Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Wenn es um AGB-Kontrolle Klauseln in Bauträgervertragspruefer geht: zerlegt Ergebnis, Frist, Zuständigkeit, Beweislast und Gegenposition; liefert eine Fristen- und Risikoampel mit Sofortschritten.
.claude/skills/klotzkette-agb-kontrolle-klauseln/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | 72% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 59% | 0% |
Sobald der Bautraeger Klauseln vorformuliert und für mehrere Erwerber verwendet — was beim Bautraegervertrag der Regelfall ist, auch wenn ein Notar den Entwurf erstellt hat. Die notarielle Beurkundung schützt nicht vor AGB-Kontrolle.
Keine eigenständigen quellenhart verifizierten Entscheidungen zur allgemeinen AGB-Kontrolle im Megaprompt ausgewiesen. Einschlägige Spezialentscheidungen werden in den thematischen Skills (Abnahme, Tatsachenbestätigung, Leistungsänderung) behandelt.
Klauselmatrix mit Originalwortlaut, Paragraf 307 bis 309-Einordnung, erwartbarem Gegenargument, Antwort, Ampel und gewünschter Neufassung oder Streichungsanweisung. Regelfolge nach Paragraf 306 BGB ausweisen.
textAGB-Klauselmatrix (Auszug) | Originalwortlaut | Decodierung | Rechtsanker | Gegenargument | Antwort | Änderung | Ampel | |---|---|---|---|---|---|---| | [Klausel aus Vertrag] | [Risiko für Erwerber] | [Norm] | [Bautraeger-/Notarargument] | [Antwort] | [Neufassung oder Streichung] | [rot/orange/grün] | Regelfolge bei Unwirksamkeit: Nach Paragraf 306 BGB entfällt die Klausel [Paragraph/Absatz]. Der Vertrag bleibt im Übrigen wirksam. An die Stelle der unwirksamen Klausel tritt die gesetzliche Regelung nach [Norm]. Eine geltungserhaltende Reduktion zugunsten des Bauträgers findet nicht statt.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-17 | fail→pass | 19,807 | 19,724 | -0% | 1 | 1 | 0% | 2,033 | 3,490 | +72% | 0 | 0 | — |
case-16 | pass→pass | 23,372 | 20,746 | -11% | 1 | 1 | 0% | 2,756 | 3,684 | +34% | 0 | 0 | — |
case-01 | fail→pass | 33,619 | 23,930 | -29% | 1 | 1 | 0% | 4,062 | 4,447 | +9% | 0 | 0 | — |
case-02 | fail→pass | 21,575 | 23,313 | +8% | 1 | 1 | 0% | 2,854 | 3,617 | +27% | 0 | 0 | — |
case-03 | pass→fail | 29,701 | 20,776 | -30% | 1 | 1 | 0% | 3,438 | 3,752 | +9% | 0 | 0 | — |
case-04 | pass→pass | 20,619 | 21,937 | +6% | 1 | 1 | 0% | 2,584 | 3,887 | +50% | 0 | 0 | — |
case-05 | pass→pass | 21,791 | 19,650 | -10% | 1 | 1 | 0% | 2,943 | 3,652 | +24% | 0 | 0 | — |
case-06 | pass→pass | 22,501 | 19,247 | -14% | 1 | 1 | 0% | 2,384 | 3,411 | +43% | 0 | 0 | — |
case-07 | fail→pass | 26,279 | 19,469 | -26% | 1 | 1 | 0% | 2,867 | 3,631 | +27% | 0 | 0 | — |
case-08 | pass→pass | 19,307 | 19,845 | +3% | 1 | 1 | 0% | 2,242 | 3,300 | +47% | 0 | 0 | — |
case-09 | pass→pass | 23,502 | 18,917 | -20% | 1 | 1 | 0% | 2,610 | 3,607 | +38% | 0 | 0 | — |
case-10 | pass→pass | 33,924 | 19,317 | -43% | 1 | 1 | 0% | 2,098 | 3,422 | +63% | 0 | 0 | — |
case-11 | fail→pass | 21,429 | 14,819 | -31% | 1 | 1 | 0% | 2,238 | 3,549 | +59% | 0 | 0 | — |
case-12 | pass→pass | 19,239 | 19,854 | +3% | 1 | 1 | 0% | 2,214 | 3,321 | +50% | 0 | 0 | — |
case-13 | fail→pass | 21,845 | 21,431 | -2% | 1 | 1 | 0% | 2,793 | 3,431 | +23% | 0 | 0 | — |
case-14 | fail→pass | 26,539 | 20,336 | -23% | 1 | 1 | 0% | 2,877 | 3,629 | +26% | 0 | 0 | — |
case-15 | pass→pass | 26,882 | 21,987 | -18% | 1 | 1 | 0% | 3,493 | 3,909 | +12% | 0 | 0 | — |
case-18 | fail→pass | 26,398 | 22,187 | -16% | 1 | 1 | 0% | 2,870 | 3,896 | +36% | 0 | 0 | — |
case-19 | pass→pass | 21,041 | 18,413 | -12% | 1 | 1 | 0% | 2,698 | 3,357 | +24% | 0 | 0 | — |
case-20 | pass→fail | 26,896 | 30,764 | +14% | 1 | 1 | 0% | 2,890 | 4,945 | +71% | 0 | 0 | — |
case-21 | pass→pass | 21,586 | 24,858 | +15% | 1 | 1 | 0% | 4,344 | 5,274 | +21% | 0 | 0 | — |
case-22 | pass→pass | 25,141 | 27,532 | +10% | 1 | 1 | 0% | 3,210 | 4,174 | +30% | 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 +27 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 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.