calloftheNIGHT2023

calloftheNIGHT2023 / gamebuddy-agent

Public

浏览器中的开源 AI 游戏分析实验室，结合结构化复盘演示与可游玩的沙盒模块，支持局势解读、策略建议与赛后分析。

www.gamebuddyagent.site ai-agent fastapi game-ai game-coaching gaming
18
1
100% credibility
Found Apr 22, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A browser-based game analysis tool that provides strategic advice from game states or screenshots, bundled with playable prototypes for genres like Pokemon battles, MOBAs, and RPGs.

How It Works

1
🔍 Discover GameBuddy

You find this fun game helper on a sharing site and open its web page in your browser.

2
🎮 Play mini-games

Jump into quick browser games like Pokemon battles or MOBA lanes to feel the strategy excitement.

3
📝 Pick your game

Choose a game type such as turn-based battles, team fights, or character building.

4
💭 Share your situation

Type a quick question like 'what next?' and paste your game details or upload a picture.

5
💡 Get smart tips

Instantly see a clear summary, best moves, common mistakes, and beginner advice tailored just for you.

🏆 Level up your skills

Save your preferences, review past tips, and play better with confidence from your new game insights.

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 gamebuddy-agent?

GameBuddy Agent is a browser-based lab for AI-driven game analysis, where you feed it structured game states or screenshot placeholders to get situation summaries, phase predictions, tactical advice, and review breakdowns tailored for beginners or post-match coaching. Built with a TypeScript Next.js frontend and Python FastAPI backend, it ships playable sandboxes like MOBA lanes, RPG build labs, and survivor prototypes right alongside analysis demos for Pokemon battles, MOBA recaps, and RPG gear checks. Developers get a one-stop repo to prototype agent behaviors without separate game engines.

Why is it gaining traction?

It stands out by bundling heuristic analysis (fallback rules when offline) with optional OpenRouter LLM boosts via function calling, all deployable via Docker Compose or Render in minutes—no heavy vision ML needed for MVP testing. The hook is the seamless mix of API endpoints like POST /api/v1/analyze/state for real analysis and browser-playable modules to validate ideas interactively, saving time on mock setups. Dual English/Chinese UI and MongoDB persistence for user profiles add polish for global experimentation.

Who should use this?

Game AI researchers prototyping coaching agents for turn-based or MOBA titles, indie devs building post-match tools without full game clients, and TypeScript enthusiasts forking browser sandboxes to test strategy heuristics. Perfect for Pokemon Showdown users wanting quick reviews or RPG makers iterating build optimizers via JSON states.

Verdict

Grab it if you're experimenting with game agents—Docker spins up fast, docs cover API shapes and samples, tests exist for core flows—but at 18 stars and 1.0% credibility, treat as an early prototype needing community polish before production. Solid foundation for forking.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.