skills.rest
akillness
akillnessCOMMUNITY
Explore all skills by this author →
Website𝕏@akillness0308GitHub

ralph

Enforce specification-first AI development workflows across multiple CLI platforms.

40|9|Updated Jan 4, 2026
Search Tags:#specification-first#double-diamond#ralph#llm-orchestration#ouroboros#persistent-loop
One-click install
npx skills add https://github.com/akillness/oh-my-skills --skill ralph-akillness
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ralph
Source: https://github.com/akillness/oh-my-skills/tree/main/.agent-skills/ralph
Command: npx skills add https://github.com/akillness/oh-my-skills --skill ralph-akillness

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

The ralph skill formalizes and enforces a specification-first AI development workflow, allowing teams to crystallize vague ideas into actionable specs before coding, while enabling persistent, converging loops across iterations.

Core Features & Use Cases

  • Socratic interviewing and Ontology-driven clarification guide rapid requirement crystallization.
  • Immutable seed generation and cross-platform orchestration enable end-to-end development with traceable evolution.
  • Use cases include early scoping of complex AI projects, risk reduction via rapid iteration, and continuous improvement with drift tracking.

Quick Start

Begin by running ooo interview to crystallize requirements, then run ooo seed to generate an immutable YAML spec, and finally run ooo run to execute the Double Diamond workflow.

Frequently Asked Questions about ralph

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

FAQPage Schema
What is specification-first AI development and how does it manage iterative loops?▼

Specification-first AI development crystallizes vague ideas into actionable specs before coding, using persistent converging loops to enable traceable evolution across platforms.

How do I start a specification-first workflow for an AI project?▼

Run an interview command to crystallize requirements, generate an immutable YAML seed spec, and execute the Double Diamond workflow to drive end-to-end development with continuous improvement.

Does this specification-first workflow support cross-platform LLM orchestration?▼

Yes, it specifies cross-platform orchestration for platforms like Claude Code, Codex CLI, Gemini CLI, and OpenCode, enforcing a root SKILL.md with mandatory name and description.

When do I need Socratic interviewing and ontology-driven clarification for AI specs?▼

You need Socratic interviewing and ontology-driven clarification during early scoping of complex AI projects to guide rapid requirement crystallization and reduce risk via rapid iteration.

Are there safety checks for toxic content in LLM orchestration loops?▼

Yes, the specification-first workflow includes safety checks for toxic content to ensure secure cross-platform orchestration and iterative loop execution.

Related Skills

Explore More Software Engineering Skills→
usepowershellusepowershell
9

{skill-name}

Define a Skill Unit with SKILL.md and YAML frontmatter.

Community
Basic
mnthemnthe
4

ultrawork

Enforce verification-first development with planning, evidence, and deterministic execution loops.

Community
Advanced
Q00Q00
5.4k

evolve

Orchestrate an evolutionary loop that interviews, seeds, executes, and evaluates ontologies.

Community
Advanced
heysuhasheysuhas
2

subagent-driven-development

Delegate implementation tasks to specialized subagents with two-stage review.

Community
Advanced