ralph

Monitor task completion with PRD-driven execution and reviewer verification.

Updated May 30, 2026
One-click install
npx skills add https://github.com/kwt00/oh-my-grok --skill ralph-kwt00
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ralph
Source: https://github.com/kwt00/oh-my-grok/tree/main/skills/ralph
Command: npx skills add https://github.com/kwt00/oh-my-grok --skill ralph-kwt00

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Ralph helps ensure comprehensive task completion with verification, ideal for projects requiring thoroughness and review.

Core Features & Use Cases

  • Task Persistence: Continuously works on tasks until all user stories pass and are verified.
  • Parallel Execution: Concurrently executes tasks for efficiency.
  • Verification Loop: Requires verification before completion to ensure quality.
  • Use Case: Ideal for complex tasks with multiple user stories that need to be completed and verified.

Quick Start

Start Ralph for a task by saying, "ralph run this task."

Frequently Asked Questions about ralph

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

FAQPage Schema
How do I ensure task completion with a verification loop for complex software development?

You ensure task completion by using a PRD-driven execution model that mandates reviewer verification before marking user stories as passed, continuously monitoring progress until all tasks are fully verified.

How does parallel execution work for concurrent task tracking?

Parallel execution works by concurrently dispatching agent calls to handle multiple user stories simultaneously, structuring the workflow to track progress and report status through a prd.json file.

What is PRD-driven execution and how does it track task progress?

PRD-driven execution uses a prd.json file as the central tracking mechanism to structure software development tasks, enabling continuous work on user stories and mandatory verification reporting until completion.

Do I need a specific environment to run parallel agent calls and verification loops?

You need a Grok environment that supports OMC's orchestration and Grok's plugin architecture to run parallel agent calls and execute the mandatory reviewer verification loop for task completion.

Can I use this task verification approach for projects with multiple user stories?

Yes, this approach is ideal for complex software development projects with multiple user stories, as it provides task persistence to continuously work until all stories pass and are verified.

How do I start a task with automated verification and parallel execution?

You start a task by issuing a run command to trigger the PRD-driven execution, which then automatically handles parallel agent dispatches and the verification loop until all user stories pass.