SkillAgentSearch skills...

MahjongCopilot

雀魂游戏实时 AI 指导 / In-game step-by-step AI Assistant for Maj-soul

Install / Use

/learn @latorc/MahjongCopilot
About this skill

Quality Score

0/100

Supported Platforms

GitHub Copilot

README

麻将 Copilot / Mahjong Copilot

麻将 AI 助手,基于 mjai (Mortal模型) 实现的机器人。会对游戏对局的每一步进行指导。现支持雀魂三人、四人麻将。

加入QQ群:1031865144 <br> Click to <a href="https://discord.gg/7hcZYTFw5r" target="_blank">Join Discord</a>

Mahjong AI Assistant for Majsoul, based on mjai (Mortal model) bot impelementaion. When you are in a Majsoul game, AI will give you step-by-step guidance. Now supports Majsoul 3-person and 4-person game modes.

下载、帮助和更多信息请访问网站 Please see website for download, help, and more information
<a href="https://mjcopilot.com/help" target="_blank">帮助信息 Help Info </a> | <a href="https://mjcopilot.com" target="_blank">https://mjcopilot.com</a>


特性:

  • 对局每一步 AI 指导,可在游戏中覆盖显示
  • 自动打牌,自动加入游戏
  • 多语言支持
  • 支持本地 Mortal 模型和在线模型,支持三麻和四麻

Features:

  • Step-by-step AI guidance for the game, with optional in-game overlay.
  • Auto play & auto joining next game
  • Multi-language support
  • Supports Mortal local models and online models, 3p and 4p mahjong modes.

<a id="instructions"></a>

使用方法 / Instructions

开发

  1. 克隆 repo
  2. 安装 Python 虚拟环境。Python 版本推荐 3.11.
  3. 安装 requirements.txt 中的依赖。
  4. 安装 Playwright + Chromium
  5. 主程序入口: main.py

To Develope

  1. Clone the repo
  2. Install Python virtual environment. Python version 3.11 recommended.
  3. Install dependencies from requirements.txt
  4. Install Playwright + Chromium
  5. Main entry: main.py

示例脚本 Sample script:

git clone https://github.com/latorc/MahjongCopilot.git
cd MahjongCopilot
python -m venv venv
CALL venv\Scripts\activate.bat
pip install -r requirements.txt
set PLAYWRIGHT_BROWSERS_PATH=0
playwright install chromium
python main.py

配置模型

本程序支持几种模型来源。其中,本地模型(Local)是基于 Akagi 兼容的 Mortal 模型。要获取 Akagi 的模型,请参见 <a href="https://github.com/shinkuan/Akagi" target="_blank"> Akagi Github </a> 的说明。

Model Configuration

This program supports different types of AI models. The 'Local' Model type uses Mortal models compatible with Akagi. To acquire Akagi's models, please refer to <a href="https://github.com/shinkuan/Akagi" target="_blank"> Akagi Github </a>.

截图 / Screenshots

界面 / GUI

游戏中覆盖显示 (HUD)/ In-game Overlay (HUD)

设计 / Design

目录说明 Description for folders:

  • gui: tkinter GUI 相关类 / tkinter GUI related classes
  • game: 雀魂游戏相关类 / classes related to Majsoul game
  • bot: AI 模型和机器人实现 / implementations for AI models and bots
  • common: 共同使用的支持代码 commonly used supporting code
  • libriichi & libriichi3p: 编译完成的 libriichi 库文件 / For compiled libriichi libraries

鸣谢 / Credit

  • 基于 Mortal 模型和 MJAI 协议 Based on Mortal Model an MJAI protocol

    Mortal: https://github.com/Equim-chan/Mortal

  • 设计和功能实现基于 Akagi Design and implementation based on Akagi

    Akagi: https://github.com/shinkuan/Akagi

  • 参考 Reference Mahjong Soul API: https://github.com/MahjongRepository/mahjong_soul_api

  • MJAI协议参考 / MJAI Protocol Reference

    MJAI: https://mjai.app

许可 / License

本项目使用 GNU GPL v3 许可协议。
协议全文请见 LICENSE

Related Skills

View on GitHub
GitHub Stars954
CategoryDevelopment
Updated2d ago
Forks124

Languages

Python

Security Score

100/100

Audited on Mar 27, 2026

No findings