skills.rest
chanhee-kang
chanhee-kangCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

ralph

Automate iterative task refinement and verification with tiered architectural checks.

2|Updated Feb 3, 2026
Search Tags:#workflow#multi-agent#iteration#prd#ralph#ultrawork
One-click install
npx skills add https://github.com/chanhee-kang/oh-my-gemini --skill ralph-chanhee-kang
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ralph
Source: https://github.com/chanhee-kang/oh-my-gemini/tree/main/skills/ralph
Command: npx skills add https://github.com/chanhee-kang/oh-my-gemini --skill ralph-chanhee-kang

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates iterative task refinement and verification with architectural oversight.

Core Features & Use Cases

  • Self-referential loop with Ralph and Ultrawork for parallel task execution and persistence
  • Optional PRD mode to generate Product Requirements Document before execution
  • Architecture-backed verification using tiered agents to ensure correct, complete results
  • Progress tracking and clean state cleanup after verification

Quick Start

Provide your task to the Ralph Skill to begin an iterative refinement loop. Optionally enable PRD mode with --prd to generate a PRD document and progress log. The Ralph loop will orchestrate parallel tasks and perform architect verification before signaling completion.

Frequently Asked Questions about ralph

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

FAQPage Schema
How does iterative task refinement work for complex multi-agent workflows?▼

Iterative task refinement automates progressive task execution using self-referential loops with parallel agents, applying architectural verification to ensure correct and complete results across complex projects.

How do I generate a Product Requirements Document before starting task execution?▼

You can generate a Product Requirements Document (PRD) by enabling PRD mode with the --prd flag, which creates a formal PRD document and progress log before orchestrating parallel task execution.

Can I use parallel task execution for projects that require progressive refinement?▼

Yes, parallel task execution supports complex projects requiring progressive refinement by orchestrating multiple agents simultaneously through a self-referential loop with persistence and tracking.

What is the best way to verify complex project tasks across multiple agents?▼

Architecture-backed verification uses tiered agents to perform formal checks on task results, ensuring correctness and completeness while tracking progress and cleaning state after verification completes.

Does multi-agent workflow execution require any external dependencies?▼

Multi-agent workflow execution operates without external dependencies, relying entirely on internal architectural oversight, tiered verification agents, and self-referential loops for persistent task completion.

Related Skills

Explore More Product & Management Skills→
The-RabakThe-Rabak

ralph-loop

Automate iterative prompt loops until a completion promise or max-iterations limit.

Community
Advanced
okwindsokwinds
50

prd-to-engineering-spec

Transform a PRD into an implementation-ready engineering specification with architecture, data modeling, API design, security, deployment, and testing.

Community
Advanced
koasis89koasis89

ralph

Execute complex tasks iteratively with parallel processing and architect verification.

Community
Advanced
suhwansuhwan

ralph

Automate iterative task loops with architect verification until completion.

Community
Advanced