ColinHong10

ColinHong10 / NRC_AI

Public

AI自动化战斗的洛克王国世界

18
1
100% credibility
Found Apr 02, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A simulator for Rokugou (洛克王国) game battles featuring AI-driven matches between predefined teams, player versus AI mode, batch statistics, and an experience system where AI improves over multiple games.

How It Works

1
🕵️ Discover the Fun Battle Simulator

You stumble upon this cool tool on GitHub that lets you simulate intense battles from the Rokugou game with smart AI creatures.

2
📥 Get It Running Easily

Download the files, double-click the starter on Windows or run the simple launch script, and a welcoming menu pops up right away.

3
Pick Your Battle Mode
🤖
Watch AI vs AI

Sit back as two AI teams battle it out with clever strategies.

🎮
Play Against AI

Take control of one team and make moves against a thinking opponent.

📈
Run Simulations

Launch batches of battles to see win rates and how AI improves over time.

4
⚔️ Epic Battles Unfold

Watch creatures use skills, switch fighters, build energy, and outsmart each other in thrilling turn-based clashes that feel just like the real game.

5
📊 Check Results and Growth

See who wins, average fight lengths, and how your AI gets smarter with each battle by learning from past fights.

🏆 Master of Rokugou Battles

You've run exciting simulations, beaten the AI or watched legends form, and can save progress to pick up where you left off next time.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is NRC_AI?

NRC_AI is a Python-powered battle simulator for Rokugou, a Pokémon-style game, pitting two fixed teams—a toxic squad against a wing king lineup—in automated fights. It solves the pain of manually grinding matches by delivering AI-vs-AI battles via Monte Carlo Tree Search, complete with player-vs-AI mode, batch runs for win-rate stats, and experience learning where AIs get smarter over games. Run it via a simple CLI menu to watch fights, experiment with 50-100 game batches, or tweak simulation depth for deeper analysis.

Why is it gaining traction?

It stands out with a rich ruleset modeling energy costs, speed priorities, stacking status effects like poison and freeze, and counter interactions between skills—far beyond basic random sims. Developers dig the learning curve visualization across phased experiments, showing AI win rates climb with battle history, plus bilingual docs and easy CSV/Excel data loading for custom tweaks. In the NRC AI workshop scene, it's a quick Python entry for testing strategic plans without building from scratch.

Who should use this?

Game AI hobbyists scripting Rokugou team tests, reinforcement learning devs prototyping MCTS in zero-sum games, or Python scripters analyzing battle metas via batch stats. It's ideal for NRC AI zone participants running productivity boosts on sims, or students at NRC AI symposiums 2025 dissecting turn-based tactics like aircraft impact rules in a fun wrapper.

Verdict

Grab it for Rokugou fans or MCTS tinkering—solid CLI and docs make it approachable despite 18 stars and 1.0% credibility signaling early-stage code. Polish tests and generalize teams to boost maturity, but it's already a clever NRC AI assist for quick strategic experiments.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.