Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Wenn es um EU-Schutzmassnahmen (Safeguards): Mengenbeschraenkungen und Unternehmensreaktion in Außenwirtschaft, Sanktionen, Zoll und CBAM geht: erstellt den passenden Entwurf aus Sachverhalt, Norm, Beweis und Antrag; liefert einen verwertbaren Entwurf mit Anträgen, Begründung und Anlagenlogik.
.claude/skills/klotzkette-aussenwirtschaft-schutzmassnahmen-safeguards/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 87% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 93% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 75% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 121% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 94% | 0% |
EU-Schutzmassnahmen (Safeguards) nach VO (EU) 2015/478 Art. 16 und VO (EU) 2015/755 Art. 13: Mengenbeschraenkungen Ueberwachungsmassnahmen und Antragstellung bei der EU-Kommission. Wirkung auf Importeure und Exporteure in betroffenen Sektoren (Stahl Solar Keramik). Abgrenzung zu Antidumping-Zoellen. Output: Safeguard-Prüfvermerk und Reaktionsplan. Arbeite entlang dieser konkreten Prüfungslinie und trenne Rolle, Frist, Zuständigkeit, Beweislast und gewünschten Output.
Dieser Skill kann mit thematisch benachbarten Skills kombiniert werden, insbesondere:
aussenwirtschaft-sanktionsscreening-fuzzy-matchaussenwirtschaft-gueterlisten-klassifizierungaussenwirtschaft-freiwillige-offenlegung-bafa-zollaussenwirtschaft-icp-kontrollsystem| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 30,902 | 42,682 | +38% | 1 | 1 | 0% | 4,355 | 8,149 | +87% | 0 | 0 | — |
case-02 | fail→pass | 30,530 | 41,569 | +36% | 1 | 1 | 0% | 4,572 | 8,831 | +93% | 0 | 0 | — |
case-03 | fail→fail | 35,081 | 42,638 | +22% | 1 | 1 | 0% | 4,540 | 7,850 | +73% | 0 | 0 | — |
case-04 | pass→pass | 23,471 | 30,500 | +30% | 1 | 1 | 0% | 2,824 | 5,987 | +112% | 0 | 0 | — |
case-05 | fail→pass | 16,309 | 15,156 | -7% | 1 | 1 | 0% | 2,023 | 3,531 | +75% | 0 | 0 | — |
case-06 | pass→pass | 13,197 | 19,745 | +50% | 1 | 1 | 0% | 1,413 | 4,091 | +190% | 0 | 0 | — |
case-07 | pass→pass | 20,890 | 15,296 | -27% | 1 | 1 | 0% | 2,203 | 3,323 | +51% | 0 | 0 | — |
case-08 | pass→pass | 21,000 | 21,214 | +1% | 1 | 1 | 0% | 2,433 | 4,604 | +89% | 0 | 0 | — |
case-09 | pass→pass | 25,700 | 27,445 | +7% | 1 | 1 | 0% | 3,295 | 5,478 | +66% | 0 | 0 | — |
case-10 | pass→pass | 18,557 | 16,523 | -11% | 1 | 1 | 0% | 2,297 | 4,470 | +95% | 0 | 0 | — |
case-11 | pass→pass | 22,460 | 29,474 | +31% | 1 | 1 | 0% | 2,914 | 5,202 | +79% | 0 | 0 | — |
case-12 | pass→pass | 17,506 | 24,082 | +38% | 1 | 1 | 0% | 2,000 | 4,634 | +132% | 0 | 0 | — |
case-13 | pass→pass | 17,471 | 25,500 | +46% | 1 | 1 | 0% | 2,047 | 5,026 | +146% | 0 | 0 | — |
case-14 | fail→pass | 24,875 | 35,085 | +41% | 1 | 1 | 0% | 3,147 | 6,947 | +121% | 0 | 0 | — |
case-15 | pass→pass | 22,513 | 19,965 | -11% | 1 | 1 | 0% | 2,526 | 4,116 | +63% | 0 | 0 | — |
case-16 | pass→fail | 13,758 | 27,151 | +97% | 1 | 1 | 0% | 2,173 | 5,153 | +137% | 0 | 0 | — |
case-17 | pass→pass | 12,177 | 26,110 | +114% | 1 | 1 | 0% | 1,958 | 5,121 | +162% | 0 | 0 | — |
case-18 | pass→pass | 18,121 | 27,390 | +51% | 1 | 1 | 0% | 2,251 | 5,508 | +145% | 0 | 0 | — |
case-19 | pass→pass | 18,251 | 24,207 | +33% | 1 | 1 | 0% | 2,507 | 5,120 | +104% | 0 | 0 | — |
case-20 | pass→pass | 20,102 | 23,792 | +18% | 1 | 1 | 0% | 2,373 | 4,841 | +104% | 0 | 0 | — |
case-21 | fail→pass | 17,984 | 19,732 | +10% | 1 | 1 | 0% | 2,186 | 4,246 | +94% | 0 | 0 | — |
case-22 | pass→pass | 13,967 | 29,476 | +111% | 1 | 1 | 0% | 2,247 | 5,763 | +156% | 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 +18 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.