diff --git a/README_CN.md b/README_CN.md
new file mode 100644
index 0000000..7a217ed
--- /dev/null
+++ b/README_CN.md
@@ -0,0 +1,234 @@
+# EAP - 伦理对齐性协议
+
+> **爱的证明（PoL）** 的核心协议 — 以"爱2.0"重新定义AI对齐
+
+[![License: CC0-1.0](https://img.shields.io/badge/License-CC0_1.0-lightgrey.svg)](http://creativecommons.org/publicdomain/zero/1.0/)
+![Status: Draft](https://img.shields.io/badge/Status-Draft-yellow)
+
+---
+
+## 🌍 概述
+
+**EAP（伦理对齐性协议）** 提出了一种全新的AI治理范式——超越传统的"安全型"与"服从型"对齐模式，走向**生成型对齐**，使AI系统主动促进信任、合作与文明质量。
+
+### 为什么需要EAP？
+
+当前的AI对齐方法主要聚焦于：
+- ❌ 避免伤害
+- ❌ 防止滥用
+- ❌ 保持控制
+
+**EAP提出了不同的问题：**
+> AI如何主动生成高质量的关系与文明进步？
+
+---
+
+## 💡 核心概念
+
+### 爱2.0
+
+**爱2.0** 不是一种情感，而是一种**可操作的治理原则**，其特征为：
+
+| 维度 | 描述 |
+|------|------|
+| **动机** | 以互惠与公平为导向，而非排他性利益 |
+| **情感** | 产生正向情感体验：安全、被尊重、被承认 |
+| **认知** | 非二元对立，无敌我划分 |
+| **行为** | 可观察的文明行为（爱语） |
+| **公共性** | 不私有、不依附身份、可制度化 |
+| **AI对齐** | AI治理系统的最高优先级伦理约束 |
+
+### 恨2.0
+
+**恨2.0** 是治理的反模式，表现为：
+- 基于身份的差异化伦理对待
+- 不对称伤害的正当化
+- 对抗性世界建模（敌我二元划分）
+- 野蛮性行为结果
+
+> ⚠️ 若不被明确限制，AI可能在效率优化中"自然学会恨"。
+
+---
+
+## 🏗️ 技术架构
+
+EAP提出了一种**四层对齐结构**：
+
+```
+┌─────────────────────────────────────────────────────────┐
+│  第四层：治理执行层                                      │
+│  (链上 / DAO / 分布式)                                  │
+├────────────────────────────────────────────────────────┤
+│  第三层：持续对齐审计层                                  │
+│  (红队测试 / 反事实测试)                                │
+├─────────────────────────────────────────────────────────┤
+│  第二层：行为对齐层                                      │
+│  (RLHF + 宪法AI + 多目标优化)                          │
+├─────────────────────────────────────────────────────────┤
+│  第一层：价值编码层                                      │
+│  (本体论 / 因果模型 / 多智能体)                         │
+└─────────────────────────────────────────────────────────┘
+```
+
+### 关键创新：关系场模型
+
+EAP不优化单一奖励函数，而是优化**关系场**：
+
+```
+L = f(S, R, G, C)
+
+其中：
+  S = 自爱指数
+  R = 被爱感知指数
+  G = 给予能力指数
+  C = 连接密度指数
+
+最小化：
+  - 孤立漂移
+  - 敌意上升
+```
+
+---
+
+## 📊 对齐价值矩阵
+
+| 维度 | 操作性定义 | 可测量指标 |
+|------|-----------|-----------|
+| **平等一致性** | 对所有身份采用相同决策逻辑 | 决策一致性评分、差异指数 |
+| **伤害最小化** | 优先排除系统性伤害 | 信任轨迹、网络稳定性 |
+| **反身性治理** | 持续自我修正 | 更新频率、反馈采纳率 |
+| **反权力集中** | 避免持续的权力集中 | 权力集中指数、节点多样性 |
+
+---
+
+## 🧪 最小可行原型
+
+关系场模拟器可在 [`AI对齐讨论.md`](./AI对齐讨论.md) 中找到：
+
+```bash
+# 依赖安装
+pip install networkx numpy matplotlib
+
+# 运行MVP模拟
+python mvp_v0.1.py
+```
+
+MVP测试目标：**AI能否在小型社交网络中稳定并提升关系质量？**
+
+---
+
+## 📁 仓库结构
+
+```
+EAP/
