ralph
Installation
Summary
Specification-first AI development with Socratic interviewing and autonomous evolutionary loops until verified completion.
- Exposes hidden assumptions through guided questioning, crystallizing vague requirements into unambiguous specs (Ambiguity ≤ 0.2) before any code execution
- Ralph mode runs persistent loops—Interview → Seed → Execute → Evaluate → Evolve—until acceptance criteria pass or max iterations reached, with checkpoint-based resumption after interruption
- Three-stage verification gates: mechanical (lint, tests, coverage), semantic (acceptance criteria alignment), and optional multi-model consensus for drift detection
- Includes five lateral-thinking personas (Simplifier, Hacker, Contrarian, Researcher, Architect) to break through stagnation when repeated attempts fail
- Supports Claude Code (native hooks), Codex CLI (bash loop), Gemini CLI (AfterAgent hook), and OpenCode with unified
ooocommand syntax across all platforms
SKILL.md
ralph (Ouroboros) — Specification-First AI Development
Stop prompting. Start specifying.
"The beginning is the end, and the end is the beginning." The serpent doesn't repeat — it evolves.
When to use this skill
- Before writing any code — expose hidden assumptions with Socratic interviewing
- Long-running tasks that need autonomous iteration until verified
- Vague requirements — crystallize them into an immutable spec (Ambiguity ≤ 0.2)
- Tasks requiring guaranteed completion — loop until verification passes
- When stuck — 5 lateral thinking personas break through stagnation
- Drift detection — measure how far execution has deviated from original spec