ar

Orchestrate multiple AI review agents to critique design documents in parallel.

6|2|Updated Jan 30, 2026
One-click install
npx skills add https://github.com/skillsets-cc/main --skill ar
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ar
Source: https://github.com/skillsets-cc/main/tree/main/skillsets/%40supercollectible/Valence/content/.claude/skills/ar
Command: npx skills add https://github.com/skillsets-cc/main --skill ar

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Adversarial design reviews are orchestrated across multiple AI reviewers to identify gaps, risks, and improvement opportunities in design documents, accelerating robust decision-making.

Core Features & Use Cases

  • Orchestrates ar-o, ar-k, and ar-glm5 agents to critique design docs in parallel.
  • Synthesizes findings into a deduplicated, prioritized set of actionable recommendations.
  • Use Case: Validate architecture, feasibility, and risk early in product and engineering cycles.

Quick Start

Initialize the ar skill with your design document and run the adversarial review workflow.

Frequently Asked Questions about ar

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

FAQPage Schema
How do I run an adversarial design review on a design doc?

To run an adversarial design review, you initialize the workflow with your design document, which orchestrates multiple AI agents to critique architecture, feasibility, and risk in parallel.

What is an adversarial review workflow for product and engineering teams?

An adversarial review workflow is a process that coordinates multiple AI review agents to run parallel critiques on design documents, producing deduplicated findings and a final risk report.

Can I use multiple AI agents to validate architecture and feasibility?

Yes, you can validate architecture and feasibility by orchestrating multiple AI review agents to critique your design docs in parallel and synthesize prioritized recommendations.

How do cross-agent workflows deduplicate risk analysis findings?

Cross-agent workflows deduplicate risk analysis findings by synthesizing the parallel critiques from multiple AI agents into a single, prioritized set of actionable recommendations.

Do I need specific AI agents to orchestrate task workflows for design validation?

Yes, design validation requires orchestrating specific AI agents (ar-o, ar-k, and ar-glm5) with bypass permissions to execute the structured task workflow and produce a final review report.