skills.rest
christopher-kapic
christopher-kapicCOMMUNITY
Explore all skills by this author →
GitHub

create-ralph

Generate test-validated dependency DAG execution plans for ralph-rs coding agent harnesses.

Updated Apr 8, 2026
Search Tags:#software development#execution planning#coding agents#test gating#ralph-rs#coding plans#dependency dag
One-click install
npx skills add https://github.com/christopher-kapic/ralph-rs --skill create-ralph
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: create-ralph
Source: https://github.com/christopher-kapic/ralph-rs/tree/main/.claude/skills/create-ralph
Command: npx skills add https://github.com/christopher-kapic/ralph-rs --skill create-ralph

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manually breaking down complex coding tasks like multi-phase features, refactors, or bug fixes into executable, test-validated steps for ralph-rs is error-prone and time-consuming, especially when accounting for dependency DAGs, verification gates, and review pipelines. This Skill eliminates that manual work by providing a structured, guided workflow to create correct, efficient ralph-rs execution plans.

Core Features & Use Cases

  • Dependency DAG Planning: Guides you to structure steps as a directed acyclic graph to enable parallel execution of independent branches, so unblocked work continues while other branches wait for human input.
  • Test-Gated Validation: Ensures every step includes concrete, mechanically checkable acceptance criteria and deterministic test commands to catch failures before they propagate.
  • Review Pipeline Integration: Supports configuring ralph-rs's built-in nondeterministic review pipeline or authoring custom review steps for high-risk changes, with guidance on harness selection and retry logic.
  • Use Case: Ideal for breaking down a 3-month migration of a legacy payment processing module into verifiable, parallelizable steps, each with test gates and optional review, to run through a coding agent harness like Claude or Codex.

Quick Start

Use the create-ralph skill to generate a ralph-rs execution plan for adding a new user profile editing feature with test-validated, parallelizable steps.

Frequently Asked Questions about create-ralph

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

FAQPage Schema
How do I create a dependency DAG execution plan for coding agent harnesses?▼

Creating a dependency DAG execution plan involves structuring coding tasks as a directed acyclic graph to enable parallel execution of independent branches. This replaces manual task breakdown by generating structured, test-validated plans for coding agent harnesses.

How do I break down a multi-phase code refactoring task into test-gated steps?▼

Breaking down multi-phase code refactoring requires guiding steps through test-gated validation, ensuring every step includes concrete, mechanically checkable acceptance criteria and deterministic test commands to catch failures before they propagate.

What is deterministic test gating in coding plans and when do I need it?▼

Deterministic test gating in coding plans ensures each execution step includes mechanically checkable acceptance criteria and deterministic test commands. You need it for complex multi-session feature implementation or refactoring scenarios requiring per-step verification.

Does ralph-rs support parallel execution of independent coding task branches?▼

Yes, ralph-rs supports parallel execution of independent branches through dependency DAG planning. Structuring steps as a directed acyclic graph allows unblocked work to continue while other branches wait for human input or dependencies.

How do I add a review pipeline to a coding plan for high-risk changes?▼

Adding a review pipeline involves configuring ralph-rs's built-in nondeterministic review pipeline or authoring custom review steps for high-risk changes, with guidance on harness selection and retry logic.

What is the best way to automate coding task breakdown for multi-session feature implementation?▼

The best way to automate coding task breakdown is using a structured workflow that generates test-validated dependency DAG execution plans, providing correct dependency edge wiring and alignment with dynamic scheduling and retry logic.

Related Skills

Explore More Software Engineering Skills→
learnzdevelopmenthublearnzdevelopmenthub
1

dev-test

Generate test plans and placeholder stubs from acceptance criteria.

Community
Advanced
dmae97dmae97
130

omk-adaptorch-orchestration-review

Review DAG multi-agent orchestration designs for correctness, authority, and evidence-gating failures.

Community
Intermediate
OliverOuyangOliverOuyang
1

ralplan

Gate vague multi-agent prompts through Planner, Architect, and Critic consensus reviews.

Community
Advanced
dmae97dmae97
130

speckit-implement

Execute spec-kit implementation plans from tasks.md and plan.md with dependency gating.

Community
Advanced