magi

Analyze engineering topics with parallel MAGI agents and synthesize structured judgments.

3|Updated Mar 1, 2026

Search Tags:#code-review #multi-dimensional #ai-agents #bias-detection #design-decision #magi #engineering-analysis

Ready to run

npx skills add https://github.com/ohta-rh/magi-system --skill magi-ohta-rh

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: magi
Source: https://github.com/ohta-rh/magi-system/tree/main/plugins/magi/skills/magi
Command: npx skills add https://github.com/ohta-rh/magi-system --skill magi-ohta-rh

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

MAGI provides a collaborative, multi-perspective engine for engineering decision-making by launching three specialized agents to analyze a topic in parallel and synthesizing their insights into a structured judgment.

Core Features & Use Cases

Parallel agent deliberations: three MAGI personas (Scientist, Mother, and Woman) analyze the topic on different axes.
Bias detection and synthesis: MAGI Core aggregates insights, detects sycophancy, and presents a unified verdict.
Flexible usage: suitable for architecture design, technology selection, code reviews, and design strategy.

Quick Start

Provide a topic and MAGI will launch three agent perspectives for parallel assessment and synthesize results.

Frequently Asked Questions about magi

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

FAQPage Schema

How do I get multi-perspective analysis for architecture decisions?▼

Multi-perspective analysis for architecture decisions is achieved by launching three specialized AI agents in parallel to evaluate different axes of the topic and synthesizing their insights into a unified verdict.

Can I use parallel AI agents to detect bias in code reviews?▼

Yes, you can use parallel AI agents to detect bias in code reviews by aggregating their individual assessments, detecting sycophancy, and calibrating the results into a structured engineering judgment.

How do I evaluate technology selection trade-offs using multiple perspectives?▼

Evaluating technology selection trade-offs involves providing a topic to trigger parallel agent deliberations, which then synthesize multidimensional insights to handle complex design strategy decisions.

What is the best way to analyze complex engineering design decisions?▼

The best way to analyze complex engineering design decisions is using a multi-perspective engine that applies parallel agent deliberations, bias detection, and core synthesis to manage trade-offs.

When should I not use multi-perspective agents for engineering analysis?▼

You should avoid multi-perspective agents for engineering analysis when a topic requires a single definitive technical answer rather than a synthesized judgment handling complex trade-offs across multiple dimensions.