autonomous-loops

Automate Claude Code loop design and execution for multi-agent tasks.

2|Updated Mar 21, 2026
One-click install
npx skills add https://github.com/Throokie/claude-code-skills --skill autonomous-loops-throokie
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: autonomous-loops
Source: https://github.com/Throokie/claude-code-skills/tree/main/skills/autonomous-loops
Command: npx skills add https://github.com/Throokie/claude-code-skills --skill autonomous-loops-throokie

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Patterns and architectures for autonomous Claude Code loops — from simple sequential pipelines to RFC-driven multi-agent DAG systems.

Core Features & Use Cases

  • Defines a spectrum of loop patterns (Sequential Pipeline, NanoClaw REPL, Infinite Agentic Loop, Continuous Claude PR Loop, De-Sloppify, Ralphinho/RFC-Driven DAG) to cover common AI automation tasks.
  • Supports multi-agent orchestration, per-stage context windows, persistent state, merge queues, and eviction recovery to ensure robust, scalable automation.
  • Use cases include autonomous development pipelines, long-running tasks, spec-driven generation, and end-to-end RFC/DAG workflows in software projects.

Quick Start

Describe a small feature RFC and run a basic sequential loop to implement and test it.

Frequently Asked Questions about autonomous-loops

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

FAQPage Schema
How do I build autonomous Claude Code loops for multi-agent tasks?

Multi-agent orchestration is managed through RFC-driven DAG systems that enforce isolated context windows per stage, explicit dependency graphs, and persistent state tracking between iterations for robust pipeline execution.

What loop patterns are available for continuous AI agent automation?

Available loop patterns include Sequential Pipeline, NanoClaw REPL, Infinite Agentic Loop, Continuous Claude PR Loop, and RFC-Driven DAG, covering tasks from simple sequential generation to complex merge-based landing workflows.

Can I use RFC-driven DAG orchestration for long-running continuous PR loops?

Yes, RFC-driven DAG orchestration supports continuous PR loops by utilizing persistent state between iterations, merge queues, and eviction recovery to ensure scalable, long-running autonomous development workflows.

How do I maintain persistent state across iterations in an infinite agentic loop?

Persistent state across iterations is maintained by isolating the context window per stage and passing explicit dependencies through the DAG, ensuring each loop iteration retains necessary state without context bleed.

Do I need isolated context windows for each stage in a sequential pipeline?

Yes, isolated context windows per stage are required in sequential pipelines to prevent context contamination, ensuring each autonomous loop stage executes independently before merging via the landing workflow.

What distinguishes RFC-driven DAG orchestration from a simple sequential pipeline for autonomous development?

RFC-driven DAG orchestration supports complex multi-agent dependencies and parallel execution paths, whereas a simple sequential pipeline processes tasks linearly without explicit dependency graphs or advanced merge queue workflows.