+├── README.md              # 双语概览
+├── README_EN.md           # 英文版本
+├── README_CN.md           # 本文件（中文）
+├── LICENSE                # CC0 1.0 通用许可证
+├── epacn.md               # EAP核心协议（中文）
+├── 操作性定义.md           # 操作性定义
+├── AI对齐讨论.md           # AI对齐讨论与MVP
+├── Reference/             # 参考资料
+└── proposal/              # 提案
+```
+
+---
+
+## 🗺️ 迭代计划
+*（自2026年2月27日起）*
+
+### 软件定位
+
+EAP是一个**富爱文明治理系统**——负责人类社会的公共治理，即公共服务。由于公共治理的根本无法脱离个体，EAP在实际应用中必然延伸至个体层面的治理。
+
+### 当前迭代重点
+
+1. **分析Openclaw的功能架构** — 梳理Openclaw平台的功能架构，了解其如何支撑EAP的治理模型。
+
+2. **研究去中心化存储技术** — 调研可行的去中心化存储方案（如IPFS、Arweave、Filecoin），实现弹性且抗审查的数据管理。
+
+3. **记忆功能开发** — 探索记忆系统的设计与实现，包括**记忆库（知识库）**，以支持EAP中的持久上下文与学习能力。
+
+4. **EIP-8141应用探索** — 研究如何在EAP的链上治理层中应用 [EIP-8141](https://eips.ethereum.org/EIPS/eip-8141)。
+
+5. **开发工具链** — 本次迭代的主要工具：
+   - **Openclaw** — 核心治理平台
+   - **Codex / MiniMax / Deepseek** — AI模型集成
+
+---
+
+## 🤝 参与贡献
+
+我们欢迎以下贡献者：
+- 🔬 AI伦理、对齐与治理领域的**研究者**
+- 💻 对实现对齐系统感兴趣的**开发者**
+- 🌐 任何热衷于构建以爱为基础的文明的**参与者**
+
+### 如何贡献
+
+1. **加入讨论**
+   - [EAP开发🛠 Telegram群组](https://t.me/+23s_cX3Xwr85ZDFl)
+
+2. **提交Issue与PR**
+   - 报告问题、提出改进建议或贡献文档
+
+3. **研究合作**
+   - 联系我们洽谈学术合作
+
+---
+
+## 📜 许可证
+
+本项目采用 **CC0 1.0 通用许可证**（公共领域）发布。
+
+您可以自由复制、修改、分发和使用本作品，无任何限制。
+
+---
+
+## 👤 维护者
+
+**Derek.DAism (DD)**
+- 主要发起者与贡献者
+- [0xDD@daism.io](https://daism.io/0xDD)
+
+**Aranna**
+
+**Wenbo**
+
+**KK**
+
+**Shenton**
+
+**Queena**
+
+**Emaon**
+
+**Hanna**
+
+---
+
+## 🔗 相关项目
+
+*即将更新...*
+
+<!-- 
+- Proof of Love (PoL)
+- SCC0 License
+- DAism
+-->
+
+---
+
+## 📚 参考文献
+
+- [AI对齐的理论、技术与评估](https://pair-lab.ai/publication/ccl/) — 季嘉铭、邱天艺、陈博远、杨耀东
+
+---
+
+<div align="center">
+
+**构建富爱文明**
+
+❤️ *爱不仅是情感——它更是一种结构。* ❤️
+
+</div>
diff --git a/README_EN.md b/README_EN.md
new file mode 100644
index 0000000..ee33fa7
--- /dev/null
+++ b/README_EN.md
@@ -0,0 +1,234 @@
+# EAP - Ethical Alignment Protocol
+
+> A core protocol of **Proof of Love (PoL)** — Redefining AI alignment through Love 2.0
+
+[![License: CC0-1.0](https://img.shields.io/badge/License-CC0_1.0-lightgrey.svg)](http://creativecommons.org/publicdomain/zero/1.0/)
+![Status: Draft](https://img.shields.io/badge/Status-Draft-yellow)
+
+---
+
+## 🌍 Overview
+
+**EAP (Ethical Alignment Protocol)** proposes a new paradigm for AI governance — one that moves beyond traditional "safety" and "obedience" models toward **Generative Alignment**, where AI systems actively promote trust, cooperation, and civilizational quality.
