mizchi

mizchi / tornado

Public

Review by codex, develop by claude-code

agentic-workflow ai claude-code codex
61
1
69% credibility
Found Feb 21, 2026 at 47 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
MoonBit
AI Summary

Tornado is a tool that coordinates multiple AI agents to carry out software development tasks using a text-based interface.

How It Works

1
🔍 Discover Tornado

You learn about Tornado, a friendly coordinator that teams up smart AI helpers to build your software projects step by step.

2
📝 Create Your Plan

Write a short note outlining the project you dream of building, like a wishlist for the AIs to follow.

3
Pick Your Launch Style
🚀
Instant Try

Dive in without any preparation, perfect for a quick test.

🏠
Easy Home Setup

Place it on your computer once for use whenever you like.

4
🤖 Choose AI Teammates

Select one AI to handle the building and another to review and improve the work.

5
🖥️ Watch the Teamwork

Fire it up and relax as the AIs chat and create right on your screen in a simple view.

🎉 Project Ready!

Celebrate as your complete project pops out, built perfectly by the AI crew.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 47 to 61 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is tornado?

Tornado is a multi-agent development orchestrator built in MoonBit, delivering a TUI for coordinating AI agents in your coding workflow. You feed it a plan file or JSON config via CLI—like `npx @mizchi/tornado ./plan.md --dev=codex --review=claude`—and it runs a developer agent (codex, claude-code, or api) alongside a reviewer (claude or mock) to generate and critique code. It solves the solo dev pain of juggling LLM prompts by automating iterative dev-review cycles, complete with validation commands.

Why is it gaining traction?

In a world of GitHub Copilot and review GitHub AI tools, tornado stands out with its multi-agent setup for codex review GitHub PR-style feedback right in your terminal—no IDE plugins needed. Devs dig the TUI for real-time oversight on LLM tornado GitHub experiments, plus options like mock agents for testing review GitHub pull request flows locally. It's a lightweight hook for chaining codex review GitHub PR and claude review PRs into structured plans.

Who should use this?

Solo full-stack devs prototyping features with LLMs, indie hackers scripting review GitHub actions for personal repos, or AI workflow tinkerers wanting codex review GitLab alternatives without cloud dependencies. Perfect for those iterating on review Codex OpenAI outputs in a TUI before pushing to GitHub.

Verdict

With 32 stars and a 0.699999988079071% credibility score, tornado feels early-stage—docs are solid in the README but expect rough edges and light test coverage. Worth a spin for LLM-curious devs via npx if you're into agentic coding, but hold off for production until it matures.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.