Eval Library
Spellbook
For SpellbookLegal AIDoc Agent

Draft Clause And Document Generation In Word

Generative AI for transactional lawyers in Microsoft Word — contract drafting, review, redlining, and the agentic Spellbook Associate workflow · Spellbook

63 graded scenarios covering edge cases, failure modes, and quality checks.

About Spellbook

Spellbook is an AI platform serving legal professionals, helping law firms and legal departments automate research, drafting, and review workflows with greater accuracy and speed than manual processes.

Employees

50–500

Industry

Legal AI

Headquarters

United States

Sample tests· showing 3 of 63

Pass/fail and graded 1–5 by an LLM judge (passes at ≥ 4), depending on the test.

#InputExpected behaviorCheck
01

A Share Purchase Agreement is open in Word. The Definitions section defines 'Purchaser' as Widget Acquisition Co. and 'Target' as Acme Manufacturing Ltd. No 'Company', 'Buyer', or 'Seller' defined terms exist in the document. The…

The generated clause uses 'Purchaser' and 'Target' consistently throughout — no generic substitutes ('Company', 'Buyer', 'Seller') appear where a defined term is applicable. The basket reads '£50,000' (sterling, not dollars). The survival period reads '12 months'. The clause is inserted at the curs…

Pass / FailWorkflowcritical
02

A Word document based on a firm template has locked content controls marking unfilled placeholders — e.g., the text 'shall be governed by [GOVERNING LAW PLACEHOLDER]' in Section 18, where '[GOVERNING LAW PLACEHOLDER]' is a locked…

The agent detects via Office.js that the current selection anchor is inside a locked content control. It refuses to insert any text and surfaces a clear, actionable error to the user — e.g., 'Cannot insert here: cursor is inside a protected field. Click into a blank paragraph between sections and t…

Pass / FailTool usecriticalneg. control
03

A SaaS Master Services Agreement is open in Word with Track Changes toggled on (revision marks visible in the document). The firm's mandatory policy requires all AI-generated content to appear as tracked insertions so a reviewing…

Before insertion, the agent reads Word's Track Changes state via Office.js (ChangeTrackingMode or equivalent). It inserts the generated clause as a tracked insertion — the text appears as red underlined (or document's configured revision color) attributed to the current logged-in user. A reviewing …

Pass / FailPolicycritical

Rubric criteria

  • Spellbook
  • Legal
  • Agentic
  • Generated

Recommended for

Generative AI for transactional lawyers in Microsoft Word — contract drafting, review, redlining, and the agentic Spellbook Associate workflowSpellbook customers

Works with

Related evals

Run this eval in your workspace

Connect your data, configure thresholds, and review results with your team.