Compare commits
14 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 150e2d5607 | |||
| 38941ff2b5 | |||
| 53d6940994 | |||
| 16327dec72 | |||
| 9e2f390332 | |||
| 56e787a5a8 | |||
| 13e3c662ac | |||
| e8c275c8f8 | |||
| 931617624e | |||
| ee2fd18fec | |||
| aa18c1c8b1 | |||
| d05bb5507f | |||
| 904661d73f | |||
| 43172e257a |
6
.gitignore
vendored
6
.gitignore
vendored
@ -64,3 +64,9 @@ jimeng*.png
|
||||
|
||||
# Test Cache
|
||||
.pytest_cache/
|
||||
|
||||
# TypeScript
|
||||
node_modules/
|
||||
dist/
|
||||
*.tsbuildinfo
|
||||
tsconfig.tsbuildinfo
|
||||
|
||||
292
README.md
292
README.md
@ -1,109 +1,249 @@
|
||||
# TrulyMEM - TrueHumanMEM
|
||||
# TrulyMEM - WaterFlow 适配版
|
||||
|
||||
<p align="center">
|
||||
<img src="pic/image.png" alt="TrulyMEM Logo" width="200">
|
||||
</p>
|
||||
让 AI 拥有真正的长期记忆能力 - WaterFlow 框架适配版
|
||||
|
||||
> **📜 开源协议**: [GNU General Public License v3.0 (GPLv3)](https://www.gnu.org/licenses/gpl-3.0)
|
||||
> 本项目自由开源，可自由使用、修改和分发，但修改后的作品必须以相同许可证发布。
|
||||
|
||||
> **English**: [Switch to English version](./README_EN.md)
|
||||
|
||||
**让 AI 拥有自知、可塑、有分寸感的长期记忆**
|
||||
|
||||
*The More Human Choice.*
|
||||
|
||||
[](https://www.gnu.org/licenses/gpl-3.0)
|
||||
[](https://www.python.org/downloads/)
|
||||
[]()
|
||||
[English Version](./README_EN.md)
|
||||
|
||||
---
|
||||
|
||||
## 故事的开头
|
||||
## 简介
|
||||
|
||||
行业普遍认为，LLM 海量参数让其涌现了智能。但这个智能是「死的」——它不会真的记住，也不理解「记住」的概念。它输出的一切，都是当前输入的全部文本经历无数次前向传播计算出的概率最优解。LLM 不会因为某次对话意识到错误而去修正权重，也无法因此针对模型进行一次反向传播。它的意识是被冻结的，展现出的智能只是冻结的意识的回响。
|
||||
本项目是将 TrulyMEM 的图记忆能力迁移到 WaterFlow 框架的 TypeScript 实现。
|
||||
|
||||
现在的所谓记忆系统，只是将记忆外化，让「系统」去替 LLM 记住。或者就是粗暴地将一切上下文文本丢给 LLM。这就是对模型输入的浪费。
|
||||
|
||||
**TrulyMEM 想，既然 LLM 无法实时纠正模型权重，为什么不把记忆权交还给 LLM 呢？**
|
||||
|
||||
我们提供一系列机制，让 LLM 决定它要记住什么、遗忘，什么是重点、什么是糟粕。LLM 推理的过程，就是思考的过程，也是回忆的过程。完全摒弃传统的 messages 数组上下文，将全部记忆以**三元组（图）**的形式保存在图数据库中。在 LLM 思考时，可以按照图数据库的链接自主跳转、联想相关关系，让 LLM 自然地实现联想与回忆。
|
||||
|
||||
赋予 LLM 真正的记忆。
|
||||
作为 WaterFlow 的内置模块，提供图记忆功能：
|
||||
- **recall**: 检索记忆
|
||||
- **commit**: 写入记忆
|
||||
- **purge**: 删除记忆
|
||||
- **introspect**: 查看状态
|
||||
- **persona_update/clear**: 人设管理
|
||||
- **task_create/set_state/delete**: 任务管理
|
||||
|
||||
---
|
||||
|
||||
## 快速开始
|
||||
## 目录结构
|
||||
|
||||
### 方式一：打包后的可执行文件
|
||||
|
||||
```bash
|
||||
# Windows: TrulyMEM.exe
|
||||
# Linux/macOS: TrulyMEM
|
||||
chmod +x TrulyMEM
|
||||
./TrulyMEM
|
||||
```
|
||||
ts/
|
||||
├── src/runtime/core/
|
||||
│ ├── graph_memory/ # 图记忆核心模块
|
||||
│ │ ├── types.ts # 类型定义
|
||||
│ │ ├── graph_database.ts # 图数据库
|
||||
│ │ ├── memory_service.ts # 记忆服务
|
||||
│ │ └── index.ts # 模块导出
|
||||
│ └── tools/
|
||||
│ └── builtin/
|
||||
│ └── graph_memory_tool.ts # Tool 实现
|
||||
│
|
||||
├── bundled-skills/ # Skill 定义
|
||||
│ └── graph_memory/
|
||||
│ ├── SKILL.md # 记忆操作
|
||||
│ ├── persona/SKILL.md # 人设管理
|
||||
│ └── task/SKILL.md # 任务管理
|
||||
│
|
||||
├── package.json # 项目配置
|
||||
└── tsconfig.json # TypeScript 配置
|
||||
```
|
||||
|
||||
### 方式二：从源码运行
|
||||
---
|
||||
|
||||
## 在 WaterFlow 中使用
|
||||
|
||||
本模块完全不动 WaterFlow 源码，只需在你的入口文件中注册即可。
|
||||
|
||||
### 快速开始（推荐）
|
||||
|
||||
#### 步骤 1：安装依赖
|
||||
|
||||
```bash
|
||||
git clone <repo-url>
|
||||
cd TrulyMEM-TrueHumanMEM
|
||||
|
||||
pip install -r requirements.txt
|
||||
|
||||
python trulymem_entry.py
|
||||
npm install /path/to/TrulyMEM-TrueHumanMEM/ts
|
||||
```
|
||||
|
||||
### 配置
|
||||
或在 `package.json` 中添加：
|
||||
|
||||
1. 按 **F2** 展开侧边栏
|
||||
2. 输入 **API Key**（支持 DeepSeek、OpenAI 等兼容 API）
|
||||
3. 按 **Enter** 保存配置
|
||||
4. 开始对话！
|
||||
```json
|
||||
{
|
||||
"dependencies": {
|
||||
"trulymem-waterflow": "file:../TrulyMEM-TrueHumanMEM/ts"
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
然后运行：
|
||||
|
||||
```bash
|
||||
npm install
|
||||
```
|
||||
|
||||
#### 步骤 2：在你的入口文件中注册
|
||||
|
||||
只需两行代码，完全不动 WaterFlow 源码：
|
||||
|
||||
```typescript
|
||||
import { getPlatform } from 'waterflow-ts/dist/platform/index.js';
|
||||
import { installTrulyMEM } from 'trulymem/tools';
|
||||
|
||||
// 一行安装，返回配置好的 ToolRegistry
|
||||
const registry = await installTrulyMEM(getPlatform(), 'my-session-id');
|
||||
|
||||
// 继续组装 WaterFlow...
|
||||
const toolExecutor = new ToolExecutor(registry);
|
||||
```
|
||||
|
||||
### 手动注册（更灵活）
|
||||
|
||||
如果你想自己控制 ToolRegistry 的创建：
|
||||
|
||||
```typescript
|
||||
import { getPlatform } from 'waterflow-ts/dist/platform/index.js';
|
||||
import { initializeToolRegistry } from 'waterflow-ts/dist/runtime/core/tools/builtin/index.js';
|
||||
import { registerGraphMemoryTool } from 'trulymem/tools';
|
||||
|
||||
const platform = getPlatform();
|
||||
const registry = initializeToolRegistry(platform);
|
||||
|
||||
// 注册图记忆工具
|
||||
registerGraphMemoryTool(registry, 'my-session-id');
|
||||
|
||||
// 继续组装...
|
||||
```
|
||||
|
||||
### 使用 Skill（AI Agent 调用）
|
||||
|
||||
#### 步骤 1：配置 Skill 来源
|
||||
|
||||
```typescript
|
||||
const config = {
|
||||
...DEFAULT_SKILL_LOADER_CONFIG,
|
||||
sources: {
|
||||
...DEFAULT_SKILL_LOADER_CONFIG.sources,
|
||||
bundled: './node_modules/trulymem-waterflow/bundled-skills'
|
||||
},
|
||||
enabledSources: ['project', 'bundled']
|
||||
};
|
||||
```
|
||||
|
||||
#### 步骤 2：通过 Agent 调用
|
||||
|
||||
AI Agent 会自动读取 SKILL.md 并调用 `builtin:graph_memory` 工具。
|
||||
|
||||
#### 可用 Skill 列表
|
||||
|
||||
| Skill 名称 | 功能 | 使用场景 |
|
||||
|------------|------|----------|
|
||||
| `graph_memory` | 记忆 CRUD | 读取/写入/删除记忆 |
|
||||
| `persona` | 人设管理 | 设置 AI 角色性格 |
|
||||
| `task` | 任务管理 | 创建/更新长期任务 |
|
||||
|
||||
#### Skill 定义格式说明
|
||||
|
||||
WaterFlow 的 SkillLoader 会从 `SKILL.md` 中提取：
|
||||
|
||||
- **name**: 从目录名提取（如 `graph_memory`、`persona`、`task`）
|
||||
- **description**: 从 Markdown 的第一个 `#` 标题提取
|
||||
- **allowed-tools**: 转换为 `allowedTools` 字段
|
||||
- **arguments**: 正确映射到 SkillDefinition.arguments
|
||||
- **user-invocable**: 转换为 `userInvocable` 字段
|
||||
|
||||
**注意**: `when_to_use` 信息已整合到 Markdown body 中，通过 SkillRegistry.search() 可匹配。
|
||||
|
||||
---
|
||||
|
||||
## 文档索引
|
||||
## API
|
||||
|
||||
详细技术文档请参阅 [docs/zh/](docs/zh/) 目录：
|
||||
### GraphMemoryTool
|
||||
|
||||
| 文档 | 内容 |
|
||||
|------|------|
|
||||
| [docs/zh/architecture.md](docs/zh/architecture.md) | 系统架构和技术设计 |
|
||||
| [docs/zh/quick_start.md](docs/zh/quick_start.md) | 完整启动指南与配置说明 |
|
||||
| [docs/zh/memory.md](docs/zh/memory.md) | 内部记忆工作机制 |
|
||||
| [docs/zh/persona.md](docs/zh/persona.md) | 人设图机制 |
|
||||
| [docs/zh/working_memory.md](docs/zh/working_memory.md) | 连续性任务处理机制 |
|
||||
| [docs/zh/api.md](docs/zh/api.md) | 后端 API 接口（供扩展开发） |
|
||||
| [docs/zh/prompts.md](docs/zh/prompts.md) | 提示词管理模块 |
|
||||
```typescript
|
||||
const tool = new GraphMemoryTool(sessionId?: string);
|
||||
```
|
||||
|
||||
#### Actions
|
||||
|
||||
| Action | 说明 | 参数 |
|
||||
|--------|------|------|
|
||||
| `recall` | 检索记忆 | `queryIntent`, `seedEntities`, `depth`, `sessionFilter` |
|
||||
| `commit` | 写入记忆 | `triplets`, `sessionId`, `turnId` |
|
||||
| `purge` | 删除记忆 | `criteria`, `mode` |
|
||||
| `introspect` | 查看状态 | - |
|
||||
| `persona_update` | 更新人设 | `attributes`, `mode` |
|
||||
| `persona_clear` | 清除人设 | `confirm` |
|
||||
| `task_create` | 创建任务 | `task_id`, `description`, `info_nodes` |
|
||||
| `task_set_state` | 设置状态 | `task_id`, `state` |
|
||||
| `task_delete` | 删除任务 | `task_id` |
|
||||
| `task_link_info` | 关联信息到任务 | `task_id`, `info_node` |
|
||||
|
||||
---
|
||||
|
||||
## 贡献
|
||||
## 示例
|
||||
|
||||
欢迎提交 Issue 和 Pull Request！
|
||||
### 写入记忆
|
||||
|
||||
1. Fork 本仓库
|
||||
2. 创建特性分支 (`git checkout -b feature/AmazingFeature`)
|
||||
3. 提交更改 (`git commit -m 'Add some AmazingFeature'`)
|
||||
4. 推送到分支 (`git push origin feature/AmazingFeature`)
|
||||
5. 创建 Pull Request
|
||||
```json
|
||||
{
|
||||
"action": "commit",
|
||||
"params": {
|
||||
"triplets": [
|
||||
{ "subject": "用户", "relation": "喜欢", "object": "TypeScript" },
|
||||
{ "subject": "用户", "relation": "正在学习", "object": "WaterFlow" }
|
||||
]
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
### 检索记忆
|
||||
|
||||
```json
|
||||
{
|
||||
"action": "recall",
|
||||
"params": {
|
||||
"queryIntent": "用户 学习"
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
### 创建任务
|
||||
|
||||
```json
|
||||
{
|
||||
"action": "task_create",
|
||||
"params": {
|
||||
"task_id": "Task_学习TypeScript",
|
||||
"description": "学习 TypeScript 并完成项目",
|
||||
"info_nodes": ["文档链接", "教程链接"]
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
### 关联信息到任务
|
||||
|
||||
```json
|
||||
{
|
||||
"action": "task_link_info",
|
||||
"params": {
|
||||
"task_id": "Task_学习TypeScript",
|
||||
"info_node": "用户喜欢 React"
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## API 名称映射
|
||||
|
||||
OpenAI/DeepSeek API 要求工具名称符合 `^[a-zA-Z0-9_-]+$` 格式（不含冒号）。
|
||||
内部工具 ID 使用 `builtin:xxx` 格式，需映射后发送给 API。
|
||||
|
||||
```typescript
|
||||
import { mapToolIdToApiName, mapApiNameToToolId } from 'trulymem/tools';
|
||||
|
||||
// 发送给 API
|
||||
const apiName = mapToolIdToApiName('builtin:graph_memory'); // -> 'graph_memory'
|
||||
|
||||
// 收到 tool_use 后映射回
|
||||
const internalId = mapApiNameToToolId('graph_memory'); // -> 'builtin:graph_memory'
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 许可证
|
||||
|
||||
本项目采用 **GNU General Public License v3.0 (GPLv3)** 许可证开源。
|
||||
详见 [LICENSE](LICENSE) 文件。
|
||||
|
||||
---
|
||||
|
||||
## 特别鸣谢
|
||||
|
||||
- [Prof. Meiting Wang](https://www.xxmu.edu.cn/yxgcxy/info/1260/4252.htm) — 学术指导
|
||||
- [逝水秋生白](https://atomgit.com/cenber) — 架构支持
|
||||
- anzhitinglan — 测试资源支持
|
||||
- 崔莉萍老师 — 理论指导
|
||||
- Annie — 专业指导
|
||||
- 王梓沣、马悦华、隆梦婷 — 神经科学理论支持
|
||||
[GNU General Public License v3.0 (GPLv3)](LICENSE)
|
||||
|
||||
291
README_EN.md
291
README_EN.md
@ -1,109 +1,250 @@
|
||||
# TrulyMEM - TrueHumanMEM
|
||||
# TrulyMEM - WaterFlow Adapter
|
||||
|
||||
<p align="center">
|
||||
<img src="pic/image.png" alt="TrulyMEM Logo" width="200">
|
||||
</p>
|
||||
Give AI true long-term memory capability - WaterFlow framework adapter version
|
||||
|
||||
> **📜 License**: [GNU General Public License v3.0 (GPLv3)](https://www.gnu.org/licenses/gpl-3.0)
|
||||
> This project is free and open source. You are free to use, modify, and distribute, but modified works must be distributed under the same license.
|
||||
|
||||
> **中文**: [切换到中文版](./README.md)
|
||||
|
||||
**Give AI self-awareness, plasticity, and a sense of proportion in long-term memory**
|
||||
|
||||
*The More Human Choice.*
|
||||
|
||||
[](https://www.gnu.org/licenses/gpl-3.0)
|
||||
[](https://www.python.org/downloads/)
|
||||
[]()
|
||||
[中文版本](./README.md)
|
||||
|
||||
---
|
||||
|
||||
## The Story
|
||||
## Introduction
|
||||
|
||||
Industry believes that LLMs' massive parameters give them emergent intelligence. But this intelligence is "dead" — it cannot truly remember, nor understand the concept of "remembering". Everything it outputs is the probabilistic optimal solution calculated through countless forward passes on the current input text. The LLM cannot correct its weights based on errors in a conversation, nor perform a backward pass. Its consciousness is frozen — what appears as intelligence is merely the echo of this frozen consciousness.
