ralph-loop

Iterate BMAD story backlogs via subprocesses until completion or blockage.

1|Updated Jan 13, 2026
One-click install
npx skills add https://github.com/jongensutrecht/demo-ai-stack --skill ralph-loop-jongensutrecht
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ralph-loop
Source: https://github.com/jongensutrecht/demo-ai-stack/tree/main/.claude/skills/ralph-loop
Command: npx skills add https://github.com/jongensutrecht/demo-ai-stack --skill ralph-loop-jongensutrecht

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Ralph Loop automates the final-mile execution of BMAD story backlogs, so you can keep progressing until every story is finished, blocked, or safely stopped by a circuit breaker instead of manually babysitting each step.

Core Features & Use Cases

  • Backlog-driven iteration: Reads the BACKLOG.md for a process and selects the next story to work on based on current status.
  • Subprocess orchestration: Forces each story to run in a fresh Claude subprocess so the main context stays clean and focused.
  • Completion safeguards: Tracks attempts, updates story status, and stops only when all stories are resolved or when progress stalls.
  • Use case: Run it after launching a BMAD workflow to automatically continue through remaining stories without losing traceability or context.

Quick Start

Ask the assistant to run Ralph Loop against the process backlog and continue iterating until every story is marked DONE or BLOCKED.

Frequently Asked Questions about ralph-loop

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate BMAD story backlog execution in Claude Code?

Automating BMAD story backlog execution requires iterating through open items in a BACKLOG.md file until completion or blockage. Ralph Loop drives this by selecting the next story, running it in an isolated subprocess, and updating statuses until all items are resolved or a circuit breaker triggers.

What is subprocess orchestration for story-driven development workflows?

Subprocess orchestration for story-driven development forces each story to execute in a fresh Claude subprocess. This isolation keeps the main context clean while tracking attempts and updating story status during backlog iteration.

Do I need a BACKLOG.md file to run workflow automation for story tracking?

Yes, a BACKLOG.md file is required as the source for workflow automation. The automation reads this file to select the next story based on its current status, ensuring traceability and continued progress without manual intervention.

How does a circuit breaker safeguard work in backlog management?

A circuit breaker in backlog management stops automated iteration when progress stalls. It tracks attempts and halts the workflow safely if stories cannot be resolved, preventing infinite loops and ensuring the process stops only when all items are done or blocked.

When should I not use automated subprocess orchestration for backlog iteration?

You should avoid automated subprocess orchestration when your backlog items require shared context between stories or lack a clear status tracking mechanism. The process relies on isolated execution and a structured BACKLOG.md to determine when to stop or continue.