andylizf

andylizf / nonstop

Public

Autonomous work mode for Claude Code. Pre-flight risk assessment, blocker decision framework, session-scoped stop hook.

39
3
100% credibility
Found Mar 29, 2026 at 39 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Shell
AI Summary

Nonstop is a simple add-on for an AI coding assistant that keeps it working autonomously overnight on user-approved tasks by anticipating issues and deciding how to proceed.

How It Works

1
🔍 Discover Nonstop

You hear about a handy tool that lets your smart coding helper work all night on big projects without getting stuck on little problems.

2
📥 Set it up quickly

Tell your coding helper to grab and install it, or use a simple one-line download to get ready in seconds.

3
🚀 Hand off a big task

Give your helper a long to-do list for your coding project that would normally take hours.

4
✅ Turn on nonstop mode

Say the magic words to start, review the step-by-step plan and approve any risky moves, then confirm it's good to go.

5
😴 Rest easy overnight

Walk away or go to sleep while your helper thinks ahead, solves issues, and keeps pushing forward without waiting for you.

🌅 Wake up to finished work

Come back to see most or all of your project done, with notes on what worked around problems or needs a quick check.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 39 to 39 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 nonstop?

Nonstop turns Claude Code into an autonomous worker for overnight coding sessions. Type `/nonstop` to trigger pre-flight risk assessment, approve dangerous ops like deletes or force pushes, and set blocker-handling rules—then it grinds through tasks without halting on errors, workarounds, or simulations. Built in Shell, it solves AI stalling on ambiguities or permissions during long autonomous workflows, letting you wake up to progress.

Why is it gaining traction?

Unlike basic stop tokens in tools like taskmaster or bulky orchestrators, nonstop's mental simulation and decision framework predict issues upfront and route around blockers intelligently. Developers try it for the seamless autonomous coder vibe on GitHub, delegating builds or tests to background agents without babysitting. It's dead simple to install via curl or Claude plugins, hooking into stop protocols for true hands-off operation.

Who should use this?

Claude Code users on multi-hour refactors, like migrating auth systems or running test suites overnight. Backend devs building autonomous workflows who hate permission prompts derailing sessions. Solo hackers experimenting with autonomous work environments for non-urgent features.

Verdict

Grab it if you're deep in Claude Code—39 stars and 1.0% credibility signal early days with minimal tests, but crisp docs and MIT license make it a safe playground. Test on throwaway tasks first; scales well for autonomous coding ui boosts.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.