Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Wenn es um Aktenvorblatt: Schriftsatz-, Brief- und Memo-Bausteine in Aktenaufbereiter Strafrecht geht: erstellt den passenden Entwurf aus Sachverhalt, Norm, Beweis und Antrag; liefert einen verwertbaren Entwurf mit Anträgen, Begründung und Anlagenlogik.
.claude/skills/klotzkette-aktenvorblatt-schriftsatz-brief-und-memo-bausteine/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 56% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 74% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 99% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 66% | 0% |
Wenn Unterlagen vorhanden sind, arbeite zuerst aus den Unterlagen. Stelle nur Rückfragen, die die nächste Weiche verändern:
STRAFSACHE Mandantenname] Aktenzeichen Verteidigung: ...
Mandant: Name, geb. ..., wohnhaft ...] Mandantenrolle: Beschuldigter / Angeklagter / Verletzter / Zeuge Verteidiger: Name, Kanzlei] Gegenseite: Staatsanwaltschaft Ort, Az ...]
Ermittlungsverfahren / Hauptverfahren / Rechtsmittelverfahren Az StA: ... Az Gericht: ... Sachbearbeiter StA: ... (Tel.) Richter: ... (Schoeffengericht / Strafkammer / Strafrichter)
Tatzeit: ... Tatort: ... Gesetzliche Merkmale: § ... StGB ... Strafrahmen: ... bis ... Jahre Freiheitsstrafe / Geldstrafe
Haftbefehl vom ... | Haftgrund ... | Pruefung OLG am ... (§ 121 StPO)
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | fail→fail | 20,846 | 22,982 | +10% | 1 | 1 | 0% | 3,297 | 4,941 | +50% | 0 | 0 | — |
case-01 | fail→fail | 17,388 | 20,083 | +15% | 1 | 1 | 0% | 3,237 | 4,934 | +52% | 0 | 0 | — |
case-02 | fail→fail | 19,217 | 24,107 | +25% | 1 | 1 | 0% | 3,259 | 5,416 | +66% | 0 | 0 | — |
case-03 | fail→fail | 20,712 | 25,636 | +24% | 1 | 1 | 0% | 3,503 | 5,861 | +67% | 0 | 0 | — |
case-04 | fail→fail | 17,449 | 12,358 | -29% | 1 | 1 | 0% | 2,921 | 3,710 | +27% | 0 | 0 | — |
case-06 | fail→pass | 13,194 | 13,496 | +2% | 1 | 1 | 0% | 2,446 | 3,809 | +56% | 0 | 0 | — |
case-07 | fail→pass | 10,493 | 10,864 | +4% | 1 | 1 | 0% | 1,857 | 3,226 | +74% | 0 | 0 | — |
case-08 | pass→pass | 19,683 | 17,515 | -11% | 1 | 1 | 0% | 3,417 | 4,481 | +31% | 0 | 0 | — |
case-09 | pass→fail | 12,975 | 17,807 | +37% | 1 | 1 | 0% | 2,417 | 4,512 | +87% | 0 | 0 | — |
case-10 | pass→fail | 15,355 | 17,450 | +14% | 1 | 1 | 0% | 2,651 | 4,390 | +66% | 0 | 0 | — |
case-11 | fail→fail | 14,542 | 19,775 | +36% | 1 | 1 | 0% | 2,801 | 4,785 | +71% | 0 | 0 | — |
case-12 | fail→pass | 25,578 | 17,764 | -31% | 1 | 1 | 0% | 4,572 | 4,186 | -8% | 0 | 0 | — |
case-13 | fail→pass | 9,897 | 13,623 | +38% | 1 | 1 | 0% | 1,838 | 3,653 | +99% | 0 | 0 | — |
case-14 | pass→pass | 12,884 | 17,179 | +33% | 1 | 1 | 0% | 2,637 | 4,579 | +74% | 0 | 0 | — |
case-15 | pass→pass | 23,862 | 23,755 | -0% | 1 | 1 | 0% | 4,030 | 5,534 | +37% | 0 | 0 | — |
case-16 | pass→pass | 14,207 | 15,423 | +9% | 1 | 1 | 0% | 2,621 | 4,370 | +67% | 0 | 0 | — |
case-17 | pass→pass | 15,811 | 17,317 | +10% | 1 | 1 | 0% | 2,663 | 4,253 | +60% | 0 | 0 | — |
case-18 | fail→pass | 16,618 | 19,283 | +16% | 1 | 1 | 0% | 2,731 | 4,531 | +66% | 0 | 0 | — |
case-19 | fail→pass | 11,819 | 9,187 | -22% | 1 | 1 | 0% | 2,049 | 2,847 | +39% | 0 | 0 | — |
case-20 | fail→pass | 19,251 | 19,533 | +1% | 1 | 1 | 0% | 3,361 | 4,686 | +39% | 0 | 0 | — |
case-21 | fail→pass | 16,554 | 19,600 | +18% | 1 | 1 | 0% | 3,006 | 4,679 | +56% | 0 | 0 | — |
case-22 | fail→pass | 12,252 | 16,872 | +38% | 1 | 1 | 0% | 2,039 | 4,082 | +100% | 0 | 0 | — |
case-23 | fail→pass | 15,490 | 19,898 | +28% | 1 | 1 | 0% | 2,550 | 4,643 | +82% | 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. 23 cases were attempted. The headline lift of +35 percentage points is the difference between those two pass rates over the 23 comparable cases. 3 cases got worse with the skill loaded, and they are 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.