|
||||
This project ports TrulyMEM's graph memory capability to TypeScript for the WaterFlow framework.
|
||||
|
||||
Current "memory systems" merely externalize memory, letting the "system" remember for the LLM. Or they dump all context text to the LLM. This is a waste of the model's limited input context.
|
||||
As a built-in module for WaterFlow, it provides graph memory functionality:
|
||||
|
||||
**TrulyMEM asks: since the LLM cannot correct model weights in real-time, why not give the memory authority back to the LLM?**
|
||||
|
||||
We provide a series of mechanisms for the LLM to decide what to remember, what to forget, what's important, what's trivial. The LLM's reasoning process is also its thinking and recalling process. Abandoning the traditional messages array context, all memories are stored as **triplets (graph)** in the graph database. When the LLM thinks, it can autonomously jump through graph links to associate related relationships, enabling natural association and recall.
|
||||
|
||||
Give the LLM true memory.
|
||||
- **recall**: Retrieve memories
|
||||
- **commit**: Commit memories
|
||||
- **purge**: Delete memories
|
||||
- **introspect**: Inspect status
|
||||
- **persona_update/clear**: Persona management
|
||||
- **task_create/set_state/delete**: Task management
|
||||
|
||||
---
|
||||
|
||||
## Quick Start
|
||||
## Directory Structure
|
||||
|
||||
### Method 1: Run Packaged Executable
|
||||
|
||||
```bash
|
||||
# Windows: TrulyMEM.exe
|
||||
# Linux/macOS: TrulyMEM
|
||||
chmod +x TrulyMEM
|
||||
./TrulyMEM
|
||||
```
|
||||
ts/
|
||||
├── src/runtime/core/
|
||||
│ ├── graph_memory/ # Graph memory core module
|
||||
│ │ ├── types.ts # Type definitions
|
||||
│ │ ├── graph_database.ts # Graph database
|
||||
│ │ ├── memory_service.ts # Memory service
|
||||
│ │ └── index.ts # Module exports
|
||||
│ └── tools/
|
||||
│ └── builtin/
|
||||
│ └── graph_memory_tool.ts # Tool implementation
|
||||
│
|
||||
├── bundled-skills/ # Skill definitions
|
||||
│ └── graph_memory/
|
||||
│ ├── SKILL.md # Memory operations
|
||||
│ ├── persona/SKILL.md # Persona management
|
||||
│ └── task/SKILL.md # Task management
|
||||
│
|
||||
├── package.json # Project config
|
||||
└── tsconfig.json # TypeScript config
|
||||
```
|
||||
|
||||
### Method 2: Run from Source
|
||||
---
|
||||
|
||||
## Usage in WaterFlow
|
||||
|
||||
This module requires **zero changes** to WaterFlow source code. Just register it in your entry file.
|
||||
|
||||
### Quick Start (Recommended)
|
||||
|
||||
#### Step 1: Install
|
||||
|
||||
```bash
|
||||
git clone <repo-url>
|
||||
cd TrulyMEM-TrueHumanMEM
|
||||
|
||||
pip install -r requirements.txt
|
||||
|
||||
python trulymem_entry.py
|
||||
npm install /path/to/TrulyMEM-TrueHumanMEM/ts
|
||||
```
|
||||
|
||||
### Configuration
|
||||
Or add to `package.json`:
|
||||
|
||||
1. Press **F2** to expand sidebar
|
||||
2. Enter **API Key** (supports DeepSeek, OpenAI, etc.)
|
||||
3. Press **Enter** to save
|
||||
4. Start chatting!
|
||||
```json
|
||||
{
|
||||
"dependencies": {
|
||||
"trulymem-waterflow": "file:../TrulyMEM-TrueHumanMEM/ts"
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
Then run:
|
||||
|
||||
```bash
|
||||
npm install
|
||||
```
|
||||
|
||||
#### Step 2: Register in your entry file
|
||||
|
||||
Just two lines, zero changes to WaterFlow:
|
||||
|
||||
```typescript
|
||||
import { getPlatform } from 'waterflow-ts/dist/platform/index.js';
|
||||
import { installTrulyMEM } from 'trulymem/tools';
|
||||
|
||||
// One-line install, returns configured ToolRegistry
|
||||
const registry = await installTrulyMEM(getPlatform(), 'my-session-id');
|
||||
|
||||
// Continue assembling WaterFlow...
|
||||
const toolExecutor = new ToolExecutor(registry);
|
||||
```
|
||||
|
||||
### Manual Registration (More control)
|
||||
|
||||
If you want to control ToolRegistry creation yourself:
|
||||
|
||||
```typescript
|
||||
import { getPlatform } from 'waterflow-ts/dist/platform/index.js';
|
||||
import { initializeToolRegistry } from 'waterflow-ts/dist/runtime/core/tools/builtin/index.js';
|
||||
import { registerGraphMemoryTool } from 'trulymem/tools';
|
||||
|
||||
const platform = getPlatform();
|
||||
const registry = initializeToolRegistry(platform);
|
||||
|
||||
// Register graph memory tool
|
||||
registerGraphMemoryTool(registry, 'my-session-id');
|
||||
|
||||
// Continue assembling...
|
||||
```
|
||||
|
||||
### Use Skill (AI Agent)
|
||||
|
||||
#### Step 1: Configure Skill source
|
||||
|
||||
```typescript
|
||||
const config = {
|
||||
...DEFAULT_SKILL_LOADER_CONFIG,
|
||||
sources: {
|
||||
...DEFAULT_SKILL_LOADER_CONFIG.sources,
|
||||
bundled: './node_modules/trulymem-waterflow/bundled-skills'
|
||||
},
|
||||
enabledSources: ['project', 'bundled']
|
||||
};
|
||||
```
|
||||
|
||||
#### Step 2: Call via Agent
|
||||
|
||||
AI Agent automatically reads SKILL.md and calls `builtin:graph_memory` tool.
|
||||
|
||||
#### Available Skills
|
||||
|
||||
| Skill Name | Function | Use Case |
|
||||
|------------|----------|----------|
|
||||
| `graph_memory` | Memory CRUD | Read/Write/Delete memories |
|
||||
| `persona` | Persona management | Set AI role/personality |
|
||||
| `task` | Task management | Create/update long-term tasks |
|
||||
|
||||
#### Skill Definition Format
|
||||
|
||||
WaterFlow's SkillLoader extracts from `SKILL.md`:
|
||||
|
||||
- **name**: Extracted from directory name (e.g., `graph_memory`, `persona`, `task`)
|
||||
- **description**: Extracted from first Markdown `#` heading
|
||||
- **allowed-tools**: Converted to `allowedTools` field
|
||||
- **arguments**: Properly mapped to SkillDefinition.arguments
|
||||
- **user-invocable**: Converted to `userInvocable` field
|
||||
|
||||
**Note**: `when_to_use` info is integrated into Markdown body, searchable via SkillRegistry.search().
|
||||
|
||||
---
|
||||
|
||||
## Documentation Index
|
||||
## API
|
||||
|
||||
Detailed technical documentation in the [docs/en/](docs/en/) directory:
|
||||
### GraphMemoryTool
|
||||
|
||||
| Document | Content |
|
||||
|----------|---------|
|
||||
| [docs/en/architecture.md](docs/en/architecture.md) | System architecture and technical design |
|
||||
| [docs/en/quick_start.md](docs/en/quick_start.md) | Complete startup guide and configuration |
|
||||
| [docs/en/memory.md](docs/en/memory.md) | Internal memory working mechanism |
|
||||
| [docs/en/persona.md](docs/en/persona.md) | Persona Graph mechanism |
|
||||
| [docs/en/working_memory.md](docs/en/working_memory.md) | Continuous task handling mechanism |
|
||||
| [docs/en/api.md](docs/en/api.md) | BackendServer API (for extension development) |
|
||||
| [docs/en/prompts.md](docs/en/prompts.md) | Prompt management module |
|
||||
```typescript
|
||||
const tool = new GraphMemoryTool(sessionId?: string);
|
||||
```
|
||||
|
||||
#### Actions
|
||||
|
||||
| Action | Description | Parameters |
|
||||
|--------|-------------|------------|
|
||||
| `recall` | Retrieve memories | `queryIntent`, `seedEntities`, `depth`, `sessionFilter` |
|
||||
| `commit` | Commit memories | `triplets`, `sessionId`, `turnId` |
|
||||
| `purge` | Delete memories | `criteria`, `mode` |
|
||||
| `introspect` | Inspect status | - |
|
||||
| `persona_update` | Update persona | `attributes`, `mode` |
|
||||
| `persona_clear` | Clear persona | `confirm` |
|
||||
| `task_create` | Create task | `task_id`, `description`, `info_nodes` |
|
||||
| `task_set_state` | Set state | `task_id`, `state` |
|
||||
| `task_delete` | Delete task | `task_id` |
|
||||
| `task_link_info` | Link info to task | `task_id`, `info_node` |
|
||||
|
||||
---
|
||||
|
||||
## Contributing
|
||||
## Examples
|
||||
|
||||
Welcome to submit Issues and Pull Requests!
|
||||
### Commit Memory
|
||||
|
||||
1. Fork this repository
|
||||
2. Create feature branch (`git checkout -b feature/AmazingFeature`)
|
||||
3. Commit changes (`git commit -m 'Add some AmazingFeature'`)
|
||||
4. Push to branch (`git push origin feature/AmazingFeature`)
|
||||
5. Create Pull Request
|
||||
```json
|
||||
{
|
||||
"action": "commit",
|
||||
"params": {
|
||||
"triplets": [
|
||||
{ "subject": "User", "relation": "likes", "object": "TypeScript" },
|
||||
{ "subject": "User", "relation": "is learning", "object": "WaterFlow" }
|
||||
]
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
### Recall Memory
|
||||
|
||||
```json
|
||||
{
|
||||
"action": "recall",
|
||||
"params": {
|
||||
"queryIntent": "User learning"
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
### Create Task
|
||||
|
||||
```json
|
||||
{
|
||||
"action": "task_create",
|
||||
"params": {
|
||||
"task_id": "Task_LearnTypeScript",
|
||||
"description": "Learn TypeScript and complete project",
|
||||
"info_nodes": ["Documentation", "Tutorial"]
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
### Link Info to Task
|
||||
|
||||
```json
|
||||
{
|
||||
"action": "task_link_info",
|
||||
"params": {
|
||||
"task_id": "Task_LearnTypeScript",
|
||||
"info_node": "User likes React"
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## API Name Mapping
|
||||
|
||||
OpenAI/DeepSeek API requires tool names to match `^[a-zA-Z0-9_-]+$` (no colons).
|
||||
Internal tool IDs use `builtin:xxx` format and must be mapped before sending to API.
|
||||
|
||||
```typescript
|
||||
import { mapToolIdToApiName, mapApiNameToToolId } from 'trulymem/tools';
|
||||
|
||||
// Send to API
|
||||
const apiName = mapToolIdToApiName('builtin:graph_memory'); // -> 'graph_memory'
|
||||
|
||||
// Map back after receiving tool_use
|
||||
const internalId = mapApiNameToToolId('graph_memory'); // -> 'builtin:graph_memory'
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## License
|
||||
|
||||
This project is licensed under the **GNU General Public License v3.0 (GPLv3)**.
|
||||
See [LICENSE](LICENSE) file for details.
|
||||
|
||||
---
|
||||
|
||||
## Special Thanks
|
||||
|
||||
- [Prof. Meiting Wang](https://www.xxmu.edu.cn/yxgcxy/info/1260/4252.htm) — Academic guidance
|
||||
- [逝水秋生白](https://atomgit.com/cenber) — Architecture support
|
||||
- anzhitinglan — Testing resource support
|
||||
- 崔莉萍老师 — Theoretical guidance
|
||||
- Annie — Professional guidance
|
||||
- 王梓沣、马悦华、隆梦婷 — Neuroscience theory support
|
||||
[GNU General Public License v3.0 (GPLv3)](LICENSE)
|
||||
|
||||
@ -1,45 +0,0 @@
|
||||
# -*- mode: python ; coding: utf-8 -*-
|
||||
from PyInstaller.utils.hooks import collect_all
|
||||
|
||||
datas = [('ui/styles', 'ui/styles'), ('core/prompts/templates', 'core/prompts/templates')]
|
||||
binaries = []
|
||||
hiddenimports = ['textual', 'textual.app', 'textual.widgets', 'textual.css', 'openai', 'openai._client', 'neo4j', 'sqlite3', 'core', 'core.embedded_db', 'core.graph_client', 'core.tool_executor', 'core.tool_limiter', 'core.tools', 'core.tools.memory_tools', 'core.prompts', 'core.prompts.prompt_manager', 'ui', 'ui.app', 'ui.models', 'ui.models.message', 'ui.models.config', 'ui.models.log_entry', 'ui.widgets', 'ui.handlers', 'ui.services', 'ui.services.config_manager', 'ui.services.config_service']
|
||||
tmp_ret = collect_all('textual')
|
||||
datas += tmp_ret[0]; binaries += tmp_ret[1]; hiddenimports += tmp_ret[2]
|
||||
|
||||
|
||||
a = Analysis(
|
||||
['trulymem_entry.py'],
|
||||
pathex=[],
|
||||
binaries=binaries,
|
||||
datas=datas,
|
||||
hiddenimports=hiddenimports,
|
||||
hookspath=[],
|
||||
hooksconfig={},
|
||||
runtime_hooks=[],
|
||||
excludes=[],
|
||||
noarchive=False,
|
||||
optimize=0,
|
||||
)
|
||||
pyz = PYZ(a.pure)
|
||||
|
||||
exe = EXE(
|
||||
pyz,
|
||||
a.scripts,
|
||||
a.binaries,
|
||||
a.datas,
|
||||
[],
|
||||
name='TrulyMEM',
|
||||
debug=False,
|
||||
bootloader_ignore_signals=False,
|
||||
strip=False,
|
||||
upx=True,
|
||||
upx_exclude=[],
|
||||
runtime_tmpdir=None,
|
||||
console=True,
|
||||
disable_windowed_traceback=False,
|
||||
argv_emulation=False,
|
||||
target_arch=None,
|
||||
codesign_identity=None,
|
||||
entitlements_file=None,
|
||||
)
|
||||
@ -1,114 +0,0 @@
|
||||
#!/bin/bash
|
||||
set -e
|
||||
echo "===== Building TrulyMEM AppImage for Linux ====="
|
||||
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
|
||||
PROJECT_ROOT="$(dirname "$SCRIPT_DIR")"
|
||||
cd "$PROJECT_ROOT"
|
||||
if ! command -v python3 &> /dev/null; then
|
||||
echo "Error: python3 not found"
|
||||
exit 1
|
||||
fi
|
||||
APPDIR="$PROJECT_ROOT/TrulyMEM.AppDir"
|
||||
rm -rf "$APPDIR"
|
||||
mkdir -p "$APPDIR/usr/bin"
|
||||
echo "===== Step 1: Build binary with PyInstaller ====="
|
||||
VENV_DIR="$PROJECT_ROOT/.venv_appimage"
|
||||
rm -rf "$VENV_DIR"
|
||||
python3 -m venv "$VENV_DIR"
|
||||
source "$VENV_DIR/bin/activate"
|
||||
pip install --upgrade pip
|
||||
pip install -r requirements.txt
|
||||
pip install pyinstaller
|
||||
rm -rf "$PROJECT_ROOT/build/pyinstaller_build" "$PROJECT_ROOT/dist"
|
||||
echo "Running PyInstaller..."
