# Setup Automation

> Inspect a game project and propose the optional Codex hooks, scoped AGENTS.md guidance, and selected local agent profiles. Use when adopting the automation pack or changing its engine-specific project setup; never apply configuration before digest confirmation.

- **Type:** Skill
- **Install:** `agentstack add skill-frabcd-codex-ai-game-studio-setup-automation`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [frabcd](https://agentstack.voostack.com/s/frabcd)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [frabcd](https://github.com/frabcd)
- **Source:** https://github.com/frabcd/codex-ai-game-studio/tree/main/plugins/ai-game-studio-automation/skills/setup-automation
- **Website:** https://frabcd.github.io/codex-ai-game-studio/

## Install

```sh
agentstack add skill-frabcd-codex-ai-game-studio-setup-automation
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Set up automation

Keep inspection and proposal read-only. Do not trust hooks, write project files, or materialize agents until the user confirms the exact transaction digest.

1. Run the core `doctor --project ` and report the detected OS, engine, existing guidance, active MCP hosts, and constraints.
2. Select one engine value: `unity`, `godot`, `unreal`, `browser`, or `generic`.
3. Run `scripts/project_setup.py plan --project  --engine ` from this plugin. For an install proposal that may later be confirmed, add `--output ` using a temporary path outside ``.
4. Present the proposed root/path-scoped `AGENTS.md` files, common roles, selected engine specialists, permissions, backups, expiry, rollback operations, and digest. Mention that hooks remain disabled until separately reviewed and trusted through `/hooks`.
5. Wait for the user to confirm that digest verbatim.
6. Apply only with `scripts/project_setup.py apply --project  --plan  --confirmed-digest `.
7. Validate the installed guidance and agent TOML, then report the transaction ID. If the user wants to undo it, generate a rollback proposal with the core `pack rollback `; never roll back without a second digest confirmation.

Preserve existing files through the transaction backup. Agents inherit the active Codex model and permission mode; do not add model pins, unsupported tool declarations, memory fields, or fixed turn limits.

For preview-only requests, present the proposal and stop after step 4. Do not solicit confirmation, write a plan inside the project, or apply the plan.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [frabcd](https://github.com/frabcd)
- **Source:** [frabcd/codex-ai-game-studio](https://github.com/frabcd/codex-ai-game-studio)
- **License:** MIT
- **Homepage:** https://frabcd.github.io/codex-ai-game-studio/

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-frabcd-codex-ai-game-studio-setup-automation
- Seller: https://agentstack.voostack.com/s/frabcd
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.