looped-builder

Automate planning and scaffolding of large coding tasks with generated spec, plan, prompt, and run files.

1|Updated Jan 28, 2026
One-click install
npx skills add https://github.com/xaneem/looped-builder-skill --skill looped-builder
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: looped-builder
Source: https://github.com/xaneem/looped-builder-skill/tree/main
Command: npx skills add https://github.com/xaneem/looped-builder-skill --skill looped-builder

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill automates the planning and scaffolding of large, multi-step coding tasks by coordinating Claude-driven iterations. It generates the necessary files (spec.md, implementation_plan.md, prompt.md, and run.sh) and orchestrates a deterministic workflow for iterative delivery.

Core Features & Use Cases

  • Automated generation of a complete planning suite for complex tasks (spec, plan, prompt, run scripts)
  • Atomic, small steps executed in Claude loops with progress tracking
  • Reusable workflow across features, refactors, or exploratory work, enabling faster experimentation and delivery

Quick Start

Describe a large feature you want scaffolded and run the looped-builder to generate the spec and plan.

Frequently Asked Questions about looped-builder

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

FAQPage Schema
How do I automate planning and scaffolding for large coding tasks?

Automating planning and scaffolding for large coding tasks involves generating a structured suite containing spec.md, implementation_plan.md, prompt.md, and run.sh. This approach coordinates iterative execution loops to break down complex features, refactors, or multi-module explorations into atomic, trackable steps.

What is iterative task scaffolding using Claude in a bash loop?

Iterative task scaffolding using Claude in a bash loop is a mechanism orchestrating deterministic, multi-step coding workflows. It generates planning artifacts and executes small, atomic steps autonomously, tracking progress to deliver complex features or refactors iteratively without manual intervention.

How do I generate a spec and implementation plan for a complex feature?

To generate a spec and implementation plan for a complex feature, describe the large task you want scaffolded and run the looped-builder. It automatically produces spec.md, implementation_plan.md, prompt.md, and run.sh files to drive autonomous multi-step execution and iterative delivery.

Can I use looped-builder for multi-module explorations and refactors?

Yes, you can use looped-builder for multi-module explorations and refactors. It targets complex features by generating a reusable workflow with planning documents and run scripts, enabling faster experimentation and iterative delivery across exploratory work and large codebase changes.

What files are needed for autonomous multi-step execution of coding tasks?

Autonomous multi-step execution of coding tasks requires spec.md, implementation_plan.md, prompt.md, and run.sh files. These artifacts satisfy requirements for deterministic workflow orchestration, allowing Claude to execute atomic steps in a loop with progress tracking.