Ombre-Brain

概述

Ombre Brain为Claude提供跨对话的持久性情感记忆。它使用Russell环形情感模型为记忆标记效价和唤醒度坐标，将记忆存储为Obsidian兼容的Markdown文件，并实现了自然衰减的遗忘曲线，其中高情绪强度的记忆衰减更慢。系统包含五个MCP工具：breath用于记忆浮现和搜索、hold用于带自动标签的记忆存储、grow用于日记归档和拆分、trace用于修改和删除、pulse用于系统状态检查。在API不可用时，它会优雅地降级到本地关键词分析。

试试问 AI

装完之后，这里有 3 个你可以让 AI 做的事：

• breath

  Surface or search memories

  试试:Show me unresolved memories试试:Search for memories about our last conversation
• hold

  Store a single memory with auto-tagging

  试试:Remember that we talked about my vacation plans试试:Store this important detail about my project
• grow

  Diary digest, auto-split into multiple buckets

  试试:Archive our conversation from today试试:Process this long chat into memories
• trace

  Modify metadata, mark resolved, delete

  试试:Mark this memory as resolved试试:Update the importance of this memory试试:Delete this old memory
• pulse

  System status + all memory buckets listing

  试试:Show me system status试试:List all my memory buckets试试:What's the current state of my memory system

可对比工具

安装

git clone https://github.com/P0lar1zzZ/Ombre-Brain.git
cd Ombre-Brain
python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
pip install -r requirements.txt
cp config.example.yaml config.yaml

{
  "mcpServers": {
    "ombre-brain": {
      "command": "python",
      "args": ["/path/to/Ombre-Brain/server.py"],
      "env": {
        "OMBRE_API_KEY": "your-api-key"
      }
    }
  }
}

Ombre-Brain 对比