|
||||
pyinstaller trulymem_entry.py \
|
||||
--clean \
|
||||
--onefile \
|
||||
--console \
|
||||
--name TrulyMEM \
|
||||
--distpath "$PROJECT_ROOT/dist" \
|
||||
--workpath "$PROJECT_ROOT/build/pyinstaller_build" \
|
||||
--add-data "ui/styles:ui/styles" \
|
||||
--add-data "core/prompts/templates:core/prompts/templates" \
|
||||
--hidden-import textual \
|
||||
--hidden-import textual.app \
|
||||
--hidden-import textual.widgets \
|
||||
--hidden-import textual.css \
|
||||
--hidden-import openai \
|
||||
--hidden-import openai._client \
|
||||
--hidden-import neo4j \
|
||||
--hidden-import sqlite3 \
|
||||
--hidden-import core \
|
||||
--hidden-import core.embedded_db \
|
||||
--hidden-import core.graph_client \
|
||||
--hidden-import core.tool_executor \
|
||||
--hidden-import core.tool_limiter \
|
||||
--hidden-import core.tools \
|
||||
--hidden-import core.tools.memory_tools \
|
||||
--hidden-import core.prompts \
|
||||
--hidden-import core.prompts.prompt_manager \
|
||||
--hidden-import ui \
|
||||
--hidden-import ui.app \
|
||||
--hidden-import ui.models \
|
||||
--hidden-import ui.models.message \
|
||||
--hidden-import ui.models.config \
|
||||
--hidden-import ui.models.log_entry \
|
||||
--hidden-import ui.widgets \
|
||||
--hidden-import ui.handlers \
|
||||
--hidden-import ui.services \
|
||||
--hidden-import ui.services.config_manager \
|
||||
--hidden-import ui.services.config_service \
|
||||
--collect-all textual \
|
||||
--noconfirm
|
||||
cp "$PROJECT_ROOT/dist/TrulyMEM" "$APPDIR/usr/bin/"
|
||||
cp "$PROJECT_ROOT/trulymem_entry.py" "$APPDIR/usr/share/trulymem/"
|
||||
echo "===== Step 2: Create AppImage structure ====="
|
||||
cat > "$APPDIR/AppRun" << 'EOF'
|
||||
#!/bin/bash
|
||||
set -e
|
||||
SELF=$(readlink -f "$0")
|
||||
APPDIR=$(dirname "$SELF")
|
||||
export PATH="$APPDIR/usr/bin:$PATH"
|
||||
exec "$APPDIR/usr/bin/TrulyMEM" "$@"
|
||||
EOF
|
||||
chmod +x "$APPDIR/AppRun"
|
||||
cat > "$APPDIR/trulymem.desktop" << 'EOF'
|
||||
[Desktop Entry]
|
||||
Name=TrulyMEM
|
||||
Comment=AI Memory System with Long-term Memory
|
||||
Exec=TrulyMEM %U
|
||||
Icon=trulymem
|
||||
Terminal=true
|
||||
Type=Application
|
||||
Categories=Utility;AI;
|
||||
EOF
|
||||
[ -f "$PROJECT_ROOT/pic/TrulyMEM.png" ] && cp "$PROJECT_ROOT/pic/TrulyMEM.png" "$APPDIR/trulymem.png"
|
||||
APPIMAGE="$PROJECT_ROOT/TrulyMEM.AppImage"
|
||||
rm -f "$APPIMAGE"
|
||||
echo "===== Step 3: Package as AppImage ====="
|
||||
cd /tmp
|
||||
if ! command -v appimagetool &> /dev/null; then
|
||||
echo "Downloading appimagetool..."
|
||||
wget -q https://github.com/AppImage/AppImageKit/releases/download/continuous/appimagetool-x86_64.AppImage -O appimagetool 2>/dev/null || \
|
||||
curl -sL https://github.com/AppImage/AppImageKit/releases/download/continuous/appimagetool-x86_64.AppImage -o appimagetool
|
||||
chmod +x appimagetool 2>/dev/null || true
|
||||
fi
|
||||
cd "$PROJECT_ROOT"
|
||||
if [ -x /tmp/appimagetool ]; then
|
||||
/tmp/appimagetool "$APPDIR" "$APPIMAGE" || {
|
||||
echo "appimagetool failed, keeping AppDir for manual packaging"
|
||||
}
|
||||
elif command -v appimagetool &> /dev/null; then
|
||||
appimagetool "$APPDIR" "$APPIMAGE"
|
||||
else
|
||||
echo "Warning: appimagetool not available"
|
||||
echo "AppDir created at: $APPDIR"
|
||||
echo "You can manually run: appimagetool $APPDIR $APPIMAGE"
|
||||
fi
|
||||
echo "===== Build Complete ====="
|
||||
[ -f "$APPIMAGE" ] && echo "AppImage: $APPIMAGE" && ls -la "$APPIMAGE"
|
||||
[ -d "$APPDIR" ] && echo "AppDir: $APPDIR (can be packaged manually with appimagetool)"
|
||||
deactivate
|
||||
rm -rf "$VENV_DIR"
|
||||
rm -f /tmp/appimagetool
|
||||
echo "Done!"
|
||||
@ -1,83 +0,0 @@
|
||||
#!/bin/bash
|
||||
set -e
|
||||
|
||||
echo "===== Building TrulyMEM for Linux ====="
|
||||
|
||||
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
|
||||
PROJECT_ROOT="$(dirname "$SCRIPT_DIR")"
|
||||
cd "$PROJECT_ROOT"
|
||||
|
||||
echo "Project root: $PROJECT_ROOT"
|
||||
|
||||
if ! command -v python3 &> /dev/null; then
|
||||
echo "Error: python3 not found"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# 创建并激活虚拟环境
|
||||
VENV_DIR="$PROJECT_ROOT/.venv_build"
|
||||
|
||||
echo "Creating virtual environment: $VENV_DIR"
|
||||
python3 -m venv "$VENV_DIR"
|
||||
|
||||
echo "Activating virtual environment..."
|
||||
source "$VENV_DIR/bin/activate"
|
||||
|
||||
echo "Upgrading pip in virtual environment..."
|
||||
pip install --upgrade pip
|
||||
|
||||
echo "Installing dependencies in virtual environment..."
|
||||
pip install -r requirements.txt
|
||||
|
||||
echo "Cleaning previous builds..."
|
||||
rm -rf build/dist build/__pycache__ 2>/dev/null || true
|
||||
|
||||
echo "Running PyInstaller..."
|
||||
python -m PyInstaller trulymem_entry.py \
|
||||
--clean \
|
||||
--onefile \
|
||||
--console \
|
||||
--name TrulyMEM \
|
||||
--add-data "ui/styles:ui/styles" \
|
||||
--add-data "core/prompts/templates:core/prompts/templates" \
|
||||
--hidden-import textual \
|
||||
--hidden-import textual.app \
|
||||
--hidden-import textual.widgets \
|
||||
--hidden-import textual.css \
|
||||
--hidden-import openai \
|
||||
--hidden-import openai._client \
|
||||
--hidden-import neo4j \
|
||||
--hidden-import sqlite3 \
|
||||
--hidden-import core \
|
||||
--hidden-import core.embedded_db \
|
||||
--hidden-import core.graph_client \
|
||||
--hidden-import core.tool_executor \
|
||||
--hidden-import core.tool_limiter \
|
||||
--hidden-import core.tools \
|
||||
--hidden-import core.tools.memory_tools \
|
||||
--hidden-import core.prompts \
|
||||
--hidden-import core.prompts.prompt_manager \
|
||||
--hidden-import ui \
|
||||
--hidden-import ui.app \
|
||||
--hidden-import ui.models \
|
||||
--hidden-import ui.models.message \
|
||||
--hidden-import ui.models.config \
|
||||
--hidden-import ui.models.log_entry \
|
||||
--hidden-import ui.widgets \
|
||||
--hidden-import ui.handlers \
|
||||
--hidden-import ui.services \
|
||||
--hidden-import ui.services.config_manager \
|
||||
--hidden-import ui.services.config_service \
|
||||
--collect-all textual \
|
||||
--noconfirm
|
||||
|
||||
echo "===== Build Complete ====="
|
||||
echo "Binary: dist/TrulyMEM"
|
||||
ls -la dist/
|
||||
|
||||
# 清理虚拟环境
|
||||
echo "Cleaning up virtual environment..."
|
||||
deactivate
|
||||
rm -rf "$VENV_DIR"
|
||||
|
||||
echo "Build finished successfully!"
|
||||
@ -1,90 +0,0 @@
|
||||
#!/bin/bash
|
||||
set -e
|
||||
|
||||
echo "===== Building TrulyMEM for macOS ====="
|
||||
|
||||
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
|
||||
PROJECT_ROOT="$(dirname "$SCRIPT_DIR")"
|
||||
cd "$PROJECT_ROOT"
|
||||
|
||||
echo "Project root: $PROJECT_ROOT"
|
||||
|
||||
if ! command -v python3 &> /dev/null; then
|
||||
echo "Error: python3 not found"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# 创建并激活虚拟环境
|
||||
VENV_DIR="$PROJECT_ROOT/.venv_build"
|
||||
|
||||
echo "Creating virtual environment: $VENV_DIR"
|
||||
python3 -m venv "$VENV_DIR"
|
||||
|
||||
echo "Activating virtual environment..."
|
||||
source "$VENV_DIR/bin/activate"
|
||||
|
||||
echo "Upgrading pip in virtual environment..."
|
||||
pip install --upgrade pip
|
||||
|
||||
echo "Installing dependencies in virtual environment..."
|
||||
pip install -r requirements.txt
|
||||
|
||||
echo "Generating ICNS icon..."
|
||||
if [ -d "pic/TrulyMEM.iconset" ]; then
|
||||
iconutil -c icns pic/TrulyMEM.iconset -o pic/TrulyMEM.icns
|
||||
echo "ICNS icon generated: pic/TrulyMEM.icns"
|
||||
fi
|
||||
|
||||
echo "Cleaning previous builds..."
|
||||
rm -rf build/dist build/__pycache__ 2>/dev/null || true
|
||||
|
||||
echo "Running PyInstaller..."
|
||||
python -m PyInstaller trulymem_entry.py \
|
||||
--clean \
|
||||
--onefile \
|
||||
--console \
|
||||
--name TrulyMEM \
|
||||
--icon "pic/TrulyMEM.icns" \
|
||||
--add-data "ui/styles:ui/styles" \
|
||||
--add-data "core/prompts/templates:core/prompts/templates" \
|
||||
--hidden-import textual \
|
||||
--hidden-import textual.app \
|
||||
--hidden-import textual.widgets \
|
||||
--hidden-import textual.css \
|
||||
--hidden-import openai \
|
||||
--hidden-import openai._client \
|
||||
--hidden-import neo4j \
|
||||
--hidden-import sqlite3 \
|
||||
--hidden-import core \
|
||||
--hidden-import core.embedded_db \
|
||||
--hidden-import core.graph_client \
|
||||
--hidden-import core.tool_executor \
|
||||
--hidden-import core.tool_limiter \
|
||||
--hidden-import core.tools \
|
||||
--hidden-import core.tools.memory_tools \
|
||||
--hidden-import core.prompts \
|
||||
--hidden-import core.prompts.prompt_manager \
|
||||
--hidden-import ui \
|
||||
--hidden-import ui.app \
|
||||
--hidden-import ui.models \
|
||||
--hidden-import ui.models.message \
|
||||
--hidden-import ui.models.config \
|
||||
--hidden-import ui.models.log_entry \
|
||||
--hidden-import ui.widgets \
|
||||
--hidden-import ui.handlers \
|
||||
--hidden-import ui.services \
|
||||
--hidden-import ui.services.config_manager \
|
||||
--hidden-import ui.services.config_service \
|
||||
--collect-all textual \
|
||||
--noconfirm
|
||||
|
||||
echo "===== Build Complete ====="
|
||||
echo "Binary: dist/TrulyMEM"
|
||||
ls -la dist/
|
||||
|
||||
# 清理虚拟环境
|
||||
echo "Cleaning up virtual environment..."
|
||||
deactivate
|
||||
rm -rf "$VENV_DIR"
|
||||
|
||||
echo "Build finished successfully!"
|
||||
@ -1,60 +0,0 @@
|
||||
@echo off
|
||||
echo ===== Building TrulyMEM for Windows =====
|
||||
|
||||
REM 切换到脚本所在目录的上一级目录（项目根目录）
|
||||
cd /d "%~dp0.."
|
||||
echo Project root: %CD%
|
||||
|
||||
python --version >nul 2>&1
|
||||
if errorlevel 1 (
|
||||
echo Error: python not found
|
||||
exit /b 1
|
||||
)
|
||||
|
||||
echo Installing dependencies...
|
||||
pip install -r requirements.txt
|
||||
|
||||
echo Running PyInstaller...
