doc-merge

Compare documents semantically and generate annotated diffs with diagnostic reports.

Updated Feb 16, 2026
One-click install
npx skills add https://github.com/zkbkb/thinkbench-skills --skill doc-merge
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: doc-merge
Source: https://github.com/zkbkb/thinkbench-skills/tree/main/skills/doc-merge
Command: npx skills add https://github.com/zkbkb/thinkbench-skills --skill doc-merge

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill streamlines the process of comparing multiple documents on the same topic, identifying their differences at a semantic level, and guiding you through merging them into a single, coherent output.

Core Features & Use Cases

  • Semantic Comparison: Goes beyond character-level diffs to analyze differences in framework, structure, arguments, and expression.
  • Difference Diagnosis: Provides a structured report detailing where and how documents diverge.
  • Interactive Resolution: Guides you through resolving conflicts step-by-step, respecting dependencies between differences.
  • Use Case: You have three AI-generated drafts of a research paper. Use this Skill to compare them, understand their unique contributions and contradictions, and merge them into the best possible final version.

Quick Start

Use the doc-merge skill to compare the three provided draft documents and help me merge them into a single, cohesive document.

Frequently Asked Questions about doc-merge

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

FAQPage Schema
How do I merge multiple drafts of the same document into one?

Semantic document comparison analyzes differences across meta, structural, argument, and expression levels rather than just character changes, generating an annotated diff and a diagnostic report to explain how documents diverge.

What's the best way to consolidate conflicting AI-generated research drafts?

The best way to consolidate conflicting drafts is using an interactive batched resolution workflow that respects dependencies between differences, ensuring contradictions are resolved step-by-step for a coherent final document.

Can I compare documents for structural and argument-level differences, not just text?

Yes, you can compare documents for structural and argument-level differences. The analysis diagnoses divergence across meta, structural, argument, and expression levels, going beyond basic character-level diffs.

Does document comparison work with more than two input files?

Yes, multi-document comparison works with more than two files. The Skill performs semantic comparison across multiple documents on the same topic and consolidates all inputs into a single cohesive output.

When should I not use semantic merge for document consolidation?

You should avoid semantic merge for documents on entirely different topics, as the Skill requires multiple inputs addressing the same subject to accurately diagnose differences and resolve conflicts during consolidation.