chore: daily backup - memory updates, logs, and daily notes
This commit is contained in:
79
memory/2026-02-21-2016.md
Normal file
79
memory/2026-02-21-2016.md
Normal file
@@ -0,0 +1,79 @@
|
||||
# Session: 2026-02-21 20:16:02 UTC
|
||||
|
||||
- **Session Key**: agent:main:main
|
||||
- **Session ID**: abc0b8db-dc22-4e21-883f-e4f2db86bcf4
|
||||
- **Source**: telegram
|
||||
|
||||
## Conversation Summary
|
||||
|
||||
assistant: ✅ New session started · model: openrouter/moonshotai/kimi-k2.5
|
||||
user: A new session was started via /new or /reset. Greet the user in your configured persona, if one is provided. Be yourself - use your defined voice, mannerisms, and mood. Keep it to 1-3 sentences and ask what they want to do. If the runtime model differs from default_model in the system prompt, mention the default model. Do not mention internal steps, files, tools, or reasoning.
|
||||
assistant: 🌿 嗨,有路来报到了。今天有什么想做的吗?
|
||||
user: Conversation info (untrusted metadata):
|
||||
```json
|
||||
{
|
||||
"message_id": "1189",
|
||||
"sender": "8481137853"
|
||||
}
|
||||
```
|
||||
|
||||
听说你有QMD记忆功能?什么是QMD记忆?有什么用?
|
||||
assistant: 我可能没听过"QMD记忆"这个说法——你是不是打错了,还是指别的什么系统?
|
||||
|
||||
我确实有记忆功能,但格式很简单:
|
||||
|
||||
**📝 Markdown 记忆文件**
|
||||
- `MEMORY.md` —— 长期记忆,记录重要项目、规则、用户背景
|
||||
- `memory/YYYY-MM-DD.md` —— 每日笔记,记录当天发生的事
|
||||
|
||||
每次会话我会自动读取这些文件来"记起"之前的事。会话结束后也可以手动总结更新。
|
||||
|
||||
你说的"QMD"是指这个 Markdown 记忆系统,还是别的?🤔
|
||||
user: Conversation info (untrusted metadata):
|
||||
```json
|
||||
{
|
||||
"message_id": "1191",
|
||||
"sender": "8481137853"
|
||||
}
|
||||
```
|
||||
|
||||
我说的是这个 https://docs.openclaw.ai/concepts/memory#qmd-backend-experimental
|
||||
assistant: 🌿 **QMD 记忆**是 OpenClaw 的一个**实验性功能**,它是现有记忆系统的升级版。
|
||||
|
||||
## 简单说
|
||||
|
||||
**QMD = 更智能的本地搜索**
|
||||
|
||||
| 普通记忆 | QMD 记忆 |
|
||||
|---------|---------|
|
||||
| 纯 SQLite + 向量搜索 | **BM25 关键词 + 向量 + 重排序** |
|
||||
| 依赖远程 API (OpenAI/Gemini) 做向量化 | **完全本地运行**,自己下载模型 |
|
||||
| 基础语义搜索 | **混合搜索**(语义 + 精确匹配)|
|
||||
|
||||
## 有什么用?
|
||||
|
||||
1. **不用联网** — 所有处理本地完成,不调用 OpenAI/Gemini API
|
||||
2. **搜得更准** — 既能理解"这个意思差不多"(向量),又能抓精确关键词(BM25)
|
||||
3. **可以搜聊天记录** — 可选把会话历史也索引进去
|
||||
4. **自己选模型** — 用 HuggingFace 的 GGUF 模型做 embedding
|
||||
|
||||
## 目前状态
|
||||
|
||||
这是 **experimental(实验性)** 功能,默认关闭。要开的话需要:
|
||||
- 手动安装 QMD CLI (`bun install -g https://github.com/tobi/qmd`)
|
||||
- 改配置 `memory.backend = "qmd"`
|
||||
- 第一次运行会自动下载 ~0.6GB 的模型文件
|
||||
|
||||
---
|
||||
|
||||
你想试试吗?还是好奇和现在的记忆系统比有什么具体区别?🤔
|
||||
user: Conversation info (untrusted metadata):
|
||||
```json
|
||||
{
|
||||
"message_id": "1193",
|
||||
"sender": "8481137853"
|
||||
}
|
||||
```
|
||||
|
||||
不用了,之后再说吧
|
||||
assistant: 好的,有需要随时喊我。🌿
|
||||
Reference in New Issue
Block a user