Brhiza/mingyu

八字、紫微、星盘、六爻、梅花、奇门、大六壬、小六壬、塔罗、雷诺曼、灵签、择日一站式玄学算命占卜工具包，输出结构化提示词与数据。提供公开 API、MCP Server 与 skill。

What it solves

This project provides a comprehensive toolkit for traditional Chinese and Western divination (such as Bazi, Zi Wei Dou Shu, Tarot, and I Ching). It automates the complex mathematical calculations required for these systems and generates structured prompts that can be fed into LLMs for interpretation, bridging the gap between traditional divination and modern AI.

How it works

It offers a core library (mingyu-core) for developers to perform calculations programmatically, a REST API for web integration, and an MCP (Model Context Protocol) server. The MCP server allows AI agents (like those in Cursor or Claude Desktop) to directly request divination charts and receive context-aware prompts, enabling the AI to act as an interpreter for the generated data.

Who it’s for

It is designed for developers building AI agents or applications that incorporate traditional divination, as well as users who want to use AI to interpret complex astrological or divination charts.

Highlights

  • Supports a wide range of systems including Bazi, Zi Wei Dou Shu, Western astrology, I Ching, and Tarot.
  • Provides a ready-to-use MCP server for seamless integration with AI agents.
  • Includes a core TypeScript library for custom development.
  • Offers a REST API for lightweight integration without managing complex local dependencies.
  • Features an Android app that connects divination results directly to local AI chat applications.

Related

  • Project
    learnwithu/mingli-master

    An AI-powered Zi Wei Dou Shu astrology tool that uses a Python rule engine for precise chart casting and an LLM for human-like interpretation and visualization.

  • Project
    jingyunstudio/jingyun-dsh

    Jingyun DSH Client is a Tauri‑based desktop front‑end (and an npm plug‑in) that layers a full commercial system—membership, subscription billing, token‑quota control, marketplace for agents/skills/plugins, and cloud sync—on top of DeepSeek Harness. It lets AI developers quickly package their agents as sellable products, either by adding the plug‑in to an existing DSH instance or by building a branded Windows client.

  • Project
    ZJU-OmniAI/Data-Copilot

    Data‑Copilot is an open‑source LLM‑based assistant that automatically turns Chinese‑financial‑data queries into full analysis pipelines—fetching data via Tushare, planning tool calls, running predictions, visualising results, and summarising them—all orchestrated by GPT‑3.5 or Qwen‑72B‑Chat. It ships a command‑line entry point and a Gradio web UI, and is released under the MIT license.

  • Project
    cclank/lanshu-awesome-ai-video-kit

    A comprehensive, automated toolkit for AI video prompt engineering that provides verified prompt formulas, model-specific methodologies, and Claude-integrated skills for 24 different video generation models.