
在 Linux 环境安装 Go 编译器，用于 Go 项目编译和测试。
bashuname -m # x86_64 = amd64 # aarch64 = arm64
amd64 (x86_64):
bashcd /tmp curl -LO https://go.dev/dl/go1.22.0.linux-amd64.tar.gz tar -xzf go1.22.0.linux-amd64.tar.gz mv go ~/go-sdk rm go1.22.0.linux-amd64.tar.gz
arm64 (aarch64):
bashcd /tmp curl -LO https://go.dev/dl/go1.22.0.linux-arm64.tar.gz tar -xzf go1.22.0.linux-arm64.tar.gz mv go ~/go-sdk rm go1.22.0.linux-arm64.tar.gz
bashexport PATH=$PATH:~/go-sdk/bin export GOPATH=~/go export GOROOT=~/go-sdk
添加到 ~/.bashrc 或 ~/.profile：
bashecho 'export PATH=$PATH:~/go-sdk/bin' >> ~/.bashrc echo 'export GOPATH=~/go' >> ~/.bashrc echo 'export GOROOT=~/go-sdk' >> ~/.bashrc
bashgo version go env GOPATH GOROOT
bash# 运行测试 go test ./... # 运行测试（详细输出） go test ./... -v # 编译项目 go build -o <output> ./cmd/<entry> # 下载依赖 go mod download # 整理依赖 go mod tidy
| 项目 | 数值 | |------|------| | 安装包大小 | ~65MB | | 解压后大小 | ~300MB | | 内存需求 | 最低 512MB | | CPU 需求 | 单核即可 |
| 版本 | 说明 | |------|------| | go1.22.x | LTS 稳定版（推荐） | | go1.21.x | 旧稳定版 | | go1.23.x | 最新版 |
下载地址: https://go.dev/dl/ --------------------------80187b9f2f8945b5--
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 11,308 | 5,600 | -50% | 1 | 1 | 0% | 2,195 | 1,923 | -12% | 0 | 0 | — |
case-02 | fail→pass | 10,844 | 5,720 | -47% | 1 | 1 | 0% | 2,012 | 2,030 | +1% | 0 | 0 | — |
case-03 | fail→pass | 12,404 | 8,847 | -29% | 1 | 1 | 0% | 2,348 | 2,078 | -11% | 0 | 0 | — |
case-04 | pass→pass | 9,929 | 7,800 | -21% | 1 | 1 | 0% | 1,923 | 2,411 | +25% | 0 | 0 | — |
case-05 | pass→pass | 10,492 | 7,185 | -32% | 1 | 1 | 0% | 1,919 | 1,987 | +4% | 0 | 0 | — |
case-06 | pass→pass | 17,297 | 12,539 | -28% | 1 | 1 | 0% | 3,015 | 3,318 | +10% | 0 | 0 | — |
case-07 | pass→pass | 9,938 | 6,933 | -30% | 1 | 1 | 0% | 1,922 | 1,696 | -12% | 0 | 0 | — |
case-08 | pass→pass | 11,109 | 5,604 | -50% | 1 | 1 | 0% | 2,048 | 1,748 | -15% | 0 | 0 | — |
case-09 | fail→pass | 10,372 | 3,513 | -66% | 1 | 1 | 0% | 1,904 | 1,403 | -26% | 0 | 0 | — |
case-10 | fail→pass | 10,531 | 5,824 | -45% | 1 | 1 | 0% | 1,837 | 1,824 | -1% | 0 | 0 | — |
case-11 | fail→pass | 5,937 | 2,103 | -65% | 1 | 1 | 0% | 1,028 | 1,079 | +5% | 0 | 0 | — |
case-12 | fail→pass | 5,362 | 2,232 | -58% | 1 | 1 | 0% | 921 | 1,225 | +33% | 0 | 0 | — |
case-13 | fail→fail | 8,211 | 5,761 | -30% | 1 | 1 | 0% | 1,498 | 1,727 | +15% | 0 | 0 | — |
case-19 | pass→pass | 6,923 | 2,204 | -68% | 1 | 1 | 0% | 1,235 | 1,076 | -13% | 0 | 0 | — |
case-14 | pass→pass | 6,038 | 2,130 | -65% | 1 | 1 | 0% | 1,126 | 1,138 | +1% | 0 | 0 | — |
case-15 | pass→pass | 5,917 | 2,027 | -66% | 1 | 1 | 0% | 1,022 | 1,184 | +16% | 0 | 0 | — |
case-16 | pass→pass | 6,214 | 2,593 | -58% | 1 | 1 | 0% | 1,133 | 1,204 | +6% | 0 | 0 | — |
case-17 | pass→pass | 4,824 | 2,300 | -52% | 1 | 1 | 0% | 872 | 1,145 | +31% | 0 | 0 | — |
case-18 | pass→pass | 13,819 | 2,146 | -84% | 1 | 1 | 0% | 2,195 | 1,176 | -46% | 0 | 0 | — |
case-20 | fail→pass | 12,568 | 2,706 | -78% | 1 | 1 | 0% | 2,152 | 1,305 | -39% | 0 | 0 | — |
case-21 | pass→pass | 3,038 | 2,255 | -26% | 1 | 1 | 0% | 439 | 1,085 | +147% | 0 | 0 | — |
case-22 | pass→pass | 3,516 | 2,054 | -42% | 1 | 1 | 0% | 570 | 1,249 | +119% | 0 | 0 | — |
case-23 | pass→pass | 7,601 | 2,932 | -61% | 1 | 1 | 0% | 1,348 | 1,415 | +5% | 0 | 0 | — |
case-24 | pass→pass | 3,628 | 1,821 | -50% | 1 | 1 | 0% | 685 | 1,139 | +66% | 0 | 0 | — |
case-25 | pass→pass | 2,974 | 1,979 | -33% | 1 | 1 | 0% | 597 | 1,125 | +88% | 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. 25 cases were attempted. The headline lift of +32 percentage points is the difference between those two pass rates over the 25 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 7/27/2026 | +45% |
Other measured skills in the registry, with their headline benchmark lift.