review-board

Orchestrate multi-LLM document reviews via external AI CLIs and synthesize findings.

Updated Feb 14, 2026

Search Tags:#code review #feedback #technical writing #ai #llm #document review

Ready to run

npx skills add https://github.com/abhichandra21/dotclaude --skill review-board

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: review-board
Source: https://github.com/abhichandra21/dotclaude/tree/main/skills/review-board
Command: npx skills add https://github.com/abhichandra21/dotclaude --skill review-board

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill automates the process of getting in-depth technical reviews for documents from multiple AI models, synthesizing their feedback into actionable insights.

Core Features & Use Cases

Multi-LLM Review: Leverages external LLM CLIs (like ChatGPT, Gemini, GitHub Copilot) to provide independent reviews.
Feedback Synthesis: Consolidates diverse feedback into structured categories like Critical Issues, Implementation Gaps, and Risk Factors.
Iterative Refinement: Supports multi-round reviews, including rebuttals and consensus building, to drive towards a final, well-vetted document.
Use Case: Submit your architecture design document to the review board to get feedback on technical correctness, failure modes, security, and operational readiness from multiple AI perspectives before implementation.

Quick Start

Use the review-board skill to get an initial review of the attached document 'design-doc.md'.

Frequently Asked Questions about review-board

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

FAQPage Schema

How do I get multi-LLM feedback on a technical document?▼

Multi-LLM document review is automated by sending technical documents to external AI CLIs for independent feedback, synthesizing diverse responses into categorized, actionable insights like Critical Issues and Implementation Gaps.

Do I need specific CLI tools installed for AI document review?▼

Yes, AI document review requires specific CLI tools like 'codex', 'gemini', and 'gh copilot' to be installed and configured beforehand to enable parallel execution, robust error handling, and independent feedback generation.

How does multi-round AI review handle conflicting feedback?▼

Multi-round AI review handles conflicting feedback by supporting iterative refinement through rebuttal and consensus rounds, driving diverse LLM perspectives towards a final, well-vetted document with resolved technical discrepancies.

Can I use this to review architecture design documents for failure modes?▼

Yes, you can submit architecture design documents to the review board to get feedback on technical correctness, failure modes, security, and operational readiness from multiple AI perspectives before implementation.

What is the best way to categorize feedback from multiple LLMs?▼

The best way to categorize feedback from multiple LLMs is through feedback synthesis, which consolidates diverse responses into structured categories like Critical Issues, Implementation Gaps, and Risk Factors for actionable review.