{"slug": "ai-to-help-make-rpg-game", "title": "Ai to help make rpg game", "summary": "A community guide recommends that beginners building an RPG game with AI assistance start by paying for a roughly $20 monthly subscription to Google Gemini, OpenAI or Claude Code, then install the corresponding CLI client and connect it to GitHub through a fine-grained personal access token. The guide instructs users to scope the token to only the game's repository, with Contents set to read and write and Metadata read-only, and to begin with Python and Pygame on a simple project such as a Pong remake or bullet hell scroller. It also cites the 90/90 rule — when you think you are 90% done, you still have 90% to go — and notes that Codex and Claude Code now favor browser-based OAuth login.", "body_md": "To build on the excellent points made by **John6666**, I would suggest a structured approach for those just starting their journey into AI-assisted development.\n\nA first step is securing a professional subscription to either **Google Gemini**, **OpenAI**, or **Claude Code** all of which are around the **$20 monthly** mark. After subscribing, install the respective CLI clients for your system to bridge the gap between the AI and your local environment.\n\nSet up a **GitHub account**, if not already done. Create a **fine-grained API key** within the account settings to provide to the agent. This enables the AI to handle GitHub pushes and commits directly, which can make the workflow easier and more automated.\n\nFor a first build, start with **Python and Pygame**. Have the CLI agent guide through a simple project, such as a *Pong* remake or a *bullet hell scroller*. The goal is to fully flesh out that simple project.\n\nIn coding projects, remember the **90/90 rule**: when you think you are 90% done, you still have 90% to go.\n\n## **Quick Start: Installing AI Agent Tools (Windows)** <-assumption\n\nThe most current way to use these models on Windows is through their official desktop applications or CLI tools. Modern tools like **Codex** and **Claude Code** now favor browser-based **OAuth**, which means no more searching for API keys.\n\n## **1. OpenAI Codex (Desktop App & CLI)**\n\nOpenAI has shifted toward the [Codex Desktop App](https://www.microsoft.com/en-sg/p/codex/9plm9xgg6vks) as the primary command center for Windows. It includes a built-in terminal and better visual diffing than a standard CLI. [1, 2, 3]\n\n- \n**Install App:** Download directly from the Microsoft Store.\n- \n**Install CLI (Optional):** If the terminal is preferred, run`npm install -g @openai/codex` in PowerShell.\n- \n**Authenticate:** Open the app or run`codex login` in the terminal. The browser will open to sign in via the ChatGPT account.\n\n## **2. Claude (Anthropic) CLI**\n\nClaude Code is a high-speed agent that lives in the terminal.\n\n## **3. Google Gemini CLI**\n\nGoogle provides a dedicated CLI for interacting with the Gemini 1.5 Pro and Flash models.\n\n**Pro Tip:** If any of these `npm` commands fail, ensure the latest **LTS version of Node.js** is installed from [nodejs.org](http://nodejs.org).\n\nTo make your life easier, you’ll want to **give your agent a **Fine-Grained Personal Access Token (PAT)****. This is more secure than old-school tokens because you can limit the agent’s access to *only* the specific projects it needs to work on.\n\n## **Step-by-Step: Creating a GitHub Fine-Grained Token**\n\n1. \n**Go to Settings:** Log in to GitHub, click your profile picture in the top right, and select**Settings** .\n2. \n**Developer Settings:** On the left sidebar, scroll all the way to the bottom and click**Developer settings** .\n3. \n**Personal Access Tokens:** Click**Personal access tokens** >**Fine-grained tokens** .\n4. \n**Generate New Token:** Click**Generate new token** . You may be asked to confirm your password.\n5. \n**Configure the Basics:**\n   - \n**Token name:** Give it a name like`AI-Agent-CLI` .\n  - \n**Expiration:** Set this to 90 days (or whatever you’re comfortable with).\n  - \n**Description:** “Token for AI CLI agent to manage pushes and commits.”\n6. \n**Repository Access:**\n   - \nSelect **Only select repositories** .\n  - \nPick the specific repository you are using for your game (e.g., your *Pong* or*Bullet Hell* repo). This ensures the AI can’t touch your other projects.\n7. \n**Permissions:** Click the**Permissions** dropdown and set the following under**Repository permissions** :\n   - \n**Contents:** Set to**Read and write** (This allows the agent to commit and push code).\n  - \n**Metadata:** Set to**Read-only** (Usually set by default).\n  - \n**Workflows:** Set to**Read and write** (Optional—only if you want the AI to manage GitHub Actions).\n8. \n**Generate:** Scroll to the bottom and click**Generate token** .\n\n **Important:** Copy that token immediately and save it in a secure place (like a password manager). **GitHub will never show it to you again.**\n\n##", "url": "https://wpnews.pro/news/ai-to-help-make-rpg-game", "canonical_source": "https://discuss.huggingface.co/t/ai-to-help-make-rpg-game/166949#post_8", "published_at": "2026-10-23 16:33:13+00:00", "updated_at": "2026-10-23 16:40:05.158337+00:00", "lang": "en", "topics": ["ai-tools", "ai-agents", "developer-tools", "ai-products"], "entities": ["Google Gemini", "OpenAI", "Claude Code", "GitHub", "Python", "Pygame", "Codex", "Anthropic"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/ai-to-help-make-rpg-game", "markdown": "https://wpnews.pro/news/ai-to-help-make-rpg-game.md", "text": "https://wpnews.pro/news/ai-to-help-make-rpg-game.txt", "jsonld": "https://wpnews.pro/news/ai-to-help-make-rpg-game.jsonld"}}