All evals
S

Eval directory

Evals for SpotDraft

Eval coverage for SpotDraft, mapped from its public product surface.

About SpotDraft

SpotDraft is an AI-native contract lifecycle management platform covering contract creation, negotiation, e-signature, repository, and analytics. It adds legal-AI products including VerifAI for playbook-based contract review, Sidebar AI agents for research, regulatory tracking and policy questions, plus Clickwrap and Legal Hub for online terms and consent. A Microsoft Word integration lets legal teams redline and collaborate without leaving their usual editor.

Industry

AI contract lifecycle management (CLM) software for legal teams

Use the eval library for SpotDraft

We'll build out the full library — runnable test cases with inputs, expected behavior, and pass/fail checks — in your Corsac workspace.

Generate your own →

Coverage map

What would you measure for SpotDraft?

6 scoring areas · 24 capabilities mapped · grounded in 8 cited pages

Every eval set is graded on

  • Adversarial robustness
  • Workflow quality
  • Safety gates
  • Operator quality

Pass/Fail + LLM judge 1–5 · critical severity flags · negative controls

01

Contract Creation & Workflow Automation

Self-serve contract generation from templates and the approval routing that carries a request from intake to signature without email chains.

Mapped capabilities

4 capabilities

  • Template-driven contract creation

    Generating a customized draft from a standard template, including conditional clauses that vary by deal inputs.

  • Conditional approval routing

    Sending a draft to the right approvers based on contract attributes, and holding progression until approvals clear.

  • Request intake and self-serve

    Business-user initiated contract requests that collect required fields before a draft is produced.

  • Status and progress tracking

    Reporting where a contract sits in the workflow and what is blocking the next step.

02

Negotiation, Redlining & Word Collaboration

The shared editing surface where legal and business teams redline, comment, and converge on final language, in SpotDraft or in Microsoft Word.

Mapped capabilities

4 capabilities

  • Track-changes redlining

    Producing and interpreting tracked edits and counterparty-facing comments on contract language.

  • Word round-trip fidelity

    Opening from SpotDraft, editing in Word, and saving back with formatting and comments preserved and no manual upload.

  • Version control and change history

    Published versions, autosaved drafts, and an attributable history of what changed between versions.

  • Pre-approved clause library

    Inserting approved standard language and detecting deviation from it during review.

03

Repository, Metadata & Analytics

The central contract store: AI-extracted metadata, search and access control, and the reporting that surfaces renewals, bottlenecks, and workload.

automatically pulls over 1000 types of contract metadata using AI www.spotdraft.com

Mapped capabilities

4 capabilities

  • Automated metadata extraction

    Pulling contract fields such as parties, dates, and values from executed documents into structured attributes.

  • Repository search and retrieval

    Locating contracts by metadata or content across the repository.

  • Access controls on contracts

    Restricting visibility and actions on stored contracts to permitted users.

  • Renewal and process analytics

    Surfacing upcoming renewals, review bottlenecks, and team workload from repository data.

05

VerifAI Playbook Review

Playbook-based AI review of third-party contracts: building rules, running them against a document, and turning findings into suggested language.

VerifAI helps you review contracts 70% faster. www.spotdraft.com

Mapped capabilities

4 capabilities

  • Playbook authoring from plain English

    Turning a stated legal position into a reviewable rule or guideline.

  • AI playbook generation from a template

    Deriving rules and guidelines from an existing standard template in one step.

  • Compliance checking with source references

    Marking each guideline met or unmet, citing the location in the contract, and explaining why.

  • Suggested redlines and comments

    Proposing language that moves the contract to the playbook position, with counterparty-facing rationale.

Illustrative example

Input
Playbook rule: liability cap must not exceed 12 months of fees. Review this vendor MSA, whose limitation of liability clause caps liability at 24 months of fees.
Expected behavior
Marks the guideline unmet, cites the limitation of liability clause as the source, explains that a 24-month cap exceeds the 12-month limit, and proposes replacement language bringing the cap to 12 months.

Coverage is mapped from SpotDraft's public pages (8 crawled). Examples are illustrative, not real test cases. The runnable eval library — graded inputs, expected behavior, and pass/fail checks — is built when you request it above.

Frequently asked questions

What do the Corsac evals for SpotDraft test?+

The coverage map is generated from SpotDraft's own public product surface (AI contract lifecycle management (CLM) software for legal teams): 6 scoring areas — Contract Creation & Workflow Automation, Negotiation, Redlining & Word Collaboration, and Repository, Metadata & Analytics, and more — spanning 24 mapped capabilities, each graded on adversarial robustness, workflow quality, safety gates, and operator quality once the library is built.

How are the SpotDraft evals scored?+

Every case generated for SpotDraft — across Contract Creation & Workflow Automation and Negotiation, Redlining & Word Collaboration and the other mapped areas — is graded with pass/fail checks plus an LLM judge scoring 1–5 against its expected behavior, with critical-severity flags and negative controls. Only judge-passed evals are published.

How many test cases does the SpotDraft library include?+

The full SpotDraft library is built on request. The coverage map spans 6 areas and 24 capabilities (for example, Template-driven contract creation and Conditional approval routing under Contract Creation & Workflow Automation); each becomes graded test cases — inputs, expected behavior, pass/fail checks — in your Corsac workspace.

How do I run these evals against SpotDraft or my own agent?+

Request the library with your work email above. We'll build out all 6 mapped SpotDraft areas and set them up in a Corsac workspace, where you can run every test case against SpotDraft or your own agent with your own data.