Gentleman-Programming

Gentleman-Programming / agent-teams-lite

Public

Spec-Driven Development with AI Sub-Agents. An orchestrator + 9 specialized sub-agents for structured feature development. Zero dependencies. Pure Markdown. Works with Claude Code, OpenCode, Cursor, and more.

241
25
100% credibility
Found Feb 21, 2026 at 23 stars 10x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Shell
AI Summary

A set of plain-text guides that teach AI coding assistants to work as coordinated teams, breaking down and building software features step-by-step with clear plans and reviews.

How It Works

1
🔍 Discover a better way

You find Agent Teams Lite while searching for help to make AI assistants build complex app features without chaos or mistakes.

2
📥 Grab and set up

Download the files and run the simple setup helper that puts everything in the right spot for your AI coding tool.

3
🔗 Link to your AI helper

Follow the easy guide to connect these team instructions to your favorite AI coding assistant.

4
🚀 Start building a feature

Open your project, chat with your AI, and say what new feature you want—like adding a dark mode toggle.

5
✅ Review and approve steps

Your AI shows simple summaries of the plan, designs, and tasks at each stage, and you give the thumbs up to continue.

6
⚡ Watch it build and check

The AI team handles exploring, planning, coding, testing, and finishing up, keeping everything organized and on track.

🎉 Feature ready to shine

Your new feature is perfectly built, verified, and saved, making your app better without the usual headaches.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 23 to 241 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 agent-teams-lite?

Agent-teams-lite brings spec-driven development to AI coding with a lightweight orchestrator and nine specialized sub-agents that handle feature building step-by-step: from codebase exploration and proposals to specs, design, tasks, implementation, verification, and archiving. It solves AI assistants' issues like context overload and unstructured outputs by delegating focused tasks to fresh-context sub-agents, using slash commands like /sdd:new and /sdd:apply in tools such as Claude Code, OpenCode, or Cursor. Zero dependencies, pure Markdown skills in a Shell-based installer, with pluggable persistence via engram or openspec folders.

Why is it gaining traction?

Unlike inline tools like OpenSpec, it uses true sub-agent delegation for parallel phases and clean context, reducing hallucinations in spec-driven AI workflows on GitHub. Developers hook into its DAG-based flow for approval gates and structured artifacts, making complex features predictable without heavy setup. Slash commands and tool-agnostic design let you start spec-driven coding GitHub-style instantly.

Who should use this?

Solo full-stack devs or AI-augmented teams using Claude or Cursor for non-trivial features like CSV exports or dark mode, where you need spec-driven development AI to enforce reviews before code lands. Ideal for those experimenting with agent teams lite in spec-driven dev GitHub repos, avoiding chat-history chaos.

Verdict

Promising for spec-driven development Claude fans, with solid docs and examples, but at 11 stars and 1.0% credibility, it's early-stage—test on a side project before production. Worth cloning if you're into agent-driven spec workflows.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.