tabular-review

Extract schema-based contract data into spreadsheet grids with source citations.

9.1k|1.8k|Updated Apr 21, 2026
One-click install
npx skills add https://github.com/anthropics/claude-for-legal --skill tabular-review-anthropics
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tabular-review
Source: https://github.com/anthropics/claude-for-legal/tree/main/corporate-legal/skills/tabular-review
Command: npx skills add https://github.com/anthropics/claude-for-legal --skill tabular-review-anthropics

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Tabular review turns large batches of legal documents into a consistent, spreadsheet-ready set of answers with source-backed evidence in every cell, reducing review inconsistency across many documents.

Core Features & Use Cases

  • Schema-driven grid extraction: Converts a user-defined set of columns into a typed schema so every column means the same thing across every document row.
  • Verbatim evidence for every lead: Produces cells that include both the structured value and the exact quoted supporting text (plus location), enabling efficient attorney verification.
  • Explicit states for missing or uncertain content: Uses not_present, unclear, and needs_review so the output distinguishes silence from ambiguity rather than leaving blanks.
  • Batch output for diligence workflows: Generates spreadsheet deliverables (xlsx / Google Sheets plus CSV fallbacks) suitable for M&A diligence and any “review X documents for Y fields” request.

Quick Start

Use the tabular-review skill to review the documents in ./vdr/02-Contracts/ and build an M&A diligence spreadsheet output with standardized columns and source-cited cells.

Frequently Asked Questions about tabular-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract contract data into a spreadsheet for M&A diligence?

You can extract contract data into a spreadsheet for M&A diligence by using a schema-driven grid that processes batches of documents, creating one row per document and one column per requested field with verbatim source evidence.

What is the best way to standardize legal review fields across many documents?

The best way to standardize legal review fields across many documents is to build or load a typed review schema that enforces consistent column definitions, ensuring every document row answers the exact same set of requested fields.

How does evidence-cited output handle missing or ambiguous contract clauses?

Evidence-cited output handles missing or ambiguous contract clauses by applying explicit answer states like not_present, unclear, and needs_review, distinguishing silence from ambiguity rather than leaving cells blank.

Can I export batch contract review results to xlsx or CSV formats?

Yes, you can export batch contract review results to spreadsheet formats including xlsx, Google Sheets, and CSV fallbacks, delivering auditable verification workflows suitable for portfolio reviews.

Does schema-driven QA capture exact source locations for change-of-control clauses?

Yes, schema-driven QA captures exact source quotes and locations per cell for change-of-control clauses, producing structured values alongside verbatim supporting text to enable efficient attorney verification.

What are the limitations of automated contract fact extraction?

Automated contract fact extraction requires a defined typed schema and cannot interpret clauses outside the requested fields, meaning unstructured or highly unusual provisions may trigger an explicit needs_review state for manual checking.