Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Wenn es um Zolltarif, TARIC und vZTA in Außenwirtschaft, Sanktionen, Zoll und CBAM geht: prüft Frist, Form, Zuständigkeit, Rechtsweg und Sofortmaßnahmen; liefert eine Fristen- und Risikoampel mit Sofortschritten.
.claude/skills/klotzkette-aussenwirtschaft-zolltarif-vzta/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 46% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 70% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 48% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 208% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 99% | 0% |
Dieser Skill macht aus Produktbeschreibung, technischen Daten und Verwendungszweck eine zolltarifliche Prüfakte.
Kläre vor der Tarifierung:
Quellenregel: Keine Kommentar-, Handbuch- oder Aufsatzfundstellen aus Modellwissen; Literatur nur mit Nutzerquelle oder lizenziertem Live-Zugriff.
Adressat: Zollabteilung / Buchhaltung — Tonfall: technisch, nomenklatur-prazise
TARIFIERUNGSVERMERK
Datum: [DATUM]
Ware: [HANDELSBEZEICHNUNG]
Technische Beschreibung: [KURZBESCHREIBUNG]
Bearbeiter: [NAME]
1. VORGESCHLAGENE EINREIHUNG
KN-Nr. (8-steller): [NUMMER]
TARIC-Code (10-steller): [NUMMER]
Kapitel / Position: [KAP.] / [POS.]
Begründung: Allgemeine Vorschrift Nr. [1/2/3a/3b/3c/4/5/6]
Angewandte KN-Erlauterung: [FUNDSTELLE ABl.]
2. ZOLLSATZ
Drittlandzollsatz: [%]
Praferenzzollsatz (Abkommen [NAME]): [%]
Antidumping/Ausgleichszoll: [%] (VO (EU) [NR.])
3. VZTA-STATUS
Bestehende vZTA: [ ] Nr. [NUMMER] gueltig bis [DATUM] / [ ] Keine
vZTA-Antrag empfohlen: [ ] Ja / [ ] Nein
4. RISIKEN
Alternativposition: [KN-NR.] — Risiko: [BESCHREIBUNG]
Empfehlung: [vZTA / interne Dokumentation / Anwaltliches Opinion Letter]> Quellenregel: Entscheidungen nur nach Prüfung einer amtlichen oder frei zugänglichen Quelle mit Gericht, Entscheidungsform, Datum, Aktenzeichen und tragender Aussage ausgeben.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 35,164 | 34,067 | -3% | 1 | 1 | 0% | 4,563 | 6,675 | +46% | 0 | 0 | — |
case-02 | fail→pass | 28,105 | 33,431 | +19% | 1 | 1 | 0% | 4,051 | 6,889 | +70% | 0 | 0 | — |
case-03 | fail→pass | 32,230 | 35,216 | +9% | 1 | 1 | 0% | 4,523 | 6,684 | +48% | 0 | 0 | — |
case-04 | fail→pass | 26,825 | 34,007 | +27% | 1 | 1 | 0% | 1,824 | 5,612 | +208% | 0 | 0 | — |
case-05 | pass→pass | 19,859 | 27,163 | +37% | 1 | 1 | 0% | 2,340 | 5,382 | +130% | 0 | 0 | — |
case-06 | fail→pass | 24,315 | 31,497 | +30% | 1 | 1 | 0% | 3,104 | 6,173 | +99% | 0 | 0 | — |
case-07 | pass→fail | 19,657 | 24,725 | +26% | 1 | 1 | 0% | 2,526 | 4,960 | +96% | 0 | 0 | — |
case-08 | fail→pass | 26,876 | 29,949 | +11% | 1 | 1 | 0% | 3,475 | 5,962 | +72% | 0 | 0 | — |
case-09 | fail→pass | 26,766 | 31,909 | +19% | 1 | 1 | 0% | 3,060 | 6,168 | +102% | 0 | 0 | — |
case-10 | fail→pass | 26,770 | 30,092 | +12% | 1 | 1 | 0% | 3,001 | 6,059 | +102% | 0 | 0 | — |
case-11 | fail→pass | 29,346 | 31,369 | +7% | 1 | 1 | 0% | 3,582 | 6,115 | +71% | 0 | 0 | — |
case-12 | fail→pass | 28,909 | 25,648 | -11% | 1 | 1 | 0% | 3,881 | 5,322 | +37% | 0 | 0 | — |
case-13 | fail→pass | 26,836 | 39,699 | +48% | 1 | 1 | 0% | 3,675 | 7,639 | +108% | 0 | 0 | — |
case-14 | fail→fail | 20,132 | 33,379 | +66% | 1 | 1 | 0% | 2,410 | 6,388 | +165% | 0 | 0 | — |
case-15 | fail→fail | 25,356 | 22,876 | -10% | 1 | 1 | 0% | 2,825 | 4,695 | +66% | 0 | 0 | — |
case-16 | fail→pass | 16,684 | 31,552 | +89% | 1 | 1 | 0% | 2,720 | 5,241 | +93% | 0 | 0 | — |
case-17 | fail→pass | 32,646 | 40,404 | +24% | 1 | 1 | 0% | 4,069 | 6,494 | +60% | 0 | 0 | — |
case-18 | pass→pass | 12,701 | 18,055 | +42% | 1 | 1 | 0% | 2,210 | 4,755 | +115% | 0 | 0 | — |
case-19 | pass→pass | 15,756 | 19,674 | +25% | 1 | 1 | 0% | 2,224 | 4,840 | +118% | 0 | 0 | — |
case-20 | pass→pass | 20,357 | 19,264 | -5% | 1 | 1 | 0% | 2,523 | 4,796 | +90% | 0 | 0 | — |
case-21 | pass→pass | 25,586 | 23,596 | -8% | 1 | 1 | 0% | 3,554 | 5,692 | +60% | 0 | 0 | — |
case-22 | pass→pass | 20,877 | 18,622 | -11% | 1 | 1 | 0% | 3,722 | 4,910 | +32% | 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 +55 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.