autoplan

Orchestrate CEO, design, eng, and DX skill checks sequentially with a six-principle decision framework.

Updated Apr 11, 2026
One-click install
npx skills add https://github.com/Scivor/helm --skill autoplan-scivor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: autoplan
Source: https://github.com/Scivor/helm/tree/main/skills/gstack/autoplan
Command: npx skills add https://github.com/Scivor/helm --skill autoplan-scivor

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates end-to-end plan reviews by orchestrating CEO, design, engineering, and DX skills from disk.

Core Features & Use Cases

  • Runs the CEO, design, eng, and DX review skills sequentially with auto-decisions using a six-decision framework.
  • Surfaces taste decisions (close approaches, borderline scope, codex disagreements) at a final approval gate.
  • One-command execution to produce a fully reviewed plan.
  • Proactively suggests next steps when the user has a plan file and wants a full gauntlet without answering many questions.
  • Voice triggers (speech-to-text aliases) like "auto plan" or "automatic review" for hands-free operation.

Quick Start

Run autoplan to automatically run the full set of reviews on a 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 multiple domains?

Automating engineering plan reviews is handled by sequentially orchestrating CEO, design, eng, and DX skill checks from disk. The workflow applies a six-principle decision framework to auto-decide standard plan elements, surfacing only subjective taste decisions at a final approval gate.

What is a six-principle decision framework for auto-decisions in code review?

A six-principle decision framework for auto-decisions is a structured evaluation method applied sequentially during an engineering plan review. It automates standard criteria processing, reserving human intervention for borderline scope, close approaches, and codex disagreements at a final gate.

How to run a full plan review gauntlet with a single command?

To run a full plan review gauntlet with a single command, execute the autoplan process when a plan file exists. It proactively prompts next steps, sequentially loading skill definitions from disk to produce a fully reviewed plan without requiring manual question answering.

Does the gstack plan-review workflow support hands-free voice triggers?

Yes, the gstack plan-review workflow supports hands-free voice triggers using speech-to-text aliases like auto plan or automatic review. This enables voice-triggered operation to execute the sequential engineering review pipeline without manual keyboard input.

When do I need to manually intervene in an automated engineering workflow?

Manual intervention in an automated engineering workflow is required at the final approval gate specifically for taste decisions. This includes reviewing close approaches, borderline scope expansions, and codex disagreements that the auto-decision framework cannot resolve automatically.