adversarial-polish

Refine documents through multi-round adversarial review with separate critic and judge roles.

629|87|Updated May 20, 2026
One-click install
npx skills add https://github.com/JimLiu/Illustrated-Agent-Skills --skill adversarial-polish
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: adversarial-polish
Source: https://github.com/JimLiu/Illustrated-Agent-Skills/tree/main/skills/adversarial-polish
Command: npx skills add https://github.com/JimLiu/Illustrated-Agent-Skills --skill adversarial-polish

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill provides a multi-round adversarial review process to refine documents, ensuring the final draft is of the highest quality.

Core Features & Use Cases

  • Adversarial Review: Uses strawman critics, revisers, synthesizers, and anonymous judges to iteratively improve the document.
  • Document Evolution: Tracks the evolution of the document through each round, providing transparency into the changes made.
  • Use Case: Ideal for technical writing or any document where high-quality, polished content is essential.

Quick Start

Run the adversarial-polish skill on your document to initiate a multi-round review process.

Frequently Asked Questions about adversarial-polish

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

FAQPage Schema
How does multi-round adversarial review improve document quality?

An adversarial review process refines documents by using isolated roles for critics, revisers, synthesizers, and judges. It iteratively improves content through multiple rounds, with a final version chosen by an anonymous blind vote.

What is the best way to refine technical writing through iterative feedback?

Refining technical writing through iterative feedback involves strawman critics, revisers, and synthesizers who evaluate and update the document across multiple rounds. This adversarial approach ensures the final draft is highly polished.

Do I need sub-agent capabilities to run an adversarial document review?

Yes, sub-agent capabilities are required for effective isolation of roles in adversarial document review. These capabilities ensure that critics, revisers, synthesizers, and judges operate independently throughout the refinement process.

Can I track document evolution during an adversarial review process?

Yes, the adversarial review process tracks document evolution through each round. This provides full transparency into the modifications made by the revisers and synthesizers as the draft converges to its final state.

When should I use an adversarial review process instead of standard writing assistance?

An adversarial review process is preferred over standard writing assistance when high-quality, polished content is essential. It is ideal for technical writing or any document requiring rigorous, role-based refinement.