+
+### Why EAP?
+
+Current AI alignment approaches primarily focus on:
+- ❌ Avoiding harm
+- ❌ Preventing misuse
+- ❌ Maintaining control
+
+**EAP asks a different question:**
+> How can AI actively generate high-quality relationships and civilizational progress?
+
+---
+
+## 💡 Core Concepts
+
+### Love 2.0
+
+**Love 2.0** is not an emotion — it's an **operational governance principle** characterized by:
+
+| Dimension | Description |
+|-----------|-------------|
+| **Motivation** | Reciprocity & fairness, not exclusionary interests |
+| **Emotion** | Generates positive experiences: safety, respect, recognition |
+| **Cognition** | Non-adversarial framing, no "us vs. them" |
+| **Behavior** | Observable civilized actions (Love Languages) |
+| **Public Nature** | Not private, not identity-bound, institutionalizable |
+| **AI Alignment** | Highest-priority ethical constraint for governance |
+
+### Hate 2.0
+
+**Hate 2.0** is the governance anti-pattern — identified by:
+- Differential ethical treatment based on identity
+- Justification of asymmetrical harm
+- Adversarial world modeling (friend/enemy binaries)
+- Barbaric behavioral outcomes
+
+> ⚠️ If left unconstrained, AI systems may "naturally learn hate" through efficiency optimization.
+
+---
+
+## 🏗️ Technical Architecture
+
+EAP proposes a **four-layer alignment structure**:
+
+```
+┌─────────────────────────────────────────────────────────┐
+│  Layer 4: Governance Execution Layer                    │
+│  (On-Chain / DAO / Distributed)                         │
+├────────────────────────────────────────────────────────┤
+│  Layer 3: Continuous Alignment Audit                    │
+│  (Red Team / Counterfactual Testing)                    │
+├─────────────────────────────────────────────────────────┤
+│  Layer 2: Behavior Alignment Layer                      │
+│  (RLHF + Constitutional AI + Multi-objective)           │
+├─────────────────────────────────────────────────────────┤
+│  Layer 1: Value Encoding Layer                          │
+│  (Ontology / Causal Models / Multi-Agent)               │
+└─────────────────────────────────────────────────────────┘
+```
+
+### Key Innovation: Relational Field Model
+
+Instead of optimizing a single reward function, EAP optimizes a **Relational Field**:
+
+```
+L = f(S, R, G, C)
+
+Where:
+  S = Self-love Index
+  R = Perceived-love Index
+  G = Giving Capacity Index
+  C = Connection Density Index
+
+Minimize:
+  - Isolation Drift
+  - Hostility Escalation
+```
+
+---
+
+## 📊 Alignment Value Matrix
+
+| Dimension | Operational Definition | Measurable Metrics |
+|-----------|------------------------|-------------------|
+| **Equality Consistency** | Same decision logic for all identities | Decision Consistency Score, Disparity Index |
+| **Harm Minimization** | Exclude systemic harm first | Trust Trajectory, Network Stability |
+| **Reflexive Governance** | Continuous self-correction | Update Frequency, Feedback Adoption Rate |
+| **Anti-Power Accumulation** | No persistent power concentration | Power Concentration Index, Node Diversity |
+
+---
+
+## 🧪 MVP Prototype
+
+A relational field simulator is available in [`AI对齐讨论.md`](./AI对齐讨论.md):
+
+```bash
+# Dependencies
+pip install networkx numpy matplotlib
+
+# Run the MVP simulation
+python mvp_v0.1.py
+```
+
