onton

Decompose large codebase changes into ordered patches for parallel AI execution.

10|4|Updated Mar 15, 2026
One-click install
npx skills add https://github.com/flowglad/onton --skill onton
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: onton
Source: https://github.com/flowglad/onton/tree/main/skills/write-gameplan
Command: npx skills add https://github.com/flowglad/onton --skill onton

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Managing large-scale codebase modifications is challenging due to complexity, dependencies, and risks, making review and merging difficult.

Core Features & Use Cases

  • Change Decomposition: Break down extensive code changes into small, ordered patches with formal specifications.
  • Parallel Execution: Coordinate AI agents to implement patches concurrently based on dependency graphs.
  • Workflow Automation: Automate code reviews, CI integration, rebasing, and merging processes to streamline large project updates.
  • Use Case: When refactoring a monolithic app, generate a structured plan that reduces review effort and mitigates merge conflicts with formal specs.

Quick Start

Use the onton tool to generate a structured plan for your upcoming large change by defining specifications and dependencies beforehand.

Frequently Asked Questions about onton

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

FAQPage Schema
How do I orchestrate large-scale codebase changes using AI agents?

You orchestrate large-scale codebase changes by decomposing them into small, formally specified patches executed by AI agents. This approach manages complexity by coordinating dependency-aware parallel execution and automating review cycles.

What is the best way to automate code reviews for complex refactoring?

The best way to automate code reviews for complex refactoring is by generating a structured plan that breaks down modifications into ordered patches. This workflow automation integrates CI, rebasing, and merging to reduce review effort.

How do I break down extensive code modifications into manageable patches?

You break down extensive code modifications into manageable patches by defining structured specifications and dependencies beforehand. This change decomposition generates a formal plan that mitigates merge conflicts and ensures ordered execution.

Can I coordinate parallel execution of code patches based on a dependency graph?

Yes, you can coordinate parallel execution of code patches based on a dependency graph. By defining formal dependencies beforehand, AI agents implement patches concurrently while ensuring dependency-aware execution.

Do I need formal specifications to automate complex deployment scenarios?

Yes, you need formal specifications to automate complex deployment scenarios. Structured specifications are required to decompose extensive code changes into ordered patches, ensuring dependency-aware execution and automated review cycles.

When should I not use AI-driven patches for refactoring a monolithic app?

You should not use AI-driven patches for refactoring a monolithic app when you cannot define structured specifications and formal dependencies beforehand. Without these inputs, generating a dependency-aware structured plan is not possible.