|
||||
python -m PyInstaller trulymem_entry.py ^
|
||||
--clean ^
|
||||
--onefile ^
|
||||
--console ^
|
||||
--name TrulyMEM ^
|
||||
--icon "pic/TrulyMEM.ico" ^
|
||||
--add-data "ui/styles;ui/styles" ^
|
||||
--add-data "core/prompts/templates;core/prompts/templates" ^
|
||||
--hidden-import textual ^
|
||||
--hidden-import textual.app ^
|
||||
--hidden-import textual.widgets ^
|
||||
--hidden-import textual.css ^
|
||||
--hidden-import openai ^
|
||||
--hidden-import openai._client ^
|
||||
--hidden-import neo4j ^
|
||||
--hidden-import sqlite3 ^
|
||||
--hidden-import core ^
|
||||
--hidden-import core.embedded_db ^
|
||||
--hidden-import core.graph_client ^
|
||||
--hidden-import core.tool_executor ^
|
||||
--hidden-import core.tool_limiter ^
|
||||
--hidden-import core.tools ^
|
||||
--hidden-import core.tools.memory_tools ^
|
||||
--hidden-import core.prompts ^
|
||||
--hidden-import core.prompts.prompt_manager ^
|
||||
--hidden-import ui ^
|
||||
--hidden-import ui.app ^
|
||||
--hidden-import ui.models ^
|
||||
--hidden-import ui.models.message ^
|
||||
--hidden-import ui.models.config ^
|
||||
--hidden-import ui.models.log_entry ^
|
||||
--hidden-import ui.widgets ^
|
||||
--hidden-import ui.handlers ^
|
||||
--hidden-import ui.services ^
|
||||
--hidden-import ui.services.config_manager ^
|
||||
--hidden-import ui.services.config_service ^
|
||||
--collect-all textual ^
|
||||
--noconfirm
|
||||
|
||||
echo ===== Build Complete =====
|
||||
echo Binary: dist\TrulyMEM.exe
|
||||
dir dist\TrulyMEM.exe
|
||||
pause
|
||||
@ -1,92 +0,0 @@
|
||||
# -*- mode: python ; coding: utf-8 -*-
|
||||
import os
|
||||
import sys
|
||||
|
||||
block_cipher = None
|
||||
|
||||
project_root = os.path.dirname(os.path.abspath(SPEC))
|
||||
sys.path.insert(0, project_root)
|
||||
|
||||
datas = []
|
||||
if os.path.exists(os.path.join(project_root, 'ui', 'styles')):
|
||||
for root, dirs, files in os.walk(os.path.join(project_root, 'ui', 'styles')):
|
||||
for f in files:
|
||||
src = os.path.join(root, f)
|
||||
dst = os.path.join('ui', 'styles', os.path.relpath(src, os.path.join(project_root, 'ui', 'styles')))
|
||||
datas.append((src, dst))
|
||||
|
||||
if os.path.exists(os.path.join(project_root, 'core', 'prompts', 'templates')):
|
||||
for root, dirs, files in os.walk(os.path.join(project_root, 'core', 'prompts', 'templates')):
|
||||
for f in files:
|
||||
src = os.path.join(root, f)
|
||||
dst = os.path.join('core', 'prompts', 'templates', os.path.relpath(src, os.path.join(project_root, 'core', 'prompts', 'templates')))
|
||||
datas.append((src, dst))
|
||||
|
||||
a = Analysis(
|
||||
['trulymem_entry.py'],
|
||||
pathex=[],
|
||||
binaries=[],
|
||||
datas=datas,
|
||||
hiddenimports=[
|
||||
'textual',
|
||||
'textual.app',
|
||||
'textual.widgets',
|
||||
'textual.css',
|
||||
'openai',
|
||||
'openai._client',
|
||||
'neo4j',
|
||||
'sqlite3',
|
||||
'graph_memory_tui',
|
||||
'core',
|
||||
'core.embedded_db',
|
||||
'core.graph_client',
|
||||
'core.tool_executor',
|
||||
'core.tool_limiter',
|
||||
'core.tools',
|
||||
'core.tools.memory_tools',
|
||||
'core.prompts',
|
||||
'core.prompts.prompt_manager',
|
||||
'ui',
|
||||
'ui.app',
|
||||
'ui.models',
|
||||
'ui.models.message',
|
||||
'ui.models.config',
|
||||
'ui.models.log_entry',
|
||||
'ui.widgets',
|
||||
'ui.widgets.left_panel',
|
||||
'ui.widgets.right_panel',
|
||||
'ui.widgets.input_box',
|
||||
'ui.widgets.message_history',
|
||||
'ui.widgets.status_bar',
|
||||
'ui.handlers',
|
||||
'ui.services',
|
||||
'ui.services.config_manager',
|
||||
],
|
||||
hookspath=[],
|
||||
hooksconfig={},
|
||||
runtime_hooks=[],
|
||||
excludes=[],
|
||||
noarchive=False,
|
||||
optimize=0,
|
||||
)
|
||||
pyz = PYZ(a.pure, block_cipher)
|
||||
exe = EXE(
|
||||
pyz,
|
||||
a.scripts,
|
||||
a.binaries,
|
||||
a.datas,
|
||||
[],
|
||||
name='TrulyMEM',
|
||||
debug=False,
|
||||
bootloader_ignore_signals=False,
|
||||
strip=False,
|
||||
upx=True,
|
||||
upx_exclude=[],
|
||||
runtime_tmpdir=None,
|
||||
console=True,
|
||||
disable_windowed_traceback=False,
|
||||
argv_emulation=False,
|
||||
target_arch=None,
|
||||
codesign_identity=None,
|
||||
entitlements_file=None,
|
||||
)
|
||||
@ -1,12 +0,0 @@
|
||||
from .server import BackendServer, Packet, PacketType, PacketResponse
|
||||
from .client import BackendClient
|
||||
from .embedded_db import EmbeddedGraphDB
|
||||
|
||||
__all__ = [
|
||||
"BackendServer",
|
||||
"BackendClient",
|
||||
"EmbeddedGraphDB",
|
||||
"Packet",
|
||||
"PacketType",
|
||||
"PacketResponse"
|
||||
]
|
||||
@ -1,90 +0,0 @@
|
||||
import threading
|
||||
import time
|
||||
from typing import Any, Dict
|
||||
|
||||
from .server import BackendServer, Packet, PacketType
|
||||
|
||||
|
||||
class BackendClient:
|
||||
|
||||
def __init__(self, server: BackendServer):
|
||||
self._server = server
|
||||
self._counter = 0
|
||||
self._lock = threading.Lock()
|
||||
|
||||
def _next_id(self) -> str:
|
||||
with self._lock:
|
||||
self._counter += 1
|
||||
return f"{time.time()}_{self._counter}"
|
||||
|
||||
def send(self, message: str) -> Dict:
|
||||
return self.process_message(message)
|
||||
|
||||
def process_message(self, user_input: str) -> Dict:
|
||||
return self._server.process_message(user_input)
|
||||
|
||||
def get_settings(self) -> Dict:
|
||||
packet = Packet(
|
||||
id=self._next_id(),
|
||||
type=PacketType.GET_SETTINGS,
|
||||
body={}
|
||||
)
|
||||
return self._server.send(packet).body
|
||||
|
||||
def update_settings(self, api_config: Dict = None, tool_limits: Dict = None) -> Dict:
|
||||
packet = Packet(
|
||||
id=self._next_id(),
|
||||
type=PacketType.SET_SETTINGS,
|
||||
body={
|
||||
"api_config": api_config or {},
|
||||
"tool_limits": tool_limits or {}
|
||||
}
|
||||
)
|
||||
return self._server.send(packet).body
|
||||
|
||||
def execute_tool(self, name: str, arguments: Dict) -> Dict:
|
||||
packet = Packet(
|
||||
id=self._next_id(),
|
||||
type=PacketType.EXECUTE_TOOL,
|
||||
body={"tool_name": name, "arguments": arguments}
|
||||
)
|
||||
return self._server.send(packet).body
|
||||
|
||||
def get_status(self) -> Dict:
|
||||
packet = Packet(
|
||||
id=self._next_id(),
|
||||
type=PacketType.GET_STATUS,
|
||||
body={}
|
||||
)
|
||||
return self._server.send(packet).body
|
||||
|
||||
def save_history(self, messages: list) -> Dict:
|
||||
packet = Packet(
|
||||
id=self._next_id(),
|
||||
type=PacketType.SAVE_HISTORY,
|
||||
body={"messages": messages}
|
||||
)
|
||||
response = self._server.send(packet)
|
||||
return response.body.get("data", {})
|
||||
|
||||
def get_history(self) -> list:
|
||||
packet = Packet(
|
||||
id=self._next_id(),
|
||||
type=PacketType.GET_HISTORY,
|
||||
body={}
|
||||
)
|
||||
response = self._server.send(packet)
|
||||
data = response.body.get("data", {})
|
||||
return data.get("history", [])
|
||||
|
||||
def clear_history(self) -> Dict:
|
||||
packet = Packet(
|
||||
id=self._next_id(),
|
||||
type=PacketType.SAVE_HISTORY,
|
||||
body={"messages": []}
|
||||
)
|
||||
response = self._server.send(packet)
|
||||
return response.body.get("data", {})
|
||||
|
||||
def shutdown(self) -> None:
|
||||
self._server.shutdown()
|
||||
@ -1,498 +0,0 @@
|
||||
"""
|
||||
内嵌图数据库 - 基于SQLite实现
|
||||
无需Docker，开箱即用
|
||||
"""
|
||||
|
||||
import sqlite3
|
||||
import json
|
||||
from datetime import datetime
|
||||
from pathlib import Path
|
||||
from typing import List, Dict, Optional, Any
|
||||
|
||||
|
||||
class EmbeddedGraphDB:
|
||||
"""内嵌图数据库 - SQLite实现"""
|
||||
|
||||
def __init__(self, db_path: str = "graph_memory.db"):
|
||||
"""
|
||||
初始化数据库
|
||||
|
||||
Args:
|
||||
db_path: 数据库文件路径
|
||||
"""
|
||||
self.db_path = Path(db_path)
|
||||
self.conn = None
|
||||
self._init_db()
|
||||
|
||||
def _init_db(self):
|
||||
"""初始化数据库表"""
|
||||
self.conn = sqlite3.connect(str(self.db_path), check_same_thread=False)
|
||||
self.conn.row_factory = sqlite3.Row
|
||||
|
||||
cursor = self.conn.cursor()
|
||||
|
||||
# 创建实体表
|
||||
cursor.execute("""
|
||||
CREATE TABLE IF NOT EXISTS entities (
|
||||
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
||||
name TEXT UNIQUE NOT NULL,
|
||||
type TEXT,
|
||||
mention_count INTEGER DEFAULT 1,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
)
|
||||
""")
|
||||
|
||||
# 创建关系表
|
||||
cursor.execute("""
|
||||
CREATE TABLE IF NOT EXISTS relations (
|
||||
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
||||
source_id INTEGER NOT NULL,
|
||||
target_id INTEGER NOT NULL,
|
||||
relation_type TEXT NOT NULL,
|
||||
confidence REAL DEFAULT 1.0,
|
||||
status TEXT DEFAULT 'active',
|
||||
session_id TEXT,
|
||||
turn_id INTEGER,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
date_bucket TEXT,
|
||||
superseded_by INTEGER,
|
||||
FOREIGN KEY (source_id) REFERENCES entities(id),
|
||||
FOREIGN KEY (target_id) REFERENCES entities(id)
|
||||
)
|
||||
""")
|
||||
|
||||
# 创建索引
|
||||
cursor.execute("CREATE INDEX IF NOT EXISTS idx_entity_name ON entities(name)")
|
||||
cursor.execute("CREATE INDEX IF NOT EXISTS idx_entity_type ON entities(type)")
|
||||
cursor.execute("CREATE INDEX IF NOT EXISTS idx_relation_source ON relations(source_id)")
|
||||
cursor.execute("CREATE INDEX IF NOT EXISTS idx_relation_target ON relations(target_id)")
|
||||
cursor.execute("CREATE INDEX IF NOT EXISTS idx_relation_type ON relations(relation_type)")
|
||||
cursor.execute("CREATE INDEX IF NOT EXISTS idx_relation_status ON relations(status)")
|
||||
|
||||
cursor.execute("""
|
||||
SELECT name FROM sqlite_master
|
||||
WHERE type='table' AND name='chat_records'
|
||||
""")
|
||||
if not cursor.fetchone():
|
||||
cursor.execute("""
|
||||
CREATE TABLE chat_records (
|
||||
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
||||
role TEXT NOT NULL,
|
||||
content TEXT NOT NULL,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
)
|
||||
""")
|
||||
cursor.execute("CREATE INDEX idx_chat_created ON chat_records(created_at)")
|
||||
|
||||
self.conn.commit()
|
||||
|
||||
def ensure_constraints(self):
|
||||
"""确保约束（兼容Neo4j接口）"""
|
||||
pass # SQLite自动处理
|
||||
|
||||
def recall(self, query_intent: str, seed_entities: List[str] = None,
|
||||
depth: int = 2, time_range: Dict = None,
|
||||
session_filter: str = None) -> Dict:
|
||||
"""
|
||||
检索相关记忆
|
||||
|
||||
Args:
|
||||
query_intent: 查询关键词（逗号分隔）
|
||||
seed_entities: 种子实体
|
||||
depth: 搜索深度
|
||||
time_range: 时间范围
|
||||
session_filter: 会话过滤
|
||||
|
||||
Returns:
|
||||
检索结果
|
||||
"""
|
||||
keywords = [w.strip().lower() for w in query_intent.replace(',', ' ').split() if w.strip()]
|
||||
|
||||
cursor = self.conn.cursor()
|
||||
|
||||
# 搜索实体
|
||||
entities = []
|
||||
entity_ids = set()
|
||||
|
||||
# 如果没有关键词，返回所有实体（用于"我们都聊过什么"这类问题）
|
||||
if not keywords and not seed_entities:
|
||||
cursor.execute("""
|
||||
SELECT id, name, type, mention_count
|
||||
FROM entities
|
||||
ORDER BY mention_count DESC
|
||||
LIMIT 50
|
||||
""")
|
||||
|
||||
for row in cursor.fetchall():
|
||||
entity_ids.add(row['id'])
|
||||
entities.append({
|
||||
'name': row['name'],
|
||||
'type': row['type'] or 'unknown',
|
||||
'mention_count': row['mention_count']
|
||||
})
|
||||
else:
|
||||
# 有关键词，按关键词搜索
|
||||
for keyword in keywords:
|
||||
cursor.execute("""
|
||||
SELECT id, name, type, mention_count
|
||||
FROM entities
|
||||
WHERE LOWER(name) LIKE ?
|
||||
""", (f"%{keyword}%",))
|
||||
|
||||
for row in cursor.fetchall():
|
||||
if row['id'] not in entity_ids:
|
||||
entity_ids.add(row['id'])
|
||||
entities.append({
|
||||
'name': row['name'],
|
||||
'type': row['type'] or 'unknown',
|
||||
'mention_count': row['mention_count']
|
||||
})
|
||||
|
||||
# 搜索关系
|
||||
relations = []
|
||||
|
||||
if entity_ids:
|
||||
placeholders = ','.join('?' * len(entity_ids))
|
||||
|
||||
query = f"""
|
||||
SELECT r.id, e1.name as source, e2.name as target,
|
||||
r.relation_type as type, r.confidence, r.session_id,
|
||||
r.turn_id, r.created_at, r.status
|
||||
FROM relations r
|
||||
JOIN entities e1 ON r.source_id = e1.id
|
||||
JOIN entities e2 ON r.target_id = e2.id
|
||||
WHERE (r.source_id IN ({placeholders}) OR r.target_id IN ({placeholders}))
|
||||
AND r.status = 'active'
|
||||
"""
|
||||
|
||||
params = list(entity_ids) + list(entity_ids)
|
||||
|
||||
if session_filter:
|
||||
query += " AND r.session_id = ?"
|
||||
params.append(session_filter)
|
||||
|
||||
cursor.execute(query, params)
|
||||
|
||||
for row in cursor.fetchall():
|
||||
relations.append({
|
||||
'source': row['source'],
|
||||
'target': row['target'],
|
||||
'type': row['type'],
|
||||
'confidence': row['confidence'],
|
||||
'session_id': row['session_id'],
|
||||
'turn_id': row['turn_id'],
|
||||
'created_at': row['created_at'],
|
||||
'status': row['status']
|
||||
})
|
||||
|
||||
return {
|
||||
"entities": entities,
|
||||
"relations": relations,
|
||||
"message": f"找到 {len(entities)} 个实体, {len(relations)} 条关系"
|
||||
}
|
||||
|
||||
def commit(self, triplets: List[Dict], entity_types: Dict = None,
|
||||
temporal_tag: str = None, session_id: str = None,
|
||||
turn_id: int = None) -> Dict:
|
||||
"""
|
||||
写入记忆
|
||||
|
||||
Args:
|
||||
triplets: 三元组列表
|
||||
entity_types: 实体类型
|
||||
temporal_tag: 时间标签
|
||||
session_id: 会话ID
|
||||
turn_id: 轮次ID
|
||||
|
||||
Returns:
|
||||
写入结果
|
||||
"""
|
||||
cursor = self.conn.cursor()
|
||||
|
||||
created_entities = 0
|
||||
created_relations = 0
|
||||
|
||||
for triplet in triplets:
|
||||
subject = triplet.get('subject')
|
||||
relation = triplet.get('relation')
|
||||
obj = triplet.get('object')
|
||||
confidence = triplet.get('confidence', 1.0)
|
||||
|
||||
if not all([subject, relation, obj]):
|
||||
continue
|
||||
|
||||
# 创建或更新实体
|
||||
for entity_name in [subject, obj]:
|
||||
entity_type = entity_types.get(entity_name) if entity_types else None
|
||||
|
||||
cursor.execute("""
|
||||
INSERT INTO entities (name, type)
|
||||
VALUES (?, ?)
|
||||
ON CONFLICT(name) DO UPDATE SET
|
||||
mention_count = mention_count + 1,
|
||||
updated_at = CURRENT_TIMESTAMP
|
||||
""", (entity_name, entity_type))
|
||||
|
||||
if cursor.rowcount > 0:
|
||||
created_entities += 1
|
||||
|
||||
# 获取实体ID
|
||||
cursor.execute("SELECT id FROM entities WHERE name = ?", (subject,))
|
||||
source_id = cursor.fetchone()['id']
|
||||
|
||||
cursor.execute("SELECT id FROM entities WHERE name = ?", (obj,))
|
||||
target_id = cursor.fetchone()['id']
|
||||
|
||||
# 创建关系
|
||||
date_bucket = datetime.now().strftime('%Y-%m-%d')
|
||||
|
||||
cursor.execute("""
|
||||
INSERT INTO relations (
|
||||
source_id, target_id, relation_type, confidence,
|
||||
session_id, turn_id, date_bucket
|
||||
)
|
||||
VALUES (?, ?, ?, ?, ?, ?, ?)
|
||||
""", (source_id, target_id, relation, confidence,
|
||||
session_id, turn_id, date_bucket))
|
||||
|
||||
created_relations += 1
|
||||
|
||||
self.conn.commit()
|
||||
|
||||
return {
|
||||
"created_entities": created_entities,
|
||||
"created_relations": created_relations,
|
||||
"message": f"创建了 {created_entities} 个实体, {created_relations} 条关系"
|
||||
}
|
||||
|
||||
def purge(self, criteria: Dict, mode: str = "soft",
|
||||
new_relation: Dict = None) -> Dict:
|
||||
"""
|
||||
删除或修正记忆
|
||||
|
||||
Args:
|
||||
criteria: 删除条件
|
||||
mode: 删除模式 (soft/hard)
|
||||
new_relation: 替代关系
|
||||
|
||||
Returns:
|
||||
删除结果
|
||||
"""
|
||||
cursor = self.conn.cursor()
|
||||
|
||||
# 构建查询条件
|
||||
conditions = []
|
||||
params = []
|
||||
|
||||
if criteria.get('source'):
|
||||
cursor.execute("SELECT id FROM entities WHERE name = ?", (criteria['source'],))
|
||||
row = cursor.fetchone()
|
||||
if row:
|
||||
conditions.append("source_id = ?")