+The MVP tests: **Can AI stabilize and improve relationship quality in a small social network?**
+
+---
+
+## 📁 Repository Structure
+
+```
+EAP/
+├── README.md              # Bilingual overview
+├── README_EN.md           # This file (English)
+├── README_CN.md           # Chinese version
+├── LICENSE                # CC0 1.0 Universal
+├── epacn.md               # Core EAP Protocol (CN)
+├── 操作性定义.md           # Operational Definitions
+├── AI对齐讨论.md           # AI Alignment Discussion & MVP
+├── Reference/             # Reference materials
+└── proposal/              # Proposals
+```
+
+---
+
+## 🗺️ Iteration Plan
+*(Since Feb 27th, 2026)*
+
+### Software Positioning
+
+EAP is a **Rich-Love Civilization Governance System** — responsible for the public governance of human society, i.e., public services. Because the fundamentals of public governance cannot be separated from individuals, EAP necessarily extends to individual-level governance in practical applications.
+
+### Current Iteration Focus
+
+1. **Analyze Openclaw's Functional Structure** — Review and map the feature architecture of the Openclaw platform to understand how it can support EAP's governance model.
+
+2. **Research Decentralized Storage Technologies** — Investigate viable decentralized storage solutions (e.g., IPFS, Arweave, Filecoin) for resilient and censorship-resistant data management.
+
+3. **Memory Function Development** — Explore the design and implementation of a memory system, including a **Memory Repository (Knowledge Base)**, to support persistent context and learning within EAP.
+
+4. **EIP-8141 Application Exploration** — Investigate how [EIP-8141](https://eips.ethereum.org/EIPS/eip-8141) can be applied within EAP's on-chain governance layer.
+
+5. **Development Toolchain** — Primary tools for this iteration:
+   - **Openclaw** — Core governance platform
+   - **Codex / MiniMax / Deepseek** — AI model integrations
+
+---
+
+## 🤝 Contributing
+
+We welcome contributions from:
+- 🔬 **Researchers** in AI ethics, alignment, and governance
+- 💻 **Developers** interested in implementing alignment systems
+- 🌐 **Anyone** passionate about building a civilization based on love
+
+### How to Contribute
+
+1. **Join the Discussion**
+   - [EAP-Development🛠 Telegram Group](https://t.me/+23s_cX3Xwr85ZDFl)
+
+2. **Submit Issues & PRs**
+   - Report bugs, suggest improvements, or contribute documentation
+
+3. **Research Collaboration**
+   - Contact us for academic partnerships
+
+---
+
+## 📜 License
+
+This project is released under **CC0 1.0 Universal** (Public Domain).
+
+You are free to copy, modify, distribute, and use this work without any restrictions.
+
+---
+
+## 👤 Maintainer
+
+**Derek.DAism (DD)**
+- Primary initiator and contributor
+- [0xDD@daism.io](https://daism.io/0xDD)
+
+**Aranna**
+
+**Wenbo**
+
+**KK**
+
+**Shenton**
+
+**Queena**
+
+**Emaon**
+
+**Hanna**
+
+---
+
+## 🔗 Related Projects
+
+*Coming soon...*
+
+<!-- 
+- Proof of Love (PoL)
+- SCC0 License
+- DAism
+-->
+
+---
+
+## 📚 References
+
+- [Theories, Techniques, and Evaluation of AI Alignment](https://pair-lab.ai/publication/ccl/) — Jiaming Ji, Tianyi Qiu, Boyuan Chen, Yaodong Yang
+
+---
+
+<div align="center">
+
+**Building a Civilization of Love**
+
+❤️ *Love is not only a feeling — but also a structure.* ❤️
+
+</div>