🥷 I Built a Procedural Ninja Stealth Game in Python – "Openstate" (with Amazon Q CLI)

Ever wanted to sneak through a maze as a shadowy ninja, dodging guards, grabbing scrolls, and escaping undetected?
That’s exactly what I built with Openstate — a 2D stealth platformer game using Python, Pygame, and scaffolding powered by Amazon Q CLI.

Let me take you through how I made it, what makes it tick, and how you can play it or even build your own.

🎮 What is Openstate?
Openstate is a procedurally generated stealth game where you:

Sneak past patrolling guards with line-of-sight vision

Collect hidden scrolls across a maze

Use shadows and bushes to hide

Escape through the exit — all without being caught

Each level gets harder, and every run is unique thanks to procedural generation.

🎥 Gameplay Demo
👉 Watch the full gameplay of Openstate here: https://youtu.be/vvg_KIyrTkg

🧠 Key Features
🎯 Cone-of-vision AI: Guards detect you based on realistic field of vision

🔁 Procedurally generated levels: New maze every time

🌌 Dynamic guard speed: Increasing challenge each level

📜 Scroll collection for stars

🥷 Crouch, hide, and sneak to survive

🌟 3-star rating system: Based on scrolls collected, exit reached, and stealth success

🔁 Game over + restart system if caught

🛠️ How It Was Built
📁 Modular Architecture
player.py: Movement, jumping, crouching, hiding

guard.py: AI, patrol logic, detection

level.py: Handles maze, scrolls, exit, guards

maze_generator.py: DFS-based maze generation

ui.py: Timer, star count, detection alerts

💻 Powered by Amazon Q CLI
Amazon Q CLI helped scaffold the basic architecture for this project — making it easier to organize the code and scale features over time.
It was part of the "Build Games with Amazon Q CLI" challenge.

🧪 Challenges Faced
Tuning AI to be “dumb but fair” (nobody likes psychic guards)

Ensuring maze generation always leads to an exit

Getting animations and UI overlays working properly in Pygame

🚀 Try the Game Yourself
💻 Clone the Repo:
bash
Copy
Edit
git clone https://github.com/Kshitij-0007/openstate
cd openstate
pip install -r requirements.txt
python main.py
🔗 GitHub Repo: https://github.com/Kshitij-0007/openstate

🧠 Lessons Learned
Procedural generation adds replayability — but also debugging nightmares 😅

Simple AI systems can feel very smart with just a cone of vision and patrols

Amazon Q CLI can boost productivity and structure — especially for solo devs

🛣️ What’s Next for Openstate?
🔮 Smoke bombs to distract guards

🧠 Smarter guard AI variations

🧱 Level editor for custom mazes

💾 Save/load progress

🛍️ In-game power-ups like invisibility or speed boost

🙌 Final Thoughts
This was built as part of the Amazon Q CLI Game Jam, and I’m super proud of how it turned out. If you're from Amazon and reading this — this ninja humbly accepts T-shirt tributes 👕✨

Thanks for reading, and go give it a try!
Got feedback? Fork it? Want to collab? Let’s connect!

🔧 Built With:
Python 3.9

Pygame

Amazon Q CLI

Canva (video editing)

🕶️ Peace, scrolls, and silent footsteps.

python #pygame #gamedev #amazonqcli #opensource #devchallenge #stealthgame #indiedev #proceduralgeneration

Let me know if you'd like a thumbnail image or a post for LinkedIn
too. Let's get this ninja everywhere. 🥷💻💥