usemozzie

usemozzie / mozzie

Public

Local-first desktop app that orchestrates AI coding agents in parallel — work items, git worktrees, dependency tracking, and review workflow in one window.

mozzie.dev ai-agents claude-code codex coding-agents developer-tools
17
2
100% credibility
Found Mar 13, 2026 at 16 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Mozzie is a local-first desktop app that orchestrates multiple AI coding agents to collaboratively build and manage software projects using isolated git worktrees and a unified review workflow.

How It Works

1
📥 Download the app

Get the free desktop program that helps AI helpers build your code projects while you relax.

2
🔗 Link your smart helpers

Connect a few AI services like Claude or Gemini so they can think and create code for you.

3
📁 Add your code folders

Point it to your project directories so the helpers know where to work.

4
💭 Describe what to build

Type what you want in plain English, like 'add a login page', and press enter.

5
📋 See the smart to-do list

It breaks your idea into safe, bite-sized tasks that won't step on each other.

6
🚀 Watch them work together

Helpers tackle tasks at the same time in their own private spaces, streaming live updates.

7
✅ Review and say yes

Check the changes side-by-side, approve the good ones, and they merge neatly.

🎉 Your project is built

Everything updates automatically, ready to use, while you slept or did other things.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 16 to 17 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 mozzie?

Mozzie is a local-first desktop app built with TypeScript and Tauri that orchestrates multiple AI coding agents like Claude Code or Gemini CLI in parallel across isolated git worktrees. You describe a feature in natural language via its LLM orchestrator, which breaks it into dependency-tracked work items, assigns agents, and queues diffs for one-window review and merge. It solves the chaos of managing AI agents manually, keeping everything offline with SQLite persistence and zero cloud lock-in.

Why is it gaining traction?

Its killer hook is the seamless workflow: auto-launch blocked items on dependency completion, rejection feedback loops that teach agents, and live streaming of tool calls without tab-switching. Developers dig the git-native isolation—no merge conflicts from parallel agents—and the floating command bar for quick orchestration. In a sea of single-agent tools, mozzie's multi-agent swarm with sub-work items and stacked branches feels like a personal dev team.

Who should use this?

Solo full-stack devs or small teams prototyping features fast, especially those juggling multiple repos and tired of babysitting Cursor or Aider sessions. Backend engineers handling dependency-heavy refactors, or indie hackers building local-first github projects without vendor APIs. If you're experimenting with agents for coding tasks like bug fixes or new endpoints, this scales your machine into a workforce.

Verdict

Try mozzie if you're into agentic coding—its demo video and docs make onboarding quick despite 12 stars and 1.0% credibility signaling early days. Polish the edges on Windows support and agent reliability, but for local-first power users, it's a steal at MIT license.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.