autonomous-loops

Automate Claude Code loop design from simple pipelines to RFC-driven DAGs.

Updated Mar 1, 2026
One-click install
npx skills add https://github.com/derekhu0002/ai4pb-orchestrator --skill autonomous-loops-derekhu0002
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: autonomous-loops
Source: https://github.com/derekhu0002/ai4pb-orchestrator/tree/main/skills/autonomous-loops
Command: npx skills add https://github.com/derekhu0002/ai4pb-orchestrator --skill autonomous-loops-derekhu0002

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Patterns, architectures, and reference implementations for running Claude Code autonomously in loops. Covers everything from simple claude -p pipelines to full RFC-driven multi-agent DAG orchestration.

Core Features & Use Cases

  • Patterns and architectures include Sequential Pipeline, NanoClaw REPL, Infinite Agentic Loop, Continuous Claude PR Loop, De-Sloppify Pattern, and RFC-driven DAG orchestration.
  • Architecture supports multi-agent coordination with separate context windows, waves, and a merge-queue style landing process, enabling iterative AI-driven software development.
  • Use cases range from fully autonomous development loops to parallel agent coordination with quality gates and context persistence.

Quick Start

Choose a loop pattern (e.g., Sequential Pipeline for simple changes or Ralphinho for RFC-driven work) and start an automation run following the pattern documentation.

Frequently Asked Questions about autonomous-loops

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

FAQPage Schema
What are autonomous Claude Code loops and how do they orchestrate multi-agent pipelines?▼

Autonomous Claude Code loops orchestrate multi-agent pipelines by automating end-to-end execution across sequential pipelines, persistent REPLs, and parallel agent loops. They enforce modular patterns and separate context windows for each stage.

How do I set up an RFC-driven DAG orchestration pipeline with Claude Code?▼

Set up an RFC-driven DAG orchestration pipeline by choosing an RFC-driven loop pattern like Ralphinho. Start an automation run following the pattern documentation to coordinate multi-unit orchestration with waves and a merge-queue style landing process.

Can I run parallel agent loops with separate context windows for iterative software development?▼

Yes, you can run parallel agent loops with separate context windows for iterative software development. The architecture supports multi-agent coordination with waves, quality gates, and a merge-queue style landing process.

What is the difference between a sequential pipeline and continuous Claude PR loops?▼

A sequential pipeline processes simple changes linearly across stages, while continuous Claude PR loops enable iterative AI-driven development with persistent context and ongoing automated pull request generation.

Does Claude Code autonomous orchestration support eviction recovery for merge queues?▼

Yes, autonomous orchestration supports eviction recovery within merge queues. The architecture enforces modular patterns with a merge-queue style landing process to handle failures and maintain iterative development continuity.

When should I avoid using multi-agent loop patterns for Claude Code automation?▼

Avoid using complex multi-agent loop patterns like RFC-driven DAGs for simple changes. Use straightforward sequential pipelines to reduce overhead when your project does not require parallel agent coordination or persistent REPLs.