NTCoding

NTCoding / autonomous-claude-agent-team

Public

An autonomous claude code development team orchestrated by a hook-driven deterministic code workflow. An example repo to inspire your own workflow

14
3
100% credibility
Found Mar 03, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A Claude Code plugin coordinating AI agents as a lead, developer, and reviewer to autonomously build software features via a structured cycle ending in a draft pull request.

How It Works

1
🔍 Discover the AI Coding Team

You hear about a helpful plugin in Claude Code that lets a team of smart assistants build software features for you.

2
📦 Add the Helper Team

In your Claude Code chat, you simply add the plugin and everything sets up automatically.

3
🚀 Start Building a Feature

Tell the team leader to begin work on your idea, like adding a new button or fixing a page.

4
✅ Approve the Smart Plan

The leader shares a clear step-by-step plan with the developer and reviewer; you give it the thumbs up to start.

5
Team Works Through Steps
🔨
Code Gets Built

Watch new code appear, tests pass, and fixes happen automatically across several rounds.

❓
Help When Stuck

If the team pauses, they explain exactly what decision or info you need to keep going.

🎉 Ready Draft Pull Request

You get a complete draft change ready to review, with all checks passed and a tidy update on GitHub.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 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 autonomous-claude-agent-team?

This TypeScript plugin for Claude Code runs a team of autonomous Claude agents—lead coordinator, TDD developer, and reviewer—through a rigid feature development cycle. Kick off with `/autonomous-claude-agent-team:start-feature-team`, and it handles planning, coding, linting, git diffs, reviews, commits, CodeRabbit checks, and draft PRs tied to GitHub issues. Hooks enforce rules like no commits before review or idle agents without signals, turning flaky LLMs into a deterministic autonomous claude code workflow.

Why is it gaining traction?

Unlike loose autonomous claude agents or GitHub Copilot, it uses hooks and state files to mechanically block bad moves, respawning fresh agents per iteration to cut context noise. Devs get draft PRs with passing CI, no humans needed until final review—pure claude autonomous coding that integrates GitHub issues and CodeRabbit out of the box. It's a concrete example repo sparking autonomous coder github experiments without vague promises.

Who should use this?

Claude Code users building features solo or in small teams, especially those tired of babysitting agent drift during TDD cycles or manual PR prep. Backend devs prototyping APIs or full-stack folks iterating on TypeScript repos will see speed gains from enforced reviews and linting. Skip if you're not deep into Claude's experimental agent teams.

Verdict

Grab it as inspiration for your own claude autonomous agent setups—installs via plugin marketplace, solid docs and tests, but at 14 stars and 1.0% credibility, it's early and rigid, not production-ready. Fork and tweak for real autonomous github agents; don't bet the farm yet.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.