mindcore

人工智能代理的仿生情感思维引擎。提供类似人类的情感 通过 5 层神经传导管道进行响应（L0 随机噪声 → L1 传感器感知 → L2 潜意识冲动 → L3 人格门 → L4 决策输出）加上 5 个心理动力学补丁。与任何完全解耦 LLM — 使用纯 Python 在 CPU 上本地运行。模拟 150 次日常脉冲 涵盖昼夜节律、情绪调节和短期的 9 个类别 记忆影响。

# MindCore — Biomimetic Subconscious Engine

> Give your AI agent autonomous thoughts, emotions, and spontaneous impulses.

## What It Does

MindCore is a standalone background daemon that simulates a **subconscious
mind**. It rolls dice every second, modeling the random emergence of thoughts
like *"I want milk tea"*, *"I'm bored"*, or *"I suddenly want to chat"*.

When a thought's probability accumulates past the firing threshold, the engine
outputs a JSON signal telling your AI Agent: **"I have something to say."**

## Architecture

```
Layer 0: Noise Generators (3000 nodes)
    ├── Pink Noise (1/f, long-range correlation)
    ├── Ornstein-Uhlenbeck (physiological baseline)
    ├── Hawkes Process (emotional chain reaction)
    └── Markov Chain (attention drift)
         ↓
Layer 1: Sensor Layer (150 sensors)
    ├── Body State (hunger/fatigue/bio-rhythms)
    ├── Environment (time/weather/noise)
    └── Social Context (interaction/neglect)
         ↓
Layer 2: Impulse Emergence (150 impulse nodes)
    ├── Synapse Matrix (sensor → impulse mapping)
    ├── Sigmoid Probability + Mood Modulation
    └── Dice Roll → Random Firing
         ↓
Layer 3: Personality Gate (Softmax Sampling)
    ├── Learnable Personality Weights
    └── Short-Term Memory Topic Boost
         ↓
Layer 4: Output Template → JSON signal
```

## Quick Start

```bash
# Install dependencies
pip install -r requirements.txt

# Start the engine
python main.py
```

Requires Python 3.8+. On first run, automatically downloads `all-MiniLM-L6-v2`
local NLP model (~80MB) for synapse matrix generation.

## Key Features

- **150 Daily Impulses** across 9 categories (food, social, entertainment, etc.)
- **Stochastic, Not Scheduled** — Pink Noise + Hawkes Process + Sigmoid probability
- **Circadian Rhythms** — real clock-driven hunger/thirst/sleep cycles
- **Short-Term Memory** — 5-slot FIFO buffer with 2-hour exponential decay
- **Mood Baseline** — continuous valence modulation of impulse probability
- **Tunable Frequency** — single `BURST_BASE_OFFSET` parameter controls activity

## Integration

MindCore outputs standard JSON and is designed for [OpenClaw](https://openclaw.ai)
but compatible with any AI Agent framework that supports external signal injection.

See `references/INTEGRATION.md` for detailed integration guide.

## File Structure

- `main.py` — Entry point and engine loop
- `engine/` — Core 5-layer pipeline implementation
- `engine_supervisor.py` — Process supervisor for daemon mode
- `data/` — Runtime data (sensor state, synapse matrix, memory)
- `js_bridge/` — JavaScript bridge for OpenClaw integration

## License

AGPL-3.0 (commercial licensing available — contact zmliu0208@gmail.com)

---

## 中文说明

# MindCore —— 仿生潜意识引擎

> 为你的 AI 代理赋予自主的思想、情感和自发的冲动。

## 它的功能

MindCore 是一个独立的后台守护进程，用于模拟**潜意识**。它每秒掷一次骰子，对各类念头的随机涌现进行建模，例如 *“我想喝奶茶”*、*“我好无聊”* 或 *“我突然想聊天”*。

当某个念头的概率累积超过触发阈值时，引擎会输出一个 JSON 信号，告知你的 AI Agent：**“我有话要说。”**

## 架构

```
Layer 0: Noise Generators (3000 nodes)
    ├── Pink Noise (1/f, long-range correlation)
    ├── Ornstein-Uhlenbeck (physiological baseline)
    ├── Hawkes Process (emotional chain reaction)
    └── Markov Chain (attention drift)
         ↓
Layer 1: Sensor Layer (150 sensors)
    ├── Body State (hunger/fatigue/bio-rhythms)
    ├── Environment (time/weather/noise)
    └── Social Context (interaction/neglect)
         ↓
Layer 2: Impulse Emergence (150 impulse nodes)
    ├── Synapse Matrix (sensor → impulse mapping)
    ├── Sigmoid Probability + Mood Modulation
    └── Dice Roll → Random Firing
         ↓
Layer 3: Personality Gate (Softmax Sampling)
    ├── Learnable Personality Weights
    └── Short-Term Memory Topic Boost
         ↓
Layer 4: Output Template → JSON signal
```

## 快速开始

```bash
# Install dependencies
pip install -r requirements.txt

# Start the engine
python main.py
```

需要 Python 3.8+。首次运行时，会自动下载本地 NLP 模型 `all-MiniLM-L6-v2`（约 80MB），用于生成突触矩阵。

## 核心特性

- **150 种日常冲动**，涵盖 9 个类别（饮食、社交、娱乐等）
- **随机触发，而非定时调度** —— Pink Noise + Hawkes Process + Sigmoid 概率
- **昼夜节律** —— 由真实时钟驱动的饥饿/口渴/睡眠周期
- **短期记忆** —— 5 个槽位的 FIFO 缓冲，2 小时指数衰减
- **情绪基线** —— 对冲动概率进行连续的效价调制
- **可调频率** —— 单个 `BURST_BASE_OFFSET` 参数控制活跃度

## 集成

MindCore 输出标准 JSON，专为 [OpenClaw](https://openclaw.ai) 设计，但兼容任何支持外部信号注入的 AI Agent 框架。

详细的集成指南请参阅 `references/INTEGRATION.md`。

## 文件结构

- `main.py` —— 入口点和引擎循环
- `engine/` —— 核心 5 层管道实现
- `engine_supervisor.py` —— 守护进程模式的进程监管器
- `data/` —— 运行时数据（传感器状态、突触矩阵、记忆）
- `js_bridge/` —— 用于 OpenClaw 集成的 JavaScript 桥接

## 许可证

AGPL-3.0（提供商业授权 —— 联系 zmliu0208@gmail.com）