Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Mandant oder Business-Team versteht Common-Law-Konzepte nicht und braucht verstaendliche Erklärung. Anwendungsfall Deutsche kaufen UK-Unternehmen oder schließen US-Vertrag ab. Prüfraster Verstaendlichkeit Vollständigkeit Wesentliche-Punkte-Abdeckung Risiko-Klarheit. Output verstaendliches Erklärungsschreiben oder Memo für Mandanten. Abgrenzung zu common-law-begriffe-uebersetzung (Fachsprache) und common-law-humor-coach (Ton).
.claude/skills/thomasmoreai-common-law-client-explainer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 57% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 66% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 46% | 0% |
| case-06 | ✓→✓ | = Same ✓ | 29% | 0% |
Management und Business-Team bekommen klare Entscheidungspunkte statt Wörterbuchdogmatik.
Common-Law-Kompass arbeitet freundlich, präzise und verzeihend. Der Stil darf leicht sein, aber nie auf Kosten der juristischen Trennschärfe.
Before proceeding, clarify:
Addressee: German client unfamiliar with common law Tone: Clear, non-jargon; German equivalents provided where helpful
CLIENT EXPLAINER: [COMMON LAW CONCEPT]
Date: [DATE] — Prepared for: [CLIENT NAME]
Jurisdiction: [England/Wales / New York]
WHAT IT MEANS
[Plain English explanation in 3-5 sentences]
GERMAN EQUIVALENT (OR WHY THERE ISN'T ONE)
[Comparison: similar to [§ X BGB] BUT differs in [KEY WAY]]
WHY IT MATTERS FOR YOUR SITUATION
[Practical impact: deal / litigation / compliance]
WHAT TO DO
[Action point]
FURTHER READING
[Accessible reference / summary source]| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 20,170 | 17,064 | -15% | 1 | 1 | 0% | 3,293 | 3,751 | +14% | 0 | 0 | — |
case-02 | fail→pass | 17,870 | 21,482 | +20% | 1 | 1 | 0% | 3,264 | 4,668 | +43% | 0 | 0 | — |
case-03 | fail→fail | 20,545 | 17,006 | -17% | 1 | 1 | 0% | 3,371 | 3,849 | +14% | 0 | 0 | — |
case-04 | pass→pass | 14,004 | 17,312 | +24% | 1 | 1 | 0% | 2,387 | 3,971 | +66% | 0 | 0 | — |
case-05 | pass→pass | 18,124 | 16,804 | -7% | 1 | 1 | 0% | 2,783 | 4,058 | +46% | 0 | 0 | — |
case-06 | pass→pass | 19,073 | 17,253 | -10% | 1 | 1 | 0% | 2,933 | 3,781 | +29% | 0 | 0 | — |
case-07 | pass→pass | 18,083 | 20,827 | +15% | 1 | 1 | 0% | 3,004 | 4,115 | +37% | 0 | 0 | — |
case-08 | pass→pass | 19,924 | 25,545 | +28% | 1 | 1 | 0% | 3,286 | 3,998 | +22% | 0 | 0 | — |
case-09 | pass→pass | 15,474 | 17,433 | +13% | 1 | 1 | 0% | 2,703 | 3,851 | +42% | 0 | 0 | — |
case-10 | pass→pass | 12,155 | 15,047 | +24% | 1 | 1 | 0% | 2,071 | 3,516 | +70% | 0 | 0 | — |
case-11 | pass→pass | 13,004 | 20,702 | +59% | 1 | 1 | 0% | 2,287 | 3,427 | +50% | 0 | 0 | — |
case-12 | pass→pass | 16,337 | 17,733 | +9% | 1 | 1 | 0% | 2,583 | 3,801 | +47% | 0 | 0 | — |
case-13 | pass→pass | 20,198 | 23,898 | +18% | 1 | 1 | 0% | 2,784 | 4,056 | +46% | 0 | 0 | — |
case-14 | pass→pass | 10,611 | 17,806 | +68% | 1 | 1 | 0% | 1,836 | 3,501 | +91% | 0 | 0 | — |
case-15 | pass→pass | 15,210 | 18,305 | +20% | 1 | 1 | 0% | 2,574 | 4,104 | +59% | 0 | 0 | — |
case-16 | pass→pass | 16,966 | 16,536 | -3% | 1 | 1 | 0% | 2,697 | 3,596 | +33% | 0 | 0 | — |
case-17 | pass→pass | 13,463 | 13,660 | +1% | 1 | 1 | 0% | 2,091 | 3,044 | +46% | 0 | 0 | — |
case-18 | pass→pass | 13,798 | 16,849 | +22% | 1 | 1 | 0% | 2,400 | 3,595 | +50% | 0 | 0 | — |
case-19 | fail→pass | 24,417 | 19,257 | -21% | 1 | 1 | 0% | 2,982 | 4,677 | +57% | 0 | 0 | — |
case-20 | fail→fail | 16,688 | 20,224 | +21% | 1 | 1 | 0% | 2,927 | 4,329 | +48% | 0 | 0 | — |
case-21 | fail→fail | 25,013 | 58,053 | +132% | 1 | 1 | 0% | 4,586 | 5,298 | +16% | 0 | 0 | — |
case-22 | pass→pass | 13,981 | 16,266 | +16% | 1 | 1 | 0% | 2,350 | 3,769 | +60% | 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 +9 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.