s3d1K0

s3d1K0 / ralph-bmarge

Public

"Ralph, be Marge\!" - Autonomous BMAD sprint runner

22
5
69% credibility
Found Feb 01, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Shell
AI Summary

Ralph-bmarge is an automation script that monitors sprint status files and autonomously executes AI-assisted development and review workflows for the BMAD method until all stories are complete.

How It Works

1
🔍 Discover the sprint helper

You hear about Ralph-bmarge, a friendly tool that takes over your development sprints so you can relax.

2
📋 Prepare your project plan

Make sure your project has a sprint plan file listing all the stories ready to work on.

3
📥 Add the helper to your project

Download the simple script into your project folder and make it ready to run.

4
🚀 Start the magic

Run the script once, and it begins watching your plan and handling stories on its own.

5
🤖 Watch it work automatically

It picks the next story, develops it, reviews it, and updates your plan until everything is finished.

6
🔔 Get updates and reports

Receive messages about progress, costs, and times, plus a full summary report at the end.

🎉 Sprint complete!

Your entire sprint is done with detailed stats, so you can enjoy the results without lifting a finger.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 22 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 ralph-bmarge?

Ralph-bmarge is a shell script that automates BMAD Method sprints using Claude Code CLI, evolving the original github ralph claude code autonomous dev loop into an organized "Marge" that handles stories from ready-for-dev to done. It scans sprint-status.yaml for pending tasks, runs workflows like /bmad:bmm:workflows:dev-story and code-review in session-managed Claude sessions, and loops until your sprint finishes—tracking costs, time, and sending Slack/Discord webhooks. Fire it up with ./ralph-bmarge.sh --webhook URL for hands-off progress while you grab coffee.

Why is it gaining traction?

Unlike generic github copilot ralph loops, it uses file-based YAML watching for reliable status detection, skips Claude output parsing pitfalls, and adds an AI fixer agent to unblock stuck stories automatically. Devs dig the real-time stats, markdown sprint reports, --dry-run validation, and sound alerts—making autonomous BMAD runs predictable without babysitting. At 20 stars, it's niche but hooks Claude-powered teams tired of manual sprint drudgery.

Who should use this?

BMAD Method users running Claude Code CLI on macOS/Linux, especially solo full-stack devs or small teams automating monorepo sprints from planning to review. Ideal for AI-assisted workflows where stories pile up in YAML—frontend/backend folks chaining dev-story to code-review without context loss via --resume sessions.

Verdict

Grab it if you're deep in BMAD and Claude; the docs and CLI flags like --validate and --max-stale make setup straightforward despite 20 stars and a 0.699999988079071% credibility score signaling early maturity. Test with --dry-run first—solid for autonomous sprints, but watch for edge cases in untested Linux/Windows setups.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.