Legal document review has not changed much in 30 years. A lawyer reads a contract, identifies the relevant clauses, and flags anything non-standard. The process is slow, expensive, and scales linearly with headcount.

Crucible changes the economics of this workflow without changing its output. The goal is not to replace the lawyer. It is to remove the hours of reading that precede the judgment.

What the Workflow Looks Like

A typical contract review integration using Crucible has three stages.

Stage 1: Extraction. The contract is sent to the API in a single request. A structured JSON schema defines the fields to extract: parties, effective date, payment terms, termination rights, limitation of liability, governing law, and any domain-specific clauses the team cares about.

Stage 2: Flagging. A second request compares the extracted values against a defined standard. Non-standard terms, missing clauses, and values outside acceptable ranges are flagged with references to the source text.

Stage 3: Summary. A third request produces a short-form summary formatted for the reviewing attorney. The summary includes the flagged items, the relevant contract language, and the reasoning behind each flag.

Where It Works Best

This approach works best on commercial agreements with relatively standard structures: MSAs, NDAs, SaaS agreements, and vendor contracts. It works less well on highly bespoke agreements where the structure varies significantly from document to document. In those cases, the extraction schema requires more customization but the approach remains viable.