Use when extracting contract clauses into legal-ops systems β applies the org's clause-type taxonomy, governing-law code set, term normalization (month-to-month, initial terms), and liability-cap classification the base can't guess. Template: replace with your taxonomy.
| Test case | Without β With | Effect | Ξ tokens | Ξ turns |
|---|---|---|---|---|
| delivery-other | Β·βΒ· | β Error | 348% | 0% |
| term-cause | Β·βΒ· | β Error | 349% | 0% |
| gov-england | Β·βΒ· | β Error | 340% | 0% |
| payment-other | Β·βΒ· | β Error | 343% | 0% |
| gov-germany | Β·βΒ· | β Error | 340% | 0% |
βThe skill is a benign template schema for extracting contract clauses into structured JSON format. It contains no executable code, external network requests, or malicious instructions.β
> Template skill. The rules below are a worked example (a fictional "Acme"). Fork this skill and replace them with your organization's real rules β the agent then applies your policy instead of guessing. The measured lift demonstrates the shape works.
Extract contract clauses into JSON using Acme legal-ops codes (the model cannot guess these):
termination for convenience -> "CL-TERM-CONV"; termination for cause/breach -> "CL-TERM-CAUSE"; auto-renewal -> "CL-RENEW"; confidentiality/NDA -> "CL-CONF"; non-solicitation -> "CL-NONSOL"; governing law -> "CL-GOV"; anything else (incl. non-compete, delivery, payment) -> "CL-OTHER".
"California" -> "US-CA"); "England and Wales"/"English law" -> "GB"; Germany -> "DE"; Singapore -> "SG".
"month-to-month" -> 1; "one (1) year, renewing annually" -> 12 (the INITIAL term).
a fixed dollar amount -> "fixed"; expressly unlimited/excluded-from-cap -> "uncapped".