autoplan

Execute an end-to-end auto-review pipeline for CEO, design, and Eng plans.

1|Updated Mar 23, 2026
One-click install
npx skills add https://github.com/txema-puch/drone-ai-saturdays --skill autoplan-txema-puch
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: autoplan
Source: https://github.com/txema-puch/drone-ai-saturdays/tree/main/.claude/skills/gstack/autoplan
Command: npx skills add https://github.com/txema-puch/drone-ai-saturdays --skill autoplan-txema-puch

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates end-to-end auto-review of CEO, design, and Eng review skills by reading them from disk and applying six decision principles, delivering a fully reviewed plan with a final approval gate.

Core Features & Use Cases

  • Auto-run the complete review pipeline by loading skill files from disk and executing a deterministic decision process.
  • Surface taste decisions (close approaches, borderline scope, codex disagreements) at a final approval gate for human review when necessary.
  • One-command execution that outputs a comprehensive plan ready for governance, QA, and deployment.

Quick Start

Run the /autoplan command to execute the entire auto-review pipeline and produce a fully reviewed plan.

Frequently Asked Questions about autoplan

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

FAQPage Schema
How do I automate engineering plan reviews across CEO, design, and Eng phases?

Automating engineering plan reviews is done by loading CEO, design, and Eng skill files from disk and applying six decision principles. This auto-review pipeline executes a deterministic workflow across all phases, outputting a fully reviewed plan.

What is an auto-review pipeline for software engineering plans?

An auto-review pipeline is an end-to-end process that evaluates software plans using predefined decision principles. It operates across CEO, design, and engineering phases to auto-decide interactions and surface contentious taste decisions at a final approval gate.

How do I surface contentious taste decisions for human review in an automated pipeline?

Surfacing contentious taste decisions is handled at a final approval gate within the auto-review pipeline. The system identifies borderline scope, close approaches, and codex disagreements, pausing these specific items for human review.

Can I generate a fully decided software plan with traceable decisions for QA?

Generating a fully decided plan with traceable decisions is the primary output of this deterministic auto-review workflow. It executes the complete review lifecycle, providing a comprehensive plan ready for governance, QA, and deployment.

Do I need specific skill files on disk to run an automated engineering review?

Running an automated engineering review requires having CEO, design, and Eng skill files loaded from disk. These files provide the necessary parameters for the six decision principles applied throughout the review lifecycle.

What are the limitations of using a deterministic auto-review process for engineering plans?

A limitation of this deterministic auto-review process is that it cannot fully resolve subjective taste decisions. While it auto-decides standard interactions, borderline scope and disagreements must be escalated to a final human approval gate.