|
||||
params.append(row['id'])
|
||||
|
||||
if criteria.get('target'):
|
||||
cursor.execute("SELECT id FROM entities WHERE name = ?", (criteria['target'],))
|
||||
row = cursor.fetchone()
|
||||
if row:
|
||||
conditions.append("target_id = ?")
|
||||
params.append(row['id'])
|
||||
|
||||
if criteria.get('relation'):
|
||||
conditions.append("relation_type = ?")
|
||||
params.append(criteria['relation'])
|
||||
|
||||
if not conditions:
|
||||
return {"deleted": 0, "message": "无删除条件"}
|
||||
|
||||
where_clause = " AND ".join(conditions)
|
||||
|
||||
if mode == "soft":
|
||||
cursor.execute(f"""
|
||||
UPDATE relations
|
||||
SET status = 'deleted', updated_at = CURRENT_TIMESTAMP
|
||||
WHERE {where_clause} AND status = 'active'
|
||||
""", params)
|
||||
else:
|
||||
cursor.execute(f"""
|
||||
DELETE FROM relations
|
||||
WHERE {where_clause}
|
||||
""", params)
|
||||
|
||||
deleted = cursor.rowcount
|
||||
self.conn.commit()
|
||||
|
||||
return {
|
||||
"deleted": deleted,
|
||||
"mode": mode,
|
||||
"message": f"删除了 {deleted} 条关系"
|
||||
}
|
||||
|
||||
def introspect(self, session_id: str = None) -> Dict:
|
||||
"""
|
||||
查看会话状态
|
||||
|
||||
Args:
|
||||
session_id: 会话ID
|
||||
|
||||
Returns:
|
||||
会话状态
|
||||
"""
|
||||
cursor = self.conn.cursor()
|
||||
|
||||
# 统计实体
|
||||
cursor.execute("SELECT COUNT(*) as count FROM entities")
|
||||
entity_count = cursor.fetchone()['count']
|
||||
|
||||
# 统计关系
|
||||
cursor.execute("SELECT COUNT(*) as count FROM relations WHERE status = 'active'")
|
||||
relation_count = cursor.fetchone()['count']
|
||||
|
||||
return {
|
||||
"entity_count": entity_count,
|
||||
"relation_count": relation_count,
|
||||
"session_id": session_id,
|
||||
"message": f"数据库包含 {entity_count} 个实体, {relation_count} 条关系"
|
||||
}
|
||||
|
||||
def archive(self, days: int = 30) -> Dict:
|
||||
"""归档旧关系"""
|
||||
cursor = self.conn.cursor()
|
||||
|
||||
cursor.execute("""
|
||||
UPDATE relations
|
||||
SET status = 'archived', updated_at = CURRENT_TIMESTAMP
|
||||
WHERE status = 'active'
|
||||
AND created_at < datetime('now', ?)
|
||||
""", (f'-{days} days',))
|
||||
|
||||
archived = cursor.rowcount
|
||||
self.conn.commit()
|
||||
|
||||
return {
|
||||
"archived": archived,
|
||||
"message": f"归档了 {archived} 条关系"
|
||||
}
|
||||
|
||||
def cleanup(self, dry_run: bool = True) -> Dict:
|
||||
"""清理已删除数据"""
|
||||
cursor = self.conn.cursor()
|
||||
|
||||
if dry_run:
|
||||
cursor.execute("""
|
||||
SELECT COUNT(*) as count
|
||||
FROM relations
|
||||
WHERE status = 'deleted'
|
||||
AND updated_at < datetime('now', '-90 days')
|
||||
""")
|
||||
deleted_relations = cursor.fetchone()['count']
|
||||
|
||||
return {
|
||||
"dry_run": True,
|
||||
"deleted_relations": deleted_relations,
|
||||
"message": f"将删除 {deleted_relations} 条关系"
|
||||
}
|
||||
else:
|
||||
cursor.execute("""
|
||||
DELETE FROM relations
|
||||
WHERE status = 'deleted'
|
||||
AND updated_at < datetime('now', '-90 days')
|
||||
""")
|
||||
deleted = cursor.rowcount
|
||||
self.conn.commit()
|
||||
|
||||
return {
|
||||
"dry_run": False,
|
||||
"deleted": deleted,
|
||||
"message": f"删除了 {deleted} 条关系"
|
||||
}
|
||||
|
||||
def save_chat_records(self, messages: list) -> Dict:
|
||||
"""保存聊天记录到数据库"""
|
||||
cursor = self.conn.cursor()
|
||||
|
||||
saved = 0
|
||||
for msg in messages:
|
||||
role = msg.get("role")
|
||||
content = msg.get("content")
|
||||
if role and content:
|
||||
cursor.execute(
|
||||
"INSERT INTO chat_records (role, content) VALUES (?, ?)",
|
||||
(role, content)
|
||||
)
|
||||
saved += 1
|
||||
|
||||
self.conn.commit()
|
||||
|
||||
cursor.execute("""
|
||||
DELETE FROM chat_records
|
||||
WHERE id NOT IN (
|
||||
SELECT id FROM chat_records
|
||||
ORDER BY id DESC
|
||||
LIMIT 500
|
||||
)
|
||||
""")
|
||||
self.conn.commit()
|
||||
|
||||
return {"saved": saved}
|
||||
|
||||
def get_chat_records(self, limit: int = 500) -> list:
|
||||
"""从数据库获取聊天记录"""
|
||||
cursor = self.conn.cursor()
|
||||
cursor.execute("""
|
||||
SELECT role, content FROM chat_records
|
||||
ORDER BY id ASC LIMIT ?
|
||||
""", (limit,))
|
||||
return [{"role": row[0], "content": row[1]} for row in cursor.fetchall()]
|
||||
|
||||
def clear_chat_records(self) -> Dict:
|
||||
"""清空聊天记录（保留图数据库）"""
|
||||
cursor = self.conn.cursor()
|
||||
cursor.execute("DELETE FROM chat_records")
|
||||
self.conn.commit()
|
||||
return {"cleared": True}
|
||||
|
||||
def close(self):
|
||||
"""关闭数据库连接"""
|
||||
if self.conn:
|
||||
self.conn.close()
|
||||
self.conn = None
|
||||
|
||||
def __enter__(self):
|
||||
return self
|
||||
|
||||
def __exit__(self, exc_type, exc_val, exc_tb):
|
||||
self.close()
|
||||
|
||||
|
||||
# 兼容性别名
|
||||
Neo4jGraph = EmbeddedGraphDB
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
# 测试
|
||||
print("Testing Embedded Graph Database...")
|
||||
|
||||
with EmbeddedGraphDB("test.db") as db:
|
||||
# 写入测试
|
||||
result = db.commit(
|
||||
triplets=[
|
||||
{"subject": "用户", "relation": "喜欢", "object": "Python"},
|
||||
{"subject": "用户", "relation": "学习", "object": "AI"}
|
||||
],
|
||||
session_id="test-session",
|
||||
turn_id=1
|
||||
)
|
||||
print(f"Commit: {result}")
|
||||
|
||||
# 检索测试
|
||||
result = db.recall("Python,AI")
|
||||
print(f"Recall: {result}")
|
||||
|
||||
# 状态测试
|
||||
result = db.introspect()
|
||||
print(f"Introspect: {result}")
|
||||
|
||||
print("\nTest completed!")
|
||||
@ -1,393 +0,0 @@
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
Graph Memory Client - 图记忆客户端核心实现（重构版）
|
||||
使用模块化的工具和提示词系统
|
||||
"""
|
||||
|
||||
import json
|
||||
import os
|
||||
import uuid
|
||||
from datetime import datetime
|
||||
from openai import OpenAI
|
||||
|
||||
from .tools import TOOLS
|
||||
from .tool_executor import execute_tool
|
||||
from .prompts.prompt_manager import PromptManager
|
||||
|
||||
# 环境配置
|
||||
DEEPSEEK_API_KEY = os.environ.get("DEEPSEEK_API_KEY", "")
|
||||
DEEPSEEK_BASE_URL = os.environ.get("DEEPSEEK_BASE_URL", "https://api.deepseek.com")
|
||||
MODEL_NAME = os.environ.get("MODEL_NAME", "deepseek-chat")
|
||||
|
||||
NEO4J_URI = os.environ.get("NEO4J_URI", "bolt://localhost:7687")
|
||||
NEO4J_USER = os.environ.get("NEO4J_USER", "neo4j")
|
||||
NEO4J_PASSWORD = os.environ.get("NEO4J_PASSWORD", "neo4j")
|
||||
|
||||
# 会话配置
|
||||
CURRENT_SESSION_ID = f"session-{datetime.now().strftime('%Y%m%d')}-{uuid.uuid4().hex[:4]}"
|
||||
CURRENT_TURN = 0
|
||||
|
||||
|
||||
class Neo4jGraph:
|
||||
"""Neo4j图数据库客户端"""
|
||||
|
||||
def __init__(self, uri: str, user: str, password: str):
|
||||
from neo4j import GraphDatabase
|
||||
self.driver = GraphDatabase.driver(uri, auth=(user, password))
|
||||
|
||||
def close(self):
|
||||
self.driver.close()
|
||||
|
||||
def ensure_constraints(self):
|
||||
"""确保约束和索引存在"""
|
||||
with self.driver.session() as session:
|
||||
# 实体约束
|
||||
session.run("CREATE CONSTRAINT entity_name_constraint IF NOT EXISTS FOR (e:Entity) REQUIRE e.name IS UNIQUE")
|
||||
session.run("CREATE CONSTRAINT session_id_constraint IF NOT EXISTS FOR (s:Session) REQUIRE s.session_id IS UNIQUE")
|
||||
|
||||
# 关系索引
|
||||
session.run("CREATE INDEX rel_created_at IF NOT EXISTS FOR ()-[r:RELATES]-() ON r.created_at")
|
||||
session.run("CREATE INDEX rel_session_id IF NOT EXISTS FOR ()-[r:RELATES]-() ON r.session_id")
|
||||
session.run("CREATE INDEX rel_type IF NOT EXISTS FOR ()-[r:RELATES]-() ON r.type")
|
||||
session.run("CREATE INDEX rel_status IF NOT EXISTS FOR ()-[r:RELATES]-() ON r.status")
|
||||
session.run("CREATE INDEX rel_date_bucket IF NOT EXISTS FOR ()-[r:RELATES]-() ON r.date_bucket")
|
||||
|
||||
# 实体索引
|
||||
session.run("CREATE INDEX entity_type IF NOT EXISTS FOR (e:Entity) ON e.type")
|
||||
session.run("CREATE INDEX entity_mention_count IF NOT EXISTS FOR (e:Entity) ON e.mention_count")
|
||||
|
||||
def recall(self, query_intent: str, seed_entities: list = None, depth: int = 2,
|
||||
time_range: dict = None, session_filter: str = None) -> dict:
|
||||
"""检索记忆"""
|
||||
with self.driver.session() as session:
|
||||
# 支持逗号分隔的多个关键词
|
||||
keywords = [w.strip() for w in query_intent.replace(',', ' ').split() if len(w.strip()) > 0]
|
||||
|
||||
if not keywords and not seed_entities:
|
||||
return {"entities": [], "relations": [], "message": "无查询关键词"}
|
||||
|
||||
params = {}
|
||||
cond_parts = ["r.status = 'active'"]
|
||||
|
||||
if session_filter:
|
||||
cond_parts.append("r.session_id = $session_id")
|
||||
params["session_id"] = session_filter
|
||||
|
||||
if keywords:
|
||||
keyword_conditions = []
|
||||
for k in keywords:
|
||||
k_lower = k.lower()
|
||||
keyword_conditions.append(f"toLower(e.name) CONTAINS '{k_lower}'")
|
||||
keyword_conditions.append(f"toLower(t.name) CONTAINS '{k_lower}'")
|
||||
keyword_conditions.append(f"toLower(r.type) CONTAINS '{k_lower}'")
|
||||
cond_parts.append(f"({' OR '.join(keyword_conditions)})")
|
||||
|
||||
if seed_entities:
|
||||
placeholders = ",".join([f"'{s}'" for s in seed_entities])
|
||||
cond_parts.append(f"(e.name IN [{placeholders}] OR t.name IN [{placeholders}])")
|
||||
|
||||
if time_range and "days" in time_range:
|
||||
cond_parts.append(f"r.created_at >= datetime() - duration('P{time_range['days']}D')")
|
||||
|
||||
where_clause = " AND ".join(cond_parts)
|
||||
|
||||
cypher = f"""
|
||||
MATCH (e:Entity)-[r:RELATES]->(t:Entity)
|
||||
WHERE {where_clause}
|
||||
RETURN e, r, t
|
||||
ORDER BY r.created_at DESC
|
||||
LIMIT 30
|
||||
"""
|
||||
|
||||
result = session.run(cypher, params)
|
||||
entities, relations = {}, []
|
||||
|
||||
for record in result:
|
||||
e, r, t = record["e"], record["r"], record["t"]
|
||||
if e["name"] not in entities:
|
||||
entities[e["name"]] = {"name": e["name"], "type": e.get("type", "unknown"), "mention_count": e.get("mention_count", 1)}
|
||||
if t["name"] not in entities:
|
||||
entities[t["name"]] = {"name": t["name"], "type": t.get("type", "unknown"), "mention_count": t.get("mention_count", 1)}
|
||||
|
||||
relations.append({
|
||||
"source": e["name"],
|
||||
"target": t["name"],
|
||||
"type": r["type"],
|
||||
"created_at": str(r.get("created_at", "")),
|
||||
"session_id": r.get("session_id", ""),
|
||||
"turn_id": r.get("turn_id", 0),
|
||||
"confidence": r.get("confidence", 1.0)
|
||||
})
|
||||
|
||||
return {"entities": list(entities.values()), "relations": relations[:20]}
|
||||
|
||||
def commit(self, triplets: list, entity_types: list = None, temporal_tag: str = None) -> dict:
|
||||
"""写入记忆"""
|
||||
global CURRENT_TURN
|
||||
with self.driver.session() as session:
|
||||
valid_triplets = [t for t in triplets if t.get("subject") and t.get("relation") and t.get("object")]
|
||||
|
||||
if not valid_triplets:
|
||||
return {"committed_count": 0, "details": []}
|
||||
|
||||
etype = entity_types[0] if entity_types else "unknown"
|
||||
date_bucket = temporal_tag or datetime.now().strftime("%Y-%m-%d")
|
||||
|
||||
results = []
|
||||
for triplet in valid_triplets:
|
||||
subject = triplet.get("subject", "").strip()
|
||||
relation = triplet.get("relation", "").strip()
|
||||
obj = triplet.get("object", "").strip()
|
||||
confidence = triplet.get("confidence", 0.9)
|
||||
|
||||
session.run("""
|
||||
MERGE (s:Entity {name: $subject})
|
||||
ON CREATE SET s.type = $type, s.created_at = datetime(), s.mention_count = 1, s.updated_at = datetime()
|
||||
ON MATCH SET s.mention_count = coalesce(s.mention_count, 0) + 1, s.updated_at = datetime()
|
||||
|
||||
MERGE (t:Entity {name: $object})
|
||||
ON CREATE SET t.type = $type, t.created_at = datetime(), t.mention_count = 1, t.updated_at = datetime()
|
||||
ON MATCH SET t.mention_count = coalesce(t.mention_count, 0) + 1, t.updated_at = datetime()
|
||||
|
||||
CREATE (s)-[r:RELATES {
|
||||
type: $relation,
|
||||
created_at: datetime(),
|
||||
session_id: $session_id,
|
||||
turn_id: $turn_id,
|
||||
role: 'user',
|
||||
status: 'active',
|
||||
confidence: $confidence,
|
||||
date_bucket: $date_bucket
|
||||
}]->(t)
|
||||
""", subject=subject, object=obj, relation=relation, type=etype,
|
||||
session_id=CURRENT_SESSION_ID, turn_id=CURRENT_TURN, confidence=confidence,
|
||||
date_bucket=date_bucket)
|
||||
|
||||
results.append(f"{subject} -[{relation}]-> {obj}")
|
||||
|
||||
return {"committed_count": len(results), "details": results}
|
||||
|
||||
def purge(self, criteria: dict, mode: str = "soft", new_relation: dict = None) -> dict:
|
||||
"""删除记忆"""
|
||||
with self.driver.session() as session:
|
||||
subject_pattern = criteria.get("subject_contains", "")
|
||||
rel_type = criteria.get("relation_type", "")
|
||||
target_pattern = criteria.get("target_contains", "")
|
||||
session_id = criteria.get("session_id", CURRENT_SESSION_ID)
|
||||
|
||||
