Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when detecting and running project build systems automatically. Supports npm/yarn/pnpm/pip/poetry/gradle/maven/cargo/go/make. Triggers on build, test run, project setup, package manager detection.
.claude/skills/sangrokjung-build-system/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 164% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -59% | 0% |
프로젝트의 빌드 시스템을 자동으로 감지하고 적절한 빌드/테스트 명령어를 실행합니다.
| 빌드 시스템 | 감지 파일 | 빌드 명령어 | 테스트 명령어 | |------------|----------|------------|--------------| | npm | package.json | npm run build | npm test | | yarn | yarn.lock | yarn build | yarn test | | pnpm | pnpm-lock.yaml | pnpm build | pnpm test | | Python (pip) | requirements.txt | pip install -r requirements.txt | pytest | | Python (poetry) | pyproject.toml | poetry install | poetry run pytest | | Gradle | build.gradle | ./gradlew build | ./gradlew test | | Maven | pom.xml | mvn package | mvn test | | Cargo | Cargo.toml | cargo build | cargo test | | Go | go.mod | go build ./... | go test ./... | | Make | Makefile | make | make test |
프로젝트 루트에서 빌드 시스템을 자동 감지하고 실행합니다.
bash# 자동 감지 후 빌드 /build # 특정 명령어로 빌드 /build --cmd="npm run build:prod"
bash# 자동 감지 후 테스트 /test # 특정 테스트만 실행 /test --filter="unit"
package-lock.json → npmyarn.lock → yarnpnpm-lock.yaml → pnpmpyproject.toml → poetryrequirements.txt → pipCargo.toml → cargogo.mod → gobuild.gradle → gradlepom.xml → mavenMakefile → make프로젝트별로 .claude/config.json에서 빌드 명령어를 오버라이드할 수 있습니다:
json{ "build": { "command": "npm run build:custom", "test_command": "npm run test:ci" } }
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 3,068 | 2,409 | -21% | 1 | 1 | 0% | 396 | 1,047 | +164% | 0 | 0 | — |
case-02 | fail→fail | 2,799 | 2,095 | -25% | 1 | 1 | 0% | 262 | 777 | +197% | 0 | 0 | — |
case-03 | fail→fail | 2,285 | 4,913 | +115% | 1 | 1 | 0% | 242 | 1,462 | +504% | 0 | 0 | — |
case-04 | fail→pass | 9,423 | 2,880 | -69% | 1 | 1 | 0% | 1,734 | 1,131 | -35% | 0 | 0 | — |
case-05 | fail→pass | 7,535 | 3,811 | -49% | 1 | 1 | 0% | 1,367 | 1,091 | -20% | 0 | 0 | — |
case-06 | fail→pass | 10,326 | 2,853 | -72% | 1 | 1 | 0% | 1,687 | 1,078 | -36% | 0 | 0 | — |
case-07 | fail→pass | 12,367 | 2,382 | -81% | 1 | 1 | 0% | 2,457 | 1,005 | -59% | 0 | 0 | — |
case-08 | pass→pass | 12,159 | 2,707 | -78% | 1 | 1 | 0% | 2,114 | 1,084 | -49% | 0 | 0 | — |
case-09 | fail→pass | 9,640 | 1,531 | -84% | 1 | 1 | 0% | 1,829 | 882 | -52% | 0 | 0 | — |
case-10 | fail→pass | 4,607 | 1,891 | -59% | 1 | 1 | 0% | 848 | 942 | +11% | 0 | 0 | — |
case-11 | pass→pass | 7,920 | 1,576 | -80% | 1 | 1 | 0% | 1,394 | 868 | -38% | 0 | 0 | — |
case-12 | pass→pass | 8,065 | 1,519 | -81% | 1 | 1 | 0% | 1,547 | 830 | -46% | 0 | 0 | — |
case-13 | fail→fail | 5,087 | 1,594 | -69% | 1 | 1 | 0% | 903 | 812 | -10% | 0 | 0 | — |
case-14 | fail→pass | 8,268 | 2,367 | -71% | 1 | 1 | 0% | 1,477 | 1,060 | -28% | 0 | 0 | — |
case-15 | pass→pass | 5,075 | 1,648 | -68% | 1 | 1 | 0% | 1,082 | 865 | -20% | 0 | 0 | — |
case-16 | pass→pass | 8,275 | 2,301 | -72% | 1 | 1 | 0% | 1,643 | 978 | -40% | 0 | 0 | — |
case-17 | pass→pass | 5,140 | 2,479 | -52% | 1 | 1 | 0% | 1,112 | 1,133 | +2% | 0 | 0 | — |
case-18 | pass→pass | 6,526 | 1,538 | -76% | 1 | 1 | 0% | 1,172 | 798 | -32% | 0 | 0 | — |
case-19 | pass→pass | 9,689 | 3,462 | -64% | 1 | 1 | 0% | 1,647 | 1,175 | -29% | 0 | 0 | — |
case-20 | pass→pass | 5,094 | 3,410 | -33% | 1 | 1 | 0% | 918 | 1,196 | +30% | 0 | 0 | — |
case-21 | pass→pass | 7,272 | 5,723 | -21% | 1 | 1 | 0% | 1,465 | 1,665 | +14% | 0 | 0 | — |
case-22 | pass→pass | 2,121 | 1,545 | -27% | 1 | 1 | 0% | 405 | 841 | +108% | 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 +36 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.