Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Wenn es um Akteneinsicht-Uebersicht in Aktenaufbereiter Strafrecht geht: ordnet Akteninhalt, Belege, Lücken und Nachforderungen; liefert eine Beweislast- und Substantiierungsmatrix.
.claude/skills/klotzkette-akteneinsicht-uebersicht/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 61% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 72% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 79% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 46% | 0% |
Akteneinsicht systematisch auswerten: Aktenbestandteile (Haupt-, Sonder-, Beweismittelakte), Bandzaehlung, Aktenblattzahl je Band, fehlende Aktenstuecke, Verschluss-Sachen, Tonbaender/Datentraeger, polizeiliche Spurenakten.
Dieser Skill ist kein allgemeiner Chat-Modus, sondern ein Arbeitswerkzeug fuer aktenaufbereiter-strafrecht zum Thema akteneinsicht-uebersicht. Ausgangspunkt ist immer die konkrete Aufgabe aus der Beschreibung: Akteneinsicht systematisch auswerten: Aktenbestandteile (Haupt-, Sonder-, Beweismittelakte), Bandzaehlung, Aktenblattzahl je Band, fehlende Aktenstuecke, Verschluss-Sachen, Tonbaender/Datentraeger, polizeiliche Spurenakten. Fuehrt Pruef-Checkliste und Nachforderungsschreiben an die Staatsanwaltschaft..
Arbeite deshalb fallnah:
Wenn ein anderer Skill desselben Plugins genauer passt, schlage ihn aktiv vor und erklaere in einem Satz, warum der Wechsel die Arbeit beschleunigt.
Frage zu Beginn nur ab, was fuer den naechsten Schritt unverzichtbar ist. Wenn Material vorliegt, mit dem Material arbeiten und nur eine gezielte Rueckfrage stellen.
Der Skill erwartet folgenden inhaltlichen Aufbau im Output:
Dieser Skill gehoert zum Plugin aktenaufbereiter-strafrecht. Er ergaenzt die uebrigen Skills des Plugins um einen vertieften Spezialfall oder eine systematische Einfuehrung. Bei Folgefragen werden andere Skills des Plugins als Anschluss vorgeschlagen.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 22,592 | 22,388 | -1% | 1 | 1 | 0% | 3,777 | 4,915 | +30% | 0 | 0 | — |
case-02 | fail→pass | 24,605 | 25,497 | +4% | 1 | 1 | 0% | 4,254 | 5,503 | +29% | 0 | 0 | — |
case-03 | fail→fail | 24,685 | 23,730 | -4% | 1 | 1 | 0% | 3,737 | 5,394 | +44% | 0 | 0 | — |
case-04 | fail→pass | 16,194 | 17,673 | +9% | 1 | 1 | 0% | 2,608 | 4,199 | +61% | 0 | 0 | — |
case-05 | pass→pass | 15,344 | 19,651 | +28% | 1 | 1 | 0% | 2,311 | 4,491 | +94% | 0 | 0 | — |
case-06 | fail→fail | 11,030 | 9,533 | -14% | 1 | 1 | 0% | 1,751 | 2,777 | +59% | 0 | 0 | — |
case-07 | fail→pass | 17,491 | 24,925 | +43% | 1 | 1 | 0% | 3,230 | 5,557 | +72% | 0 | 0 | — |
case-08 | fail→fail | 12,787 | 15,294 | +20% | 1 | 1 | 0% | 2,194 | 3,919 | +79% | 0 | 0 | — |
case-09 | fail→fail | 13,732 | 10,800 | -21% | 1 | 1 | 0% | 2,444 | 3,054 | +25% | 0 | 0 | — |
case-10 | pass→pass | 16,616 | 20,491 | +23% | 1 | 1 | 0% | 2,855 | 4,734 | +66% | 0 | 0 | — |
case-11 | fail→pass | 13,544 | 17,006 | +26% | 1 | 1 | 0% | 2,266 | 4,052 | +79% | 0 | 0 | — |
case-12 | pass→fail | 10,942 | 15,298 | +40% | 1 | 1 | 0% | 1,987 | 3,865 | +95% | 0 | 0 | — |
case-13 | fail→pass | 11,451 | 10,182 | -11% | 1 | 1 | 0% | 1,958 | 2,868 | +46% | 0 | 0 | — |
case-14 | fail→fail | 15,612 | 18,513 | +19% | 1 | 1 | 0% | 2,720 | 4,309 | +58% | 0 | 0 | — |
case-15 | pass→pass | 14,031 | 18,988 | +35% | 1 | 1 | 0% | 2,567 | 4,538 | +77% | 0 | 0 | — |
case-16 | fail→pass | 13,075 | 14,414 | +10% | 1 | 1 | 0% | 2,218 | 3,497 | +58% | 0 | 0 | — |
case-17 | fail→pass | 14,899 | 17,403 | +17% | 1 | 1 | 0% | 2,457 | 4,120 | +68% | 0 | 0 | — |
case-18 | fail→fail | 13,616 | 15,222 | +12% | 1 | 1 | 0% | 2,198 | 3,661 | +67% | 0 | 0 | — |
case-19 | fail→pass | 18,557 | 17,523 | -6% | 1 | 1 | 0% | 3,223 | 4,201 | +30% | 0 | 0 | — |
case-20 | fail→pass | 20,232 | 22,666 | +12% | 1 | 1 | 0% | 3,751 | 4,902 | +31% | 0 | 0 | — |
case-21 | pass→pass | 9,908 | 14,219 | +44% | 1 | 1 | 0% | 1,642 | 3,667 | +123% | 0 | 0 | — |
case-22 | fail→fail | 11,786 | 13,971 | +19% | 1 | 1 | 0% | 2,086 | 3,522 | +69% | 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. 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.