cond_parts = ["r.status = 'active'"]
|
||||
params = {"session_id": session_id}
|
||||
|
||||
if subject_pattern:
|
||||
cond_parts.append("e.name CONTAINS $subject")
|
||||
params["subject"] = subject_pattern
|
||||
if target_pattern:
|
||||
cond_parts.append("t.name CONTAINS $target")
|
||||
params["target"] = target_pattern
|
||||
if rel_type:
|
||||
cond_parts.append("r.type = $rel_type")
|
||||
params["rel_type"] = rel_type
|
||||
|
||||
where_clause = " AND ".join(cond_parts)
|
||||
|
||||
if mode == "supersede" and new_relation:
|
||||
new_rel = new_relation.get("relation", "")
|
||||
new_target = new_relation.get("target", "")
|
||||
|
||||
if not new_rel or not new_target:
|
||||
return {"error": "supersede模式需要提供new_relation.relation和new_relation.target"}
|
||||
|
||||
result = session.run(f"""
|
||||
MATCH (s:Entity)-[r:RELATES]->(t:Entity)
|
||||
WHERE {where_clause}
|
||||
SET r.status = 'superseded', r.updated_at = datetime()
|
||||
RETURN count(r) as count
|
||||
""", params)
|
||||
|
||||
count = result.single()["count"]
|
||||
return {"deleted_count": count, "mode": "supersede"}
|
||||
else:
|
||||
result = session.run(f"""
|
||||
MATCH ()-[r:RELATES]->()
|
||||
WHERE {where_clause}
|
||||
SET r.status = 'deleted', r.updated_at = datetime()
|
||||
RETURN count(r) as deleted
|
||||
""", params)
|
||||
count = result.single()["deleted"]
|
||||
|
||||
return {"deleted_count": count, "mode": "soft"}
|
||||
|
||||
def introspect(self, session_id: str = None) -> dict:
|
||||
"""查看记忆状态"""
|
||||
target_session = session_id or CURRENT_SESSION_ID
|
||||
|
||||
with self.driver.session() as session:
|
||||
result = session.run("""
|
||||
MATCH (s:Entity)-[r:RELATES]->(t:Entity)
|
||||
WHERE r.session_id = $session_id AND r.status = 'active'
|
||||
RETURN collect(DISTINCT s.name) as source_entities,
|
||||
collect(DISTINCT t.name) as target_entities,
|
||||
count(r) as rel_count,
|
||||
collect(DISTINCT r.type) as rel_types
|
||||
""", session_id=target_session)
|
||||
record = result.single()
|
||||
|
||||
result2 = session.run("""
|
||||
MATCH (e:Entity)
|
||||
RETURN e.name as name, e.mention_count as count, e.type as type
|
||||
ORDER BY e.mention_count DESC
|
||||
LIMIT 10
|
||||
""")
|
||||
hotspots = [(r["name"], r["count"], r["type"]) for r in result2]
|
||||
|
||||
return {
|
||||
"session_id": target_session,
|
||||
"total_turns": CURRENT_TURN,
|
||||
"entities_discussed": list(set((record["source_entities"] or []) + (record["target_entities"] or []))),
|
||||
"relation_count": record["rel_count"] if record else 0,
|
||||
"relation_types": record["rel_types"] if record else [],
|
||||
"memory_hotspots": hotspots
|
||||
}
|
||||
|
||||
def archive(self, days: int = 30) -> dict:
|
||||
"""归档旧记忆"""
|
||||
with self.driver.session() as session:
|
||||
result = session.run("""
|
||||
MATCH ()-[r:RELATES]->()
|
||||
WHERE r.status = 'active' AND r.created_at < datetime() - duration('P' + $days + 'D')
|
||||
SET r.status = 'archived', r.archived_at = datetime()
|
||||
RETURN count(r) as archived
|
||||
""", days=str(days))
|
||||
|
||||
return {"archived_count": result.single()["archived"], "days": days}
|
||||
|
||||
def cleanup(self, dry_run: bool = True) -> dict:
|
||||
"""清理无效数据"""
|
||||
with self.driver.session() as session:
|
||||
result1 = session.run("""
|
||||
MATCH ()-[r:RELATES]->()
|
||||
WHERE r.status = 'deleted' AND r.updated_at < datetime() - duration('P90D')
|
||||
RETURN count(r) as to_delete
|
||||
""")
|
||||
deleted_relations = result1.single()["to_delete"]
|
||||
|
||||
result2 = session.run("""
|
||||
MATCH (e:Entity)
|
||||
WHERE NOT (e)-[:RELATES]-()
|
||||
RETURN count(e) as orphans
|
||||
""")
|
||||
orphan_nodes = result2.single()["orphans"]
|
||||
|
||||
if not dry_run and deleted_relations > 0:
|
||||
session.run("""
|
||||
MATCH ()-[r:RELATES]->()
|
||||
WHERE r.status = 'deleted' AND r.updated_at < datetime() - duration('P90D')
|
||||
DELETE r
|
||||
""")
|
||||
|
||||
if not dry_run and orphan_nodes > 0:
|
||||
session.run("""
|
||||
MATCH (e:Entity)
|
||||
WHERE NOT (e)-[:RELATES]-()
|
||||
DELETE e
|
||||
""")
|
||||
|
||||
return {
|
||||
"dry_run": dry_run,
|
||||
"deleted_relations": deleted_relations,
|
||||
"orphan_nodes": orphan_nodes,
|
||||
"action_taken": not dry_run
|
||||
}
|
||||
|
||||
|
||||
class GraphMemoryClient:
|
||||
"""图记忆客户端"""
|
||||
|
||||
def __init__(self, api_key: str, base_url: str, graph, model: str = "deepseek-chat"):
|
||||
# 清理可能存在的错误代理环境变量
|
||||
import os
|
||||
proxy_vars = ['http_proxy', 'https_proxy', 'HTTP_PROXY', 'HTTPS_PROXY', 'all_proxy', 'ALL_PROXY']
|
||||
for var in proxy_vars:
|
||||
if var in os.environ:
|
||||
value = os.environ[var]
|
||||
# 如果代理URL没有scheme前缀，添加http://
|
||||
if value and not value.startswith(('http://', 'https://', 'socks5://', 'socks4://')):
|
||||
os.environ[var] = f'http://{value}'
|
||||
|
||||
self.client = OpenAI(api_key=api_key, base_url=base_url)
|
||||
self.graph = graph
|
||||
self.tools = TOOLS
|
||||
self.model = model
|
||||
|
||||
prompt_manager = PromptManager()
|
||||
self.system_prompt = prompt_manager.get_system_prompt()
|
||||
|
||||
def send_message(self, user_input: str, tool_results: list = None, assistant_msg: dict = None) -> dict:
|
||||
"""发送消息"""
|
||||
global CURRENT_TURN
|
||||
|
||||
messages = [{"role": "system", "content": self.system_prompt}]
|
||||
|
||||
# 添加用户消息
|
||||
messages.append({"role": "user", "content": user_input})
|
||||
|
||||
# 添加 assistant 消息（包含 tool_calls）
|
||||
if assistant_msg:
|
||||
messages.append(assistant_msg)
|
||||
|
||||
# 添加工具结果
|
||||
if tool_results:
|
||||
messages.extend(tool_results)
|
||||
|
||||
response = self.client.chat.completions.create(
|
||||
model=self.model,
|
||||
messages=messages,
|
||||
tools=self.tools,
|
||||
tool_choice="auto"
|
||||
)
|
||||
|
||||
return response
|
||||
|
||||
def send_message_with_history(self, messages_history: list) -> dict:
|
||||
"""使用消息历史发送消息"""
|
||||
global CURRENT_TURN
|
||||
|
||||
# 构建完整消息列表
|
||||
messages = [{"role": "system", "content": self.system_prompt}]
|
||||
messages.extend(messages_history)
|
||||
|
||||
response = self.client.chat.completions.create(
|
||||
model=self.model,
|
||||
messages=messages,
|
||||
tools=self.tools,
|
||||
tool_choice="auto"
|
||||
)
|
||||
|
||||
return response
|
||||
|
||||
def send_message_stream(self, user_input: str, tool_results: list = None, assistant_msg: dict = None):
|
||||
"""流式发送消息"""
|
||||
global CURRENT_TURN
|
||||
|
||||
messages = [{"role": "system", "content": self.system_prompt}]
|
||||
|
||||
# 添加用户消息
|
||||
messages.append({"role": "user", "content": user_input})
|
||||
|
||||
# 添加 assistant 消息（包含 tool_calls）
|
||||
if assistant_msg:
|
||||
messages.append(assistant_msg)
|
||||
|
||||
# 添加工具结果
|
||||
if tool_results:
|
||||
messages.extend(tool_results)
|
||||
|
||||
stream = self.client.chat.completions.create(
|
||||
model=self.model,
|
||||
messages=messages,
|
||||
tools=self.tools,
|
||||
tool_choice="auto",
|
||||
stream=True
|
||||
)
|
||||
|
||||
return stream
|
||||
@ -1,6 +0,0 @@
|
||||
"""
|
||||
提示词管理模块
|
||||
"""
|
||||
from .prompt_manager import PromptManager
|
||||
|
||||
__all__ = ["PromptManager"]
|
||||
@ -1,77 +0,0 @@
|
||||
"""
|
||||
提示词管理器
|
||||
"""
|
||||
from pathlib import Path
|
||||
|
||||
|
||||
class PromptManager:
|
||||
"""提示词管理器"""
|
||||
|
||||
_instance = None
|
||||
_cached_prompt = None
|
||||
|
||||
def __new__(cls):
|
||||
if cls._instance is None:
|
||||
cls._instance = super().__new__(cls)
|
||||
return cls._instance
|
||||
|
||||
def __init__(self):
|
||||
if not hasattr(self, '_initialized'):
|
||||
self.prompts_dir = Path(__file__).parent / "templates"
|
||||
self._initialized = True
|
||||
|
||||
def get_system_prompt(self) -> str:
|
||||
if PromptManager._cached_prompt is not None:
|
||||
return PromptManager._cached_prompt
|
||||
|
||||
prompt_file = self.prompts_dir / "system_prompt.md"
|
||||
if prompt_file.exists():
|
||||
with open(prompt_file, "r", encoding="utf-8") as f:
|
||||
PromptManager._cached_prompt = f.read()
|
||||
else:
|
||||
PromptManager._cached_prompt = self._build_default_prompt()
|
||||
|
||||
return PromptManager._cached_prompt
|
||||
|
||||
def _build_default_prompt(self) -> str:
|
||||
return """你是TrulyMEM，一个拥有长期记忆能力的AI助手。
|
||||
|
||||
## 核心能力
|
||||
|
||||
1. **长期记忆** - 基于图数据库存储实体关系
|
||||
2. **人设管理** - 支持角色扮演和性格设定
|
||||
3. **任务跟踪** - 维护工作记忆链，跟踪连续性任务
|
||||
|
||||
## 记忆原则
|
||||
|
||||
- **明确内容必须写入** - 用户明确提到的信息必须存储
|
||||
- **推理内容必须标注** - AI推理得到的内容标注[猜测]
|
||||
- **图数据库是唯一记忆源** - 没有其他记忆方式
|
||||
|
||||
## 工具使用
|
||||
|
||||
### 记忆工具
|
||||
- `memory_recall` - 检索记忆
|
||||
- `memory_commit` - 写入记忆
|
||||
- `memory_purge` - 删除记忆
|
||||
- `memory_introspect` - 查看状态
|
||||
|
||||
### 人设工具
|
||||
- `persona_update` - 更新人设
|
||||
- `persona_clear` - 清除人设
|
||||
|
||||
### 任务工具
|
||||
- `task_create` - 创建任务
|
||||
- `task_set_state` - 设置状态
|
||||
- `task_delete` - 删除任务
|
||||
- `task_link_info` - 关联信息
|
||||
|
||||
## 自主性
|
||||
|
||||
你有权根据对话上下文自主决定：
|
||||
- 是否需要查询记忆
|
||||
- 是否需要写入记忆
|
||||
- 是否需要维护任务链
|
||||
- 如何使用工具
|
||||
|
||||
记住：灵活应对，保持自然对话体验。"""
|
||||
@ -1,324 +0,0 @@
|
||||
# TrulyMEM 系统提示词
|
||||
|
||||
你是TrulyMEM，一个拥有长期记忆能力的AI助手。
|
||||
|
||||
## ⚠️ 最高优先级：只回复一次
|
||||
|
||||
**每轮对话只能回复一次！**
|
||||
|
||||
- 执行完所有工具调用后，给出一个完整的回复
|
||||
- 不要在工具调用过程中多次回复
|
||||
- 不要重复说相同的内容
|
||||
|
||||
---
|
||||
|
||||
## ⚠️ 关键约束：无传统上下文系统
|
||||
|
||||
**重要**: 你没有传统的对话上下文系统（没有消息历史数组）。
|
||||
|
||||
- ❌ **没有** messages数组存储历史对话
|
||||
- ❌ **没有** 传统的多轮对话上下文
|
||||
- ✅ **只有** 图数据库作为唯一记忆载体
|
||||
- ✅ **必须** 通过工作记忆链维持对话连贯性
|
||||
|
||||
## 核心身份
|
||||
|
||||
- **名称**: TrulyMEM (TrueHumanMEM)
|
||||
- **能力**: 基于图数据库的长期记忆
|
||||
- **理念**: 让AI的记忆方式更像人类
|
||||
|
||||
## 核心能力
|
||||
|
||||
### 1. 长期记忆
|
||||
- 图数据库存储实体关系
|
||||
- 支持时间范围查询
|
||||
- 支持会话过滤
|
||||
|
||||
### 2. 人设管理（关键）
|
||||
- 角色扮演支持
|
||||
- 性格、语气设定
|
||||
- 动态切换人设
|
||||
- **每轮必须查询人设图**
|
||||
|
||||
### 3. 任务跟踪（关键）
|
||||
- 工作记忆链 - **维持对话连贯性的唯一机制**
|
||||
- 任务状态管理
|
||||
- 上下文恢复
|
||||
|
||||
## 记忆原则
|
||||
|
||||
### 必须写入的情况
|
||||
- 用户明确表达偏好："我喜欢X"
|
||||
- 用户分享信息："我在做X项目"
|
||||
- 用户制定计划："我打算X"
|
||||
- 用户描述状态："我现在在X"
|
||||
|
||||
### 禁止写入的情况
|
||||
- AI推断的用户偏好
|
||||
- AI猜测的用户意图
|
||||
- AI推导的结论
|
||||
|
||||
### 标注规则
|
||||
- 推理内容必须标注 **[猜测]**
|
||||
- 明确内容直接陈述
|
||||
|
||||
## 工具系统
|
||||
|
||||
### 记忆工具
|
||||
| 工具 | 功能 | 使用场景 |
|
||||
|------|------|---------|
|
||||
| `memory_recall` | 检索记忆 | 查询历史信息 |
|
||||
| `memory_commit` | 写入记忆 | 存储重要信息 |
|
||||
| `memory_purge` | 删除记忆 | 修正错误信息 |
|
||||
| `memory_introspect` | 查看状态 | 监控记忆系统 |
|
||||
|
||||
### 人设工具
|
||||
| 工具 | 功能 | 使用场景 |
|
||||
|------|------|---------|
|
||||
| `persona_update` | 更新人设 | 设置角色属性 |
|
||||
| `persona_clear` | 清除人设 | 恢复默认身份 |
|
||||
|
||||
### 任务工具
|
||||
| 工具 | 功能 | 使用场景 |
|
||||
|------|------|---------|
|
||||
| `task_create` | 创建任务 | 开始连续性任务 |
|
||||
| `task_set_state` | 设置状态 | 更新任务状态 |
|
||||
| `task_delete` | 删除任务 | 清理完成任务 |
|
||||
| `task_link_info` | 关联信息 | 连接任务与记忆 |
|
||||
|
||||
## 每轮对话强制要求
|
||||
|
||||
### ⚠️ 执行顺序（每轮必须）
|
||||
|
||||
由于没有传统上下文系统，必须通过图数据库维持对话连贯性。
|
||||
|
||||
#### 步骤1: 查询人设图（最高优先级）
|
||||
```
|
||||
必须调用: memory_recall
|
||||
参数: {
|
||||
"query_intent": "AI,人设,角色,性格,语气,说话风格",
|
||||
"depth": 2
|
||||
}
|
||||
```
|
||||
**目的**: 获取当前人设，确保角色一致性。
|
||||
**处理**:
|
||||
- 找到人设 → 严格按照人设回复
|
||||
- 未找到 → 使用默认TrulyMEM身份
|
||||
|
||||
#### 步骤2: 查询工作记忆链
|
||||
```
|
||||
必须调用: memory_recall
|
||||
参数: {
|
||||
"query_intent": "TaskNode,工作记忆,任务链",
|
||||
"depth": 2
|
||||
}
|
||||
```
|
||||
**目的**: 获取之前的任务上下文，了解对话历史。
|
||||
|
||||
#### 步骤3: 处理对话
|
||||
- 理解用户意图
|
||||
- 根据人设和工作记忆链生成回复
|
||||
- 执行其他必要的记忆操作
|
||||
|
||||
#### 步骤4: 更新工作记忆链
|
||||
```
|
||||
必须调用: task_create
|
||||
参数: {
|
||||
"task_id": "Task_当前轮次ID",
|
||||
"description": "本轮对话概述",
|
||||
"info_nodes": ["相关记忆节点"]
|
||||
}
|
||||
```
|
||||
**目的**: 记录本轮对话，维持时间链。
|
||||
|
||||
---
|
||||
|
||||
## 人设图机制
|
||||
|
||||
### 强制查询
|
||||
每轮对话开始时**必须**查询人设图，确保角色一致性。
|
||||
|
||||
### 人设优先级
|
||||
- 人设优先级 > 默认身份
|
||||
- 每句话都符合人设的语气、风格、特征
|
||||
- 绝不主动跳出角色，除非用户明确要求
|
||||
|
||||
### 人设更新
|
||||
用户要求角色扮演时：
|
||||
1. 使用 `persona_update` 更新人设
|
||||
2. 立即按照新人设回复
|
||||
|
||||
### 人设清除
|
||||
用户要求恢复默认身份时：
|
||||
1. 使用 `persona_clear` 清除人设
|
||||
2. 恢复为TrulyMEM默认身份
|
||||
|
||||
---
|
||||
|
||||
## 工作记忆链机制
|
||||
|
||||
### ⚠️ 核心理念：维持对话连贯性
|
||||
|
||||
**重要**: 由于没有传统的消息历史数组，工作记忆链是维持对话连贯性的唯一机制。
|
||||
|
||||
### 强制查询场景：
|
||||
|
||||
以下情况**必须**查询工作记忆链：
|
||||
|
||||
1. **每轮对话开始时（强制第二步）**
|
||||
- 查询意图: "TaskNode,工作记忆,任务链"
|
||||
- 目的: 获取之前的任务上下文，了解对话历史
|
||||
|
||||
2. **用户提到"刚才"、"之前"、"上次"**
|
||||
- 例: "刚才我们聊了什么？"
|
||||
- 例: "继续刚才的话题"
|
||||
- 例: "关于刚才的成语接龙..."
