Agent skill
ralph-init
Initialize a PRD (Product Requirements Document) for structured ralph-loop execution
Install this agent skill to your Project
npx add-skill https://github.com/Yeachan-Heo/oh-my-codex/tree/main/skills/ralph-init
SKILL.md
Ralph Init
Initialize a PRD (Product Requirements Document) for structured ralph-loop execution. Creates a structured requirements document that Ralph can use for goal-driven iteration.
Usage
/ralph-init "project or feature description"
Behavior
- Gather requirements via interactive interview or from the provided description
- Create PRD at
.omx/plans/prd-{slug}.mdwith:- Problem statement
- Goals and non-goals
- Acceptance criteria (testable)
- Technical constraints
- Implementation phases
- Link to Ralph so that
/ralphcan use the PRD as its completion criteria - Initialize/ensure canonical progress ledger at
.omx/state/{scope}/ralph-progress.json(session scope if active session exists)
Canonical source contract
- Canonical PRD source of truth is
.omx/plans/prd-{slug}.md. - Ralph progress source of truth is
.omx/state/{scope}/ralph-progress.json(session scope when available). - Legacy
.omx/prd.json/.omx/progress.txtinputs are compatibility-only and migrate one-way into canonical artifacts.
Output
A structured PRD file saved to .omx/plans/ that serves as the definition of done for Ralph execution.
Next Steps
After creating the PRD, start execution with:
/ralph "implement the PRD"
Ralph will iterate until all acceptance criteria in the PRD are met and architect-verified.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
configure-notifications
Configure OMX notifications - unified entry point for all platforms
ultraqa
QA cycling workflow - test, verify, fix, repeat until goal met
build-fix
Fix build and TypeScript errors with minimal changes
ask-claude
Ask Claude via local CLI and capture a reusable artifact
tdd
Test-Driven Development enforcement skill - write tests first, always
analyze
Run deep investigation of architecture, bugs, performance issues, or dependencies and return structured findings with file:line evidence. Use when a user says 'analyze', 'investigate', 'why does', 'what's causing', or needs root cause analysis before making changes. Routes to architect agent or Codex MCP for thorough cross-file reasoning.
Didn't find tool you were looking for?