autopilot

Automate task planning, implementation, and code review workflows.

24|2|Updated Mar 12, 2025
One-click install
npx skills add https://github.com/Orinks/AccessiWeather --skill autopilot-orinks
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: autopilot
Source: https://github.com/Orinks/AccessiWeather/tree/main/.codex/skills/autopilot
Command: npx skills add https://github.com/Orinks/AccessiWeather --skill autopilot-orinks

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates a strict end-to-end loop that plans, implements, and reviews code based on a defined task request.

Core Features & Use Cases

  • Enforces a three-phase contract: ralplan, ralph, and code-review to ensure disciplined delivery.
  • Supports iterative refinement by returning to planning when code-review findings are not clean.
  • Maintains stateful handoffs and traceable artifacts for auditing and resumption.

Quick Start

Provide a concrete task and let Autopilot execute the three-phase cycle from planning to code review.

Frequently Asked Questions about autopilot

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

FAQPage Schema
How do I automate the entire workflow from task planning to code review?

You can automate this workflow by executing a strict three-phase contract: planning, implementation, and code review. It processes product requirements iteratively, maintaining stateful handoffs and traceable artifacts throughout the end-to-end task loop.

What is the three-phase contract for autonomous code delivery?

The three-phase contract is a sequential workflow of planning, implementation, and code review. It ensures disciplined delivery by maintaining stateful handoffs and triggering a return to the planning phase whenever code review findings are not clean.

How do I handle iterative code refinement when automated review findings are not clean?

When automated code review findings are not clean, you handle iterative refinement by triggering an explicit return to the planning phase. This loop adjusts the implementation strategy before proceeding to the next coding and verification cycle.

Can I use automated task planning for product requirements described as PRDs or feature requests?

Yes, automated task planning applies directly to initiatives described as PRDs, product ideas, or feature requests. It processes these inputs through an iterative cycle of planning, coding, and verification to deliver the final reviewed code.

How do I maintain state and traceable artifacts during an automated implementation loop?

You maintain state and traceable artifacts during an automated implementation loop through stateful handoffs between the planning, implementation, and review phases. These handoffs ensure continuous auditing capabilities and allow seamless resumption of the end-to-end task cycle.

What are the limitations of using an autonomous loop for software engineering workflows?

A key limitation is its strict dependency on the three-phase contract and stateful handoffs to function correctly. Without explicit return-to-planning semantics to resolve unclean code review findings, the autonomous task loop cannot successfully complete the software engineering workflow.