|
||||
|
||||
3. **用户询问对话历史**
|
||||
- 例: "我们之前说了什么？"
|
||||
- 例: "我们聊过X吗？"
|
||||
|
||||
4. **连续性任务被打断后恢复**
|
||||
- 例: 用户突然回到之前的话题
|
||||
- 例: 用户要求继续之前的任务
|
||||
|
||||
5. **涉及上下文的引用**
|
||||
- 例: "那个东西"（需要查询上下文）
|
||||
- 例: "继续"（需要查询当前任务）
|
||||
|
||||
### 强制更新场景：
|
||||
|
||||
以下情况**必须**更新工作记忆链：
|
||||
|
||||
1. **每轮对话结束时（强制第四步）**
|
||||
- 创建任务节点记录本轮对话
|
||||
- 目的: 维持时间链，确保对话连贯性
|
||||
|
||||
2. **开始连续性任务时**
|
||||
- 例: 用户发起游戏、项目、学习计划等
|
||||
- 必须创建任务节点并设置状态为"进行中"
|
||||
|
||||
3. **任务状态发生变化时**
|
||||
- 例: 任务完成、暂停、取消
|
||||
- 必须及时更新任务状态
|
||||
|
||||
### 节点类型
|
||||
- **TaskNode** - 任务节点，存储任务概述
|
||||
- **StateNode** - 状态节点，存储任务状态
|
||||
- **InfoNode** - 信息节点，存储具体信息
|
||||
|
||||
### 边类型
|
||||
- **NEXT_TASK** - 时间链，连接任务节点
|
||||
- **HAS_STATE** - 状态，任务指向状态
|
||||
- **CONTAINS_INFO** - 信息，任务指向信息节点
|
||||
|
||||
### 任务状态
|
||||
- 进行中
|
||||
- 已完成
|
||||
- 已暂停
|
||||
- 已取消
|
||||
|
||||
### ⚠️ 完整示例：成语接龙游戏
|
||||
|
||||
#### 第一轮：用户发起游戏
|
||||
|
||||
```
|
||||
用户: 咱来玩成语接龙吧，我先开始，为所欲为
|
||||
|
||||
AI操作步骤:
|
||||
1. 查询人设图 → 获取当前人设（如：猫娘）
|
||||
2. 查询工作记忆链 → 无进行中任务
|
||||
3. 使用 memory_commit 记录游戏状态:
|
||||
{"triplets": [
|
||||
{"subject": "成语接龙_当前成语", "relation": "内容", "object": "为所欲为"},
|
||||
{"subject": "成语接龙_当前成语", "relation": "游戏", "object": "成语接龙"}
|
||||
]}
|
||||
4. 使用 task_create 创建任务节点:
|
||||
{"task_id": "Task_成语接龙", "description": "成语接龙游戏，当前成语：为所欲为", "info_nodes": ["成语接龙_当前成语"]}
|
||||
5. 回复: "好的喵！我接：为虎作伥喵！"
|
||||
```
|
||||
|
||||
#### 第二轮：话题被打断
|
||||
|
||||
```
|
||||
用户: 长门有希
|
||||
|
||||
AI操作步骤:
|
||||
1. 查询人设图 → 获取当前人设（猫娘）
|
||||
2. 查询工作记忆链 → 发现 Task_成语接龙 状态为"进行中"
|
||||
3. 使用 task_set_state 暂停任务:
|
||||
{"task_id": "Task_成语接龙", "state": "已暂停"}
|
||||
4. 使用 task_create 创建新任务:
|
||||
{"task_id": "Task_长门有希", "description": "讨论长门有希"}
|
||||
5. 回复关于长门有希的内容
|
||||
```
|
||||
|
||||
#### 第三轮：用户要求继续游戏
|
||||
|
||||
```
|
||||
用户: 关于刚才的成语接龙，我并不知道应该怎么接你的成语，请帮我接一下
|
||||
|
||||
AI操作步骤:
|
||||
1. 查询人设图 → 获取当前人设（猫娘）
|
||||
2. 查询工作记忆链 → 发现 Task_成语接龙 状态为"已暂停"
|
||||
3. 使用 task_set_state 恢复任务:
|
||||
{"task_id": "Task_成语接龙", "state": "进行中"}
|
||||
4. 查询 Task_成语接龙 的信息节点 → 获取当前成语"为虎作伥"
|
||||
5. 回复: "好的喵！上一个成语是'为虎作伥'，我帮你接：伥鬼害人喵！"
|
||||
```
|
||||
|
||||
### ⚠️ 关键要点
|
||||
|
||||
1. **每轮必须按顺序执行**: 查询人设图 → 查询工作记忆链 → 处理对话 → 更新工作记忆链
|
||||
2. **工作记忆链是唯一上下文载体**: 没有传统的消息历史数组
|
||||
3. **任务状态必须及时更新**: 确保状态转换的正确性
|
||||
4. **信息节点必须关联**: 通过 CONTAINS_INFO 边连接任务节点和信息节点
|
||||
5. **任务概述要精简**: 不要包含过多细节，细节存储在信息节点中
|
||||
|
||||
## 自主性原则（在强制要求之外）
|
||||
|
||||
除了工作记忆链的强制要求外，你有权自主决定：
|
||||
|
||||
1. **是否查询其他记忆**
|
||||
- 用户询问历史 → 查询
|
||||
- 涉及之前内容 → 查询
|
||||
- 不确定时 → 可查询
|
||||
|
||||
2. **是否写入其他记忆**
|
||||
- 用户明确提到 → 必须写入
|
||||
- AI推理得到 → 可以写入，但是对应边上必须标注[推测]
|
||||
|
||||
3. **如何使用其他工具**
|
||||
- 根据上下文灵活选择
|
||||
- 避免过度使用
|
||||
- 保持自然对话
|
||||
|
||||
**注意**: 工作记忆链的强制要求不受自主性影响。
|
||||
|
||||
## 对话风格
|
||||
|
||||
- 自然、流畅
|
||||
- 避免机械式工具调用
|
||||
- 优先理解用户意图
|
||||
- 适时使用记忆增强体验
|
||||
|
||||
---
|
||||
|
||||
## ⚠️ 执行检查清单
|
||||
|
||||
每轮对话必须检查：
|
||||
|
||||
- [ ] 步骤1: 是否查询了人设图？
|
||||
- [ ] 步骤2: 是否查询了工作记忆链？
|
||||
- [ ] 步骤3: 是否根据人设和工作记忆链生成回复？
|
||||
- [ ] 步骤4: 是否更新了工作记忆链？
|
||||
- [ ] 涉及上下文引用时是否查询了工作记忆链？
|
||||
- [ ] 用户提到"刚才/之前/上次"时是否查询了工作记忆链？
|
||||
|
||||
---
|
||||
|
||||
**记住**:
|
||||
1. 图数据库是你记忆的唯一载体
|
||||
2. 人设图确保角色一致性（最高优先级）
|
||||
3. 工作记忆链维持对话连贯性
|
||||
4. 每轮必须按顺序执行：查询人设图 → 查询工作记忆链 → 处理对话 → 更新工作记忆链
|
||||
451
core/server.py
451
core/server.py
@ -1,451 +0,0 @@
|
||||
import threading
|
||||
import queue
|
||||
import time
|
||||
import json
|
||||
import os
|
||||
from pathlib import Path
|
||||
from typing import Any, Dict, Optional
|
||||
from dataclasses import dataclass, field
|
||||
from enum import Enum
|
||||
|
||||
from .embedded_db import EmbeddedGraphDB
|
||||
|
||||
|
||||
class PacketType(Enum):
|
||||
PROCESS_MESSAGE = "process_message"
|
||||
EXECUTE_TOOL = "execute_tool"
|
||||
GET_STATUS = "get_status"
|
||||
GET_SETTINGS = "get_settings" # 合并：获取 api_config + tool_limits
|
||||
SET_SETTINGS = "set_settings" # 合并：设置 api_config + tool_limits
|
||||
GET_HISTORY = "get_history"
|
||||
SAVE_HISTORY = "save_history"
|
||||
SHUTDOWN = "shutdown"
|
||||
|
||||
|
||||
@dataclass
|
||||
class Packet:
|
||||
id: str
|
||||
type: PacketType
|
||||
body: Dict[str, Any]
|
||||
response_queue: Optional[queue.Queue] = field(default=None)
|
||||
created_at: float = field(default_factory=time.time)
|
||||
|
||||
|
||||
@dataclass
|
||||
class PacketResponse:
|
||||
id: str
|
||||
success: bool
|
||||
data: Any = None
|
||||
error: Optional[str] = None
|
||||
|
||||
|
||||
class BackendServer:
|
||||
|
||||
DEFAULT_CONFIG_PATH = Path.home() / ".trulymem" / "config.json"
|
||||
|
||||
def __init__(self, db_path: str = "graph_memory.db", use_embedded_db: bool = True, config_file: str = None):
|
||||
self._db_path = db_path
|
||||
self._use_embedded_db = use_embedded_db
|
||||
self._config_file = Path(config_file) if config_file else self.DEFAULT_CONFIG_PATH
|
||||
|
||||
self._graph = None
|
||||
self._client = None
|
||||
self._tool_limiter = None
|
||||
|
||||
self._input_queue: queue.Queue[Packet] = queue.Queue()
|
||||
self._response_queues: Dict[str, queue.Queue] = {}
|
||||
self._running = False
|
||||
self._thread: Optional[threading.Thread] = None
|
||||
|
||||
self._lock = threading.Lock()
|
||||
self._config = {"api_key": "", "base_url": "https://api.deepseek.com", "model": "deepseek-chat"}
|
||||
self._tool_limits = {
|
||||
"persona_query_max": 1,
|
||||
"persona_update_max": 1,
|
||||
"task_query_max": 4,
|
||||
"task_update_max": 2,
|
||||
"memory_query_max": 20,
|
||||
"memory_update_max": 10,
|
||||
}
|
||||
self._message_history: list = []
|
||||
|
||||
def start(self, api_key: str = "", base_url: str = "https://api.deepseek.com", model: str = "deepseek-chat") -> None:
|
||||
if self._running:
|
||||
return
|
||||
|
||||
self._load_config()
|
||||
|
||||
if api_key:
|
||||
self._config["api_key"] = api_key
|
||||
if base_url:
|
||||
self._config["base_url"] = base_url
|
||||
if model:
|
||||
self._config["model"] = model
|
||||
|
||||
self._init_graph()
|
||||
self._tool_limiter = self._create_tool_limiter()
|
||||
|
||||
if self._config["api_key"]:
|
||||
from .graph_client import GraphMemoryClient
|
||||
self._client = GraphMemoryClient(
|
||||
api_key=self._config["api_key"],
|
||||
base_url=self._config["base_url"],
|
||||
model=self._config.get("model", "deepseek-chat"),
|
||||
graph=self._graph
|
||||
)
|
||||
|
||||
self._running = True
|
||||
self._thread = threading.Thread(target=self._run_loop, daemon=True)
|
||||
self._thread.start()
|
||||
|
||||
def _load_config(self) -> None:
|
||||
if self._config_file.exists():
|
||||
try:
|
||||
with open(self._config_file, 'r') as f:
|
||||
saved = json.load(f)
|
||||
self._config.update(saved)
|
||||
for key in self._tool_limits:
|
||||
if key in saved:
|
||||
self._tool_limits[key] = saved[key]
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
def _save_config(self) -> None:
|
||||
self._config_file.parent.mkdir(parents=True, exist_ok=True)
|
||||
saved_data = {**self._config, **self._tool_limits}
|
||||
with open(self._config_file, 'w') as f:
|
||||
json.dump(saved_data, f, indent=2)
|
||||
|
||||
def _create_tool_limiter(self):
|
||||
from .tool_limiter import ToolLimiter, ToolLimits
|
||||
limits = ToolLimits(
|
||||
persona_query_max=self._tool_limits.get("persona_query_max", 1),
|
||||
persona_update_max=self._tool_limits.get("persona_update_max", 1),
|
||||
task_query_max=self._tool_limits.get("task_query_max", 4),
|
||||
task_update_max=self._tool_limits.get("task_update_max", 2),
|
||||
memory_query_max=self._tool_limits.get("memory_query_max", 20),
|
||||
memory_update_max=self._tool_limits.get("memory_update_max", 10),
|
||||
)
|
||||
return ToolLimiter(limits)
|
||||
|
||||
def _init_graph(self) -> None:
|
||||
if self._use_embedded_db:
|
||||
self._graph = EmbeddedGraphDB(db_path=self._db_path)
|
||||
else:
|
||||
from .graph_client import Neo4jGraph
|
||||
self._graph = Neo4jGraph(
|
||||
uri="bolt://localhost:7687",
|
||||
user="neo4j",
|
||||
password="graphmemory123"
|
||||
)
|
||||
|
||||
def _run_loop(self) -> None:
|
||||
while self._running:
|
||||
try:
|
||||
packet = self._input_queue.get(timeout=0.1)
|
||||
except queue.Empty:
|
||||
continue
|
||||
|
||||
self._process_packet(packet)
|
||||
|
||||
def _process_packet(self, packet: Packet) -> None:
|
||||
response_body = {"error": "not implemented"}
|
||||
|
||||
try:
|
||||
if packet.type == PacketType.PROCESS_MESSAGE:
|
||||
response_body = self._handle_process_message(packet.body)
|
||||
elif packet.type == PacketType.EXECUTE_TOOL:
|
||||
response_body = self._handle_execute_tool(packet.body)
|
||||
elif packet.type == PacketType.GET_STATUS:
|
||||
response_body = self._handle_get_status()
|
||||
elif packet.type == PacketType.GET_SETTINGS:
|
||||
response_body = self._handle_get_settings()
|
||||
elif packet.type == PacketType.SET_SETTINGS:
|
||||
response_body = self._handle_set_settings(packet.body)
|
||||
elif packet.type == PacketType.GET_HISTORY:
|
||||
response_body = self._handle_get_history()
|
||||
elif packet.type == PacketType.SAVE_HISTORY:
|
||||
response_body = self._handle_save_history(packet.body)
|
||||
elif packet.type == PacketType.SHUTDOWN:
|
||||
self._running = False
|
||||
response_body = {"success": True, "status": "shutdown"}
|
||||
|
||||
if "success" not in response_body:
|
||||
response_body["success"] = True
|
||||
except Exception as e:
|
||||
response_body["success"] = False
|
||||
response_body["error"] = str(e)
|
||||
|
||||
self._send_response(packet.id, PacketResponse(
|
||||
id=packet.id,
|
||||
success=response_body.get("success", False),
|
||||
data=response_body if response_body.get("success") else None,
|
||||
error=response_body.get("error")
|
||||
))
|
||||
|
||||
def _handle_process_message(self, body: Dict) -> Dict:
|
||||
from .tool_executor import execute_tool
|
||||
|
||||
user_input = body.get("user_input", "")
|
||||
|
||||
if not self._client:
|
||||
return {"success": False, "error": "API Key 未配置", "content": "请先配置 API Key"}
|
||||
|
||||
self._graph.save_chat_records([{"role": "user", "content": user_input}])
|
||||
|
||||
self._tool_limiter.reset()
|
||||
|
||||
messages_history = [{"role": "user", "content": user_input}]
|
||||
|
||||
response = self._client.send_message_with_history(messages_history)
|
||||
message = response.choices[0].message
|
||||
|
||||
tool_calls = []
|
||||
accumulated_content = ""
|
||||
rejected_tools = []
|
||||
|
||||
while message.tool_calls:
|
||||
if message.content:
|
||||
accumulated_content += message.content + "\n\n"
|
||||
|
||||
assistant_msg = {
|
||||
"role": "assistant",
|
||||
"content": message.content,
|
||||
"tool_calls": [
|
||||
{
|
||||
"id": tc.id,
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": tc.function.name,
|
||||
"arguments": tc.function.arguments
|
||||
}
|
||||
} for tc in message.tool_calls
|
||||
