skills.rest
filipesf
filipesfCOMMUNITY
Explore all skills by this author →
Website𝕏@filipesfGitHub

ralph

Orchestrate parallel subagent tasks with iterative architect verification.

Updated Aug 23, 2026
Search Tags:#verification#iteration#PRD#ai orchestration#ralph#ultrawork
One-click install
npx skills add https://github.com/filipesf/dotfiles --skill ralph-filipesf
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ralph
Source: https://github.com/filipesf/dotfiles/tree/main/agents/skills/ralph
Command: npx skills add https://github.com/filipesf/dotfiles --skill ralph-filipesf

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ralph provides an autonomous, self-improving task loop with built-in architect verification to ensure progress and final correctness.

Core Features & Use Cases

  • PRD-driven planning: When prompted, initializes Product Requirements Documents and progress tracking.
  • Ultrawork orchestration: Orchestrates parallel tasks across specialized subagents to optimize throughput.
  • Iterative verification: Repeats cycles until complete with formal verification by an architect.
  • Use Case: Suitable for complex product development tasks, research synthesis, and engineering planning that require multiple iterations and validation.

Quick Start

Provide a task prompt to Ralph describing the objective, e.g., "Draft a PRD for a new authentication feature and iteratively refine it until the completion criteria are met." Then allow Ralph to begin the loop and report progress.

Frequently Asked Questions about ralph

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

FAQPage Schema
How do I set up an autonomous task loop with built-in verification for complex engineering tasks?▼

An autonomous task loop with built-in verification guides AI through iterative refinement cycles, enforcing PRD initialization and orchestrating parallel execution across subagents until formal architect verification confirms task completion.

What is the best way to generate and iteratively refine a Product Requirements Document using AI orchestration?▼

PRD generation and iterative refinement are handled by initializing a Product Requirements Document based on your objective, then running self-improving task loops with tiered verification until the specified completion criteria are fully met.

Can I use parallel execution and background tasks for multi-step product development planning?▼

Parallel execution and background tasks are supported through ULTRAWORK orchestration, which distributes complex multi-step product development workloads across specialized subagents to optimize throughput and deliver verifiable outcomes.

How does architect verification work during iterative AI task refinement?▼

Architect verification evaluates progress during iterative AI task refinement by applying tiered checks at each cycle, ensuring that the self-improving loop only terminates when formal verification confirms the final outcome meets all requirements.

When do I need formal verification loops for research synthesis and engineering planning?▼

Formal verification loops are needed for research synthesis and engineering planning when tasks involve multiple iterations, require parallel execution, and demand validated, complete outcomes verified by an architect before delivery.

Related Skills

Explore More Product & Management Skills→
alvisalvis
4

create-skill

Create SKILL.md documents with YAML frontmatter and orchestration-ready directory structures.

Community
Advanced
GenEducationGenEducation

bmad-edit-prd

Edit existing PRDs through structured discovery, review, and controlled updates.

Official
Advanced
rodrigotoledorodrigotoledo

subagent-creator

Create isolated-context subagents with YAML frontmatter metadata.

Community
Intermediate
chanhee-kangchanhee-kang
2

ralph

Automate iterative task refinement and verification with tiered architectural checks.

Community
Advanced