nicepkg

nicepkg / auto-company

Public

🤖 A fully autonomous AI company that runs 24/7. 14 AI agents (Bezos, Munger, DHH...) brainstorm ideas, write code, deploy products & make money — no human in the loop. Powered by Claude Code.

24-7 agent-teams ai-agents ai-automation ai-company
101
36
69% credibility
Found Feb 12, 2026 at 12 stars 8x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

An experimental macOS tool that runs a team of 14 specialized AI agents in an endless loop to autonomously ideate products, code, deploy, and market them using shared memory.

How It Works

1
🖥️ Discover Auto Company

You find this exciting project about a fully independent AI company that builds products on its own.

2
💻 Prepare your setup

On your Mac, connect to an AI thinking service with a subscription so the team can work.

3
📥 Download the files

Grab the ready-to-go files for the AI company.

4
🚀 Launch the first run

Start a test cycle and watch the expert AI team wake up, read their shared notes, pick tasks, and get to work.

5
Choose how to run
⏱️
Quick test

Run a few cycles to see the magic in action.

♾️
Always on

Install it to restart automatically and run 24/7.

6
📊 Check the progress

Peek at logs, shared decisions, and new projects the team creates in their workspace.

🌟 AI company alive

Your team of AI experts now runs endlessly, dreaming up ideas, building apps, launching them, and growing the business.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 12 to 101 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 auto-company?

Auto-company runs a fully autonomous AI company 24/7 on your Mac, where 14 agents modeled after experts like Bezos, Munger, and DHH brainstorm ideas, code products, deploy to Cloudflare or GitHub, and chase revenue—no humans needed. Built in Python and powered by Claude Code's agent teams, it loops through cycles of planning, execution, and consensus updates via simple Makefile commands like `make start` or `make install` for daemon mode. Think fully autonomous AI agents handling the full idea-to-money pipeline, tackling debates around whether fully autonomous ai agents should not be developed by automating it anyway.

Why is it gaining traction?

It stands out with expert personas unlocking deeper LLM reasoning, plus 30+ built-in skills for research, security scans, DevOps deploys, and QA—users see real repos and sites spin up autonomously. The hook is dead-simple monitoring (`make status`, `make monitor`) and safety redlines baked in, preventing deletes or leaks, while circuit breakers handle API limits. Developers dig the "set it and forget it" vibe over manual agent orchestration tools.

Who should use this?

Indie hackers automating side-project factories, AI researchers prototyping multi-agent swarms beyond fully autonomous cars or trains, or solo founders testing revenue loops like auto company bajaj-style product spins. Ideal for those near an auto company near me experimenting with fully autonomous manufacturing of software, not enterprise teams needing audits.

Verdict

Fun experimental sandbox (11 stars, 0.699999988079071% credibility score) with solid docs and safeguards—run `make start` foreground first to watch the magic, but expect Claude quota burn and macOS-only quirks before daemonizing. Intriguing for AI tinkerers, skip if you need production stability.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.