Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Anwalt oder Mandant ist in UK/US-Gerichtsverfahren und will pleadings discovery disclosure depositions privilege evidence settlement verstehen. Prüfraster Verfahrensphasen-Überblick Privilege-Prüfung Discovery-Scope Settlement-Optionen. Output Verfahrens-Erklärung Checkliste. Abgrenzung zu common-law-interpretation-precedent (Praezedenz) und common-law-governing-law-jurisdiction (Rechtswahl).
.claude/skills/thomasmoreai-common-law-litigation-discovery/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 40% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 69% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 66% | 0% |
Deutsche ZPO-Erwartungen werden von discovery, disclosure, privilege, litigation hold und depositions getrennt.
Common-Law-Kompass arbeitet freundlich, präzise und verzeihend. Der Stil darf leicht sein, aber nie auf Kosten der juristischen Trennschärfe.
Before proceeding, clarify:
Addressee: Internal litigation team Tone: Process-focused; privilege risks highlighted
DISCLOSURE / DISCOVERY REVIEW
Date: [DATE] — Matter: [DESCRIPTION]
Jurisdiction: [England/Wales CPR / US FRCP]
1. SCOPE OF DISCLOSURE
UK Standard (CPR 31.6): [relevant documents defined as: ...]
US Scope (FRCP 26(b)): [proportional; relevance to claims/defenses]
Estimated document volume: [X documents / Y GB data]
2. PRIVILEGE REVIEW
Legal advice privilege: [YES — comms with [solicitor/attorney] re legal advice]
Litigation privilege: [YES — dominant purpose litigation / NO]
In-house counsel: [covered UK Three Rivers / covered US Upjohn]
Waiver risk: [document shared with [PARTY] — waived?]
3. E-DISCOVERY / ESI
Litigation hold issued: [YES — [DATE] / NOT YET]
Custodians: [NAME, ROLE]
Data formats: [email / Slack / WhatsApp / cloud]
Cost sharing (Zubulake): [proportionality request filed / not filed]
4. THIRD PARTY DISCLOSURE
Norwich Pharmacal order applicable?: [YES / NO]
FRCP 45 subpoena?: [YES / NO]
5. NEXT STEPS
[ ] Issue litigation hold to [CUSTODIANS]
[ ] Privilege log by [DATE]
[ ] Disclosure list CPR / FRCP 26 disclosures by [DATE]| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 25,875 | 22,953 | -11% | 1 | 1 | 0% | 4,737 | 5,242 | +11% | 0 | 0 | — |
case-02 | fail→pass | 22,829 | 20,581 | -10% | 1 | 1 | 0% | 4,353 | 4,790 | +10% | 0 | 0 | — |
case-03 | fail→pass | 21,482 | 19,808 | -8% | 1 | 1 | 0% | 3,775 | 5,024 | +33% | 0 | 0 | — |
case-04 | fail→pass | 17,820 | 21,107 | +18% | 1 | 1 | 0% | 3,513 | 4,923 | +40% | 0 | 0 | — |
case-05 | pass→pass | 20,860 | 18,197 | -13% | 1 | 1 | 0% | 3,841 | 4,902 | +28% | 0 | 0 | — |
case-06 | pass→pass | 21,561 | 16,959 | -21% | 1 | 1 | 0% | 4,093 | 4,689 | +15% | 0 | 0 | — |
case-07 | pass→pass | 15,231 | 18,117 | +19% | 1 | 1 | 0% | 2,733 | 4,513 | +65% | 0 | 0 | — |
case-08 | pass→pass | 11,182 | 15,022 | +34% | 1 | 1 | 0% | 2,089 | 4,301 | +106% | 0 | 0 | — |
case-09 | pass→pass | 12,866 | 14,681 | +14% | 1 | 1 | 0% | 2,383 | 4,063 | +70% | 0 | 0 | — |
case-10 | pass→pass | 14,382 | 18,204 | +27% | 1 | 1 | 0% | 2,682 | 4,699 | +75% | 0 | 0 | — |
case-11 | fail→pass | 13,394 | 14,941 | +12% | 1 | 1 | 0% | 2,493 | 4,220 | +69% | 0 | 0 | — |
case-12 | pass→pass | 20,771 | 15,781 | -24% | 1 | 1 | 0% | 3,554 | 4,243 | +19% | 0 | 0 | — |
case-13 | fail→pass | 12,824 | 15,335 | +20% | 1 | 1 | 0% | 2,365 | 3,929 | +66% | 0 | 0 | — |
case-14 | pass→pass | 13,628 | 16,994 | +25% | 1 | 1 | 0% | 2,643 | 4,649 | +76% | 0 | 0 | — |
case-15 | pass→pass | 12,082 | 13,730 | +14% | 1 | 1 | 0% | 2,172 | 3,749 | +73% | 0 | 0 | — |
case-16 | pass→pass | 14,043 | 14,305 | +2% | 1 | 1 | 0% | 2,506 | 3,913 | +56% | 0 | 0 | — |
case-17 | fail→fail | 17,161 | 15,741 | -8% | 1 | 1 | 0% | 3,014 | 4,234 | +40% | 0 | 0 | — |
case-18 | pass→pass | 10,240 | 12,792 | +25% | 1 | 1 | 0% | 2,012 | 3,781 | +88% | 0 | 0 | — |
case-19 | fail→fail | 16,894 | 25,197 | +49% | 1 | 1 | 0% | 3,099 | 5,542 | +79% | 0 | 0 | — |
case-20 | pass→pass | 13,972 | 18,258 | +31% | 1 | 1 | 0% | 3,037 | 4,923 | +62% | 0 | 0 | — |
case-21 | fail→fail | 14,862 | 15,336 | +3% | 1 | 1 | 0% | 2,783 | 4,291 | +54% | 0 | 0 | — |
case-22 | fail→fail | 21,001 | 20,663 | -2% | 1 | 1 | 0% | 4,015 | 5,355 | +33% | 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 +23 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.