start

Inspects project environment and routes users to the appropriate game-development workflow.

13|Updated Jul 30, 2026
One-click install
npx skills add https://github.com/frabcd/codex-ai-game-studio --skill start-frabcd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: start
Source: https://github.com/frabcd/codex-ai-game-studio/tree/main/plugins/ai-game-studio/skills/start
Command: npx skills add https://github.com/frabcd/codex-ai-game-studio --skill start-frabcd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Starting a game project with an AI assistant often leads to wrong assumptions about the environment, premature installs, or unsafe changes. This Skill performs read-only onboarding: it inspects the project, asks only the minimal blocking questions, and routes you to the correct game-studio workflow without mutating anything. ## Core Features & Use Cases - Read-only environment inspection: Detects OS, architecture, engine, renderer, DCC tools, languages, tests, and existing .ai-game-studio/ state without creating files or installing software. - Intent clarification: Asks only the smallest blocking questions about outcome, platform, commercial use, time budget, and asset rights. - Workflow routing table: Maps your need to one of 16+ downstream skills such as brainstorm, adopt, prompt-to-game, sprite-generate, or engine-automation. - Confirmation boundary: Produces a consolidated transaction plan with actions, licenses, rollback steps, and a digest before any mutating operation. - Use Case: You open an existing Unity prototype and say "help me get started". The Skill inspects the repo, confirms your goal, and recommends $ai-game-studio:adopt with a ready-to-send prompt instead of guessing. ## Quick Start Ask the assistant to use the start skill to inspect this project and route you to the safest next game-development workflow.

Frequently Asked Questions about start

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

FAQPage Schema
How do I start an AI-assisted game development project safely?

Begin with a read-only onboarding step that inspects your OS, engine, tools, and repository state before changing anything. The start skill then asks minimal questions about your goal and routes you to the matching workflow with a ready-to-send prompt.

How do I choose the right workflow for a Unity, Godot, or Unreal project?

The routing table maps your need to a specific skill: adopt for existing projects, setup-engine for engine configuration, prompt-to-game for playable proofs, and sprite-generate or asset-3d-generate for asset work. Detection of your engine and project markers drives the recommendation.

Does the onboarding skill install tools or change my configuration?

No. Routing is strictly read-only: it never creates files, installs software, starts MCP servers, launches editors, or downloads models. Any mutating action first requires a consolidated transaction plan with rollback steps and your explicit confirmation of its digest.

Can it detect my API keys or credentials during inspection?

It only checks credential presence by environment-variable name. It never prints, reads, or stores secret values, so your keys remain untouched during environment inspection.

What happens if the environment changes after a plan is proposed?

The confirmation digest becomes invalid when the environment changes or differs from the plan. A new consolidated transaction plan must be created and confirmed again before any action proceeds.