Numerologist_skills
AI divination engineering framework. Uses prompt constraints, structured knowledge bases, and external logic scripts to make LLMs accurately understand and apply Qi Men Dun Jia and Zi Wei Dou Shu. Eliminates LLM mystical hallucinations, making calculation processes transparent and controllable.
详见 GitHub 仓库说明：https://github.com/FANzR-arch/Numerologist_skillsBefore / After Comparison
1 组AI fortune-telling relied on vague built-in knowledge, full of hallucinations and errors
Engineered architecture: strict prompt constraints + knowledge base + logic scripts, transparent auditable process
description SKILL.md
Numerologist Skills (AI 术数工程化)
本项目致力于通过“工程化”手段，让大语言模型（LLM）精准理解和应用传统东方术数（如奇门遁甲、紫微斗数）。
核心模块 (Available Skills)
目前仓库中包含以下主要术数模型与知识外挂（Skills）：
- 🀄 qimen-dunjia (奇门遁甲)
- 奇门排盘逻辑脚本与约束分析。
- 囊括用神取法、星门神仪格局、主客关系判断等。
- 🔮 ziwei-doushu (紫微斗数)
- 紫微斗数安星诀与核心算理。
- 包含四化体系、命盘格局分析、以及星曜特性的知识引用。
设计理念与愿景
不再依赖 AI 模糊的内置“玄学幻觉”，而是通过严格的 Prompt 约束 + 结构化知识库 (References) + 外部逻辑脚本 (Scripts) 的架构方式让算理过程透明可控。
Created as part of the AI Engineering of Traditional Divination (AI 术数工程化) initiative.
forumUser Reviews (0)
Write a Review
No reviews yet
Statistics
User Rating
Rate this Skill