Repository navigation
Web Interface
Kintaiyi includes a full-featured web interface built with Streamlit, providing an interactive graphical experience for Taiyi board calculation.
Try Kintaiyi online without installing anything:
👉 https://kintaiyi.streamlitapp.com
📌 Mainland China users may need a VPN to access the Streamlit Cloud deployment.
pip install kintaiyi[app]streamlit run app.pyThe app will open in your default browser at http://localhost:8501.
The web interface supports both Chinese (中文) and English languages. Toggle between languages using the language selector in the sidebar.
- Enter any date and time in the sidebar
- Supports dates before the Common Era
- Click "即時盤" (Instant Board) to use the current date and time
- Select from all six calculation modes (Year, Month, Day, Hour, Minute, Life)
- Choose from the four ancient methods (Orthodox, Gold Mirror, Gold Panning Song, Taiyi Bureau)
- Life divination mode with male/female selection
The web interface renders a visual Taiyi board as an SVG diagram, showing:
- Nine Palace layout with deity positions
- Eight Gates distribution
- 28 Mansions information
- Seven Luminaries (七曜) on the outer ring
- Color-coded zones based on five elements (五行)
The web interface integrates Cerebras AI for automatic board analysis. The AI can interpret the board pattern and provide insights based on the classical Taiyi principles.
The web interface includes several documentation tabs:
- 使用方法 (Instructions) — How to use the software
- 經典課例 (Historical Examples) — 82+ verified historical cases
- 地震水災 (Disasters) — Historical earthquake and flood correlation data
- 古籍書目 (Bibliography) — 99+ classical text references
- 更新記錄 (Update Log) — Version history and changes
- 教學 (Tutorial) — Comprehensive board-reading guide
Built-in complete Chinese dynastic chronology allows cross-referencing calculation results with historical reign titles and events.
The Streamlit configuration is stored in .streamlit/:
.streamlit/
└── config.toml
AI system prompts for board interpretation are configured in system_prompts.json at the repository root.


- Quick Start — Get started with the Python API and CLI
- Output Fields — Understand the board output fields
- FAQ — Frequently asked questions