adversarial-self-refine

Community

Isolated critique loop to sharpen any solution.

Authorrd162
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Iterative improvement of any solution through blind, assertive critique using isolated CRITIC and AUTHOR agents to reduce bias and converge on better outcomes.

Core Features & Use Cases

  • Isolated multi-agent loop (CRITIC + AUTHOR) for objective critique and revision.
  • Flexible modes (PARALLEL, SEQUENTIAL, INLINE) with termination signals (DEFENSE, CONVERGE, TIMEOUT).
  • Applicable to drafts, code, prompts, designs, and plans, especially when quality is critical.

Quick Start

Provide your initial solution s0 and trigger the self-refine loop to iterate critiques and revisions until convergence or timeout.

Dependency Matrix

Required Modules

None required

Components

references

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: adversarial-self-refine
Download link: https://github.com/rd162/skills/archive/main.zip#adversarial-self-refine

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.