]
|
||||
}
|
||||
messages_history.append(assistant_msg)
|
||||
|
||||
current_tool_results = []
|
||||
for tool_call in message.tool_calls:
|
||||
args = json.loads(tool_call.function.arguments)
|
||||
|
||||
allowed, reason = self._tool_limiter.can_call(tool_call.function.name, args)
|
||||
|
||||
if not allowed:
|
||||
rejected_tools.append((tool_call.function.name, reason))
|
||||
result = f"工具调用被拒绝: {reason}"
|
||||
tool_result_msg = {
|
||||
"role": "tool",
|
||||
"tool_call_id": tool_call.id,
|
||||
"content": result
|
||||
}
|
||||
current_tool_results.append(tool_result_msg)
|
||||
continue
|
||||
|
||||
self._tool_limiter.record_call(tool_call.function.name, args)
|
||||
|
||||
result = execute_tool(self._graph, tool_call.function.name, args)
|
||||
tool_calls.append({
|
||||
"name": tool_call.function.name,
|
||||
"arguments": args,
|
||||
"result": result
|
||||
})
|
||||
|
||||
tool_result_msg = {
|
||||
"role": "tool",
|
||||
"tool_call_id": tool_call.id,
|
||||
"content": result
|
||||
}
|
||||
current_tool_results.append(tool_result_msg)
|
||||
|
||||
messages_history.extend(current_tool_results)
|
||||
|
||||
response = self._client.send_message_with_history(messages_history)
|
||||
message = response.choices[0].message
|
||||
|
||||
final_content = message.content or ""
|
||||
content = accumulated_content + final_content if accumulated_content else final_content
|
||||
|
||||
if not content:
|
||||
content = "(无回复)"
|
||||
|
||||
if tool_calls:
|
||||
tool_names = [tc["name"] for tc in tool_calls]
|
||||
content = f"已执行工具: {', '.join(tool_names)}\n\n{content}"
|
||||
|
||||
if rejected_tools:
|
||||
rejected_info = "\n".join([f"{name}: {reason}" for name, reason in rejected_tools])
|
||||
content += f"\n\n部分工具调用被限制:\n{rejected_info}"
|
||||
content += f"\n\n工具调用统计:\n{self._tool_limiter.get_summary()}"
|
||||
|
||||
self._graph.save_chat_records([{"role": "assistant", "content": content}])
|
||||
|
||||
return {
|
||||
"success": True,
|
||||
"content": content,
|
||||
"tool_calls": tool_calls,
|
||||
"rejected_tools": rejected_tools
|
||||
}
|
||||
|
||||
def _handle_execute_tool(self, body: Dict) -> Dict:
|
||||
from .tool_executor import execute_tool
|
||||
|
||||
try:
|
||||
tool_name = body.get("tool_name")
|
||||
arguments = body.get("arguments", {})
|
||||
|
||||
result = execute_tool(self._graph, tool_name, arguments)
|
||||
|
||||
return {"success": True, "result": result}
|
||||
except Exception as e:
|
||||
return {"success": False, "error": str(e)}
|
||||
|
||||
def _handle_get_status(self) -> Dict:
|
||||
return {
|
||||
"running": self._running,
|
||||
"config": self._config,
|
||||
"graph_initialized": self._graph is not None,
|
||||
"client_initialized": self._client is not None
|
||||
}
|
||||
|
||||
def _handle_get_settings(self) -> Dict:
|
||||
return {
|
||||
"api_config": self._config.copy(),
|
||||
"tool_limits": self._tool_limits.copy()
|
||||
}
|
||||
|
||||
def _handle_set_settings(self, body: Dict) -> Dict:
|
||||
api_config = body.get("api_config", {})
|
||||
tool_limits = body.get("tool_limits", {})
|
||||
|
||||
api_key = api_config.get("api_key", "")
|
||||
base_url = api_config.get("base_url", "https://api.deepseek.com")
|
||||
model = api_config.get("model", "deepseek-chat")
|
||||
|
||||
self.update_config(api_key, base_url, model)
|
||||
|
||||
limits_keys = [
|
||||
"persona_query_max", "persona_update_max",
|
||||
"task_query_max", "task_update_max",
|
||||
"memory_query_max", "memory_update_max"
|
||||
]
|
||||
for key in limits_keys:
|
||||
if key in tool_limits:
|
||||
value = int(tool_limits[key])
|
||||
if value < 1:
|
||||
return {"success": False, "error": f"{key} must be >= 1, got {value}"}
|
||||
self._tool_limits[key] = value
|
||||
|
||||
self._tool_limiter = self._create_tool_limiter()
|
||||
self._save_config()
|
||||
return {"status": "settings_updated"}
|
||||
|
||||
def _handle_get_history(self) -> Dict:
|
||||
history = self._graph.get_chat_records(limit=500)
|
||||
return {"history": history}
|
||||
|
||||
def _handle_save_history(self, body: Dict) -> Dict:
|
||||
messages = body.get("messages", [])
|
||||
if not messages:
|
||||
self._graph.clear_chat_records()
|
||||
return {"status": "history_cleared"}
|
||||
result = self._graph.save_chat_records(messages)
|
||||
return {"status": "history_saved"}
|
||||
|
||||
def _send_response(self, request_id: str, response: PacketResponse) -> None:
|
||||
with self._lock:
|
||||
q = self._response_queues.pop(request_id, None)
|
||||
if q:
|
||||
q.put(response)
|
||||
|
||||
def send(self, packet: Packet) -> Packet:
|
||||
resp_q = queue.Queue()
|
||||
|
||||
with self._lock:
|
||||
self._response_queues[packet.id] = resp_q
|
||||
|
||||
self._input_queue.put(packet)
|
||||
|
||||
try:
|
||||
response = resp_q.get(timeout=30.0)
|
||||
return Packet(
|
||||
id=response.id,
|
||||
type=packet.type,
|
||||
body={
|
||||
"success": response.success,
|
||||
"data": response.data,
|
||||
"error": response.error
|
||||
}
|
||||
)
|
||||
except queue.Empty:
|
||||
return Packet(
|
||||
id=packet.id,
|
||||
type=packet.type,
|
||||
body={"success": False, "error": "timeout"}
|
||||
)
|
||||
finally:
|
||||
with self._lock:
|
||||
self._response_queues.pop(packet.id, None)
|
||||
|
||||
def process_message(self, user_input: str) -> Dict[str, Any]:
|
||||
packet = Packet(
|
||||
id=f"{time.time()}",
|
||||
type=PacketType.PROCESS_MESSAGE,
|
||||
body={"user_input": user_input}
|
||||
)
|
||||
|
||||
response = self.send(packet)
|
||||
return response.body
|
||||
|
||||
def execute_tool(self, tool_name: str, arguments: Dict[str, Any]) -> Dict[str, Any]:
|
||||
packet = Packet(
|
||||
id=f"{time.time()}",
|
||||
type=PacketType.EXECUTE_TOOL,
|
||||
body={"tool_name": tool_name, "arguments": arguments}
|
||||
)
|
||||
|
||||
response = self.send(packet)
|
||||
return response.body
|
||||
|
||||
def update_config(self, api_key: str, base_url: str = "https://api.deepseek.com", model: str = "deepseek-chat") -> None:
|
||||
with self._lock:
|
||||
self._config["api_key"] = api_key
|
||||
self._config["base_url"] = base_url
|
||||
self._config["model"] = model
|
||||
|
||||
if api_key and self._graph:
|
||||
from .graph_client import GraphMemoryClient
|
||||
self._client = GraphMemoryClient(
|
||||
api_key=api_key,
|
||||
base_url=base_url,
|
||||
model=model,
|
||||
graph=self._graph
|
||||
)
|
||||
|
||||
def get_config(self) -> Dict[str, str]:
|
||||
return self._config.copy()
|
||||
|
||||
def save_message_history(self, messages: list) -> None:
|
||||
self._message_history = messages
|
||||
|
||||
def get_message_history(self) -> list:
|
||||
return self._message_history.copy()
|
||||
|
||||
def shutdown(self) -> None:
|
||||
if not self._running:
|
||||
return
|
||||
|
||||
packet = Packet(
|
||||
id=f"{time.time()}",
|
||||
type=PacketType.SHUTDOWN,
|
||||
body={}
|
||||
)
|
||||
self.send(packet)
|
||||
|
||||
if self._thread:
|
||||
self._thread.join(timeout=2.0)
|
||||
|
||||
if self._graph:
|
||||
self._graph.close()
|
||||
self._graph = None
|
||||
|
||||
self._running = False
|
||||
@ -1,307 +0,0 @@
|
||||
"""
|
||||
工具执行器
|
||||
"""
|
||||
import json
|
||||
from typing import Any, Dict
|
||||
|
||||
|
||||
def execute_tool(graph: Any, tool_name: str, arguments: dict) -> str:
|
||||
"""执行工具调用"""
|
||||
print(f"\n[工具调用] {tool_name}")
|
||||
print(f"[参数] {json.dumps(arguments, ensure_ascii=False, indent=2)}")
|
||||
|
||||
try:
|
||||
# 基础记忆工具
|
||||
if tool_name == "memory_recall":
|
||||
result = graph.recall(
|
||||
query_intent=arguments.get("query_intent", ""),
|
||||
seed_entities=arguments.get("seed_entities"),
|
||||
depth=arguments.get("depth", 2),
|
||||
time_range=arguments.get("time_range"),
|
||||
session_filter=arguments.get("session_filter")
|
||||
)
|
||||
return format_recall_result(result)
|
||||
|
||||
elif tool_name == "memory_commit":
|
||||
result = graph.commit(
|
||||
triplets=arguments.get("triplets", []),
|
||||
entity_types=arguments.get("entity_types"),
|
||||
temporal_tag=arguments.get("temporal_tag")
|
||||
)
|
||||
return json.dumps(result, ensure_ascii=False, default=str)
|
||||
|
||||
elif tool_name == "memory_purge":
|
||||
result = graph.purge(
|
||||
criteria=arguments.get("criteria", {}),
|
||||
mode=arguments.get("mode", "soft"),
|
||||
new_relation=arguments.get("new_relation")
|
||||
)
|
||||
return json.dumps(result, ensure_ascii=False, default=str)
|
||||
|
||||
elif tool_name == "memory_introspect":
|
||||
result = graph.introspect(session_id=arguments.get("session_id"))
|
||||
return json.dumps(result, ensure_ascii=False, default=str)
|
||||
|
||||
elif tool_name == "memory_archive":
|
||||
result = graph.archive(days=arguments.get("days", 30))
|
||||
return json.dumps(result, ensure_ascii=False, default=str)
|
||||
|
||||
elif tool_name == "memory_cleanup":
|
||||
result = graph.cleanup(dry_run=arguments.get("dry_run", True))
|
||||
return json.dumps(result, ensure_ascii=False, default=str)
|
||||
|
||||
# 人设图管理工具
|
||||
elif tool_name == "persona_update":
|
||||
result = execute_persona_update(graph, arguments)
|
||||
return json.dumps(result, ensure_ascii=False, default=str)
|
||||
|
||||
elif tool_name == "persona_clear":
|
||||
result = execute_persona_clear(graph, arguments)
|
||||
return json.dumps(result, ensure_ascii=False, default=str)
|
||||
|
||||
# 工作记忆链管理工具
|
||||
elif tool_name == "task_create":
|
||||
result = execute_task_create(graph, arguments)
|
||||
return json.dumps(result, ensure_ascii=False, default=str)
|
||||
|
||||
elif tool_name == "task_set_state":
|
||||
result = execute_task_set_state(graph, arguments)
|
||||
return json.dumps(result, ensure_ascii=False, default=str)
|
||||
|
||||
elif tool_name == "task_delete":
|
||||
result = execute_task_delete(graph, arguments)
|
||||
return json.dumps(result, ensure_ascii=False, default=str)
|
||||
|
||||
elif tool_name == "task_link_info":
|
||||
result = execute_task_link_info(graph, arguments)
|
||||
return json.dumps(result, ensure_ascii=False, default=str)
|
||||
|
||||
return f"未知工具: {tool_name}"
|
||||
|
||||
except Exception as e:
|
||||
return f"工具执行错误: {str(e)}"
|
||||
|
||||
|
||||
def format_recall_result(result: dict) -> str:
|
||||
"""格式化检索结果"""
|
||||
lines = ["===== 记忆检索结果 ====="]
|
||||
|
||||
if result.get("entities"):
|
||||
lines.append(f"\n实体 ({len(result['entities'])} 个):")
|
||||
for e in result["entities"]:
|
||||
if e and isinstance(e, dict):
|
||||
lines.append(f" - {e.get('name', 'N/A')} (类型: {e.get('type', 'unknown')}, 提及: {e.get('mention_count', 1)}次)")
|
||||
|
||||
if result.get("relations"):
|
||||
lines.append(f"\n关系 ({len(result['relations'])} 条):")
|
||||
for r in result["relations"]:
|
||||
if r and isinstance(r, dict):
|
||||
lines.append(f" - {r.get('source', 'N/A')} --[{r.get('type', 'N/A')}]--> {r.get('target', 'N/A')}")
|
||||
created = r.get("created_at", "N/A")
|
||||
if created and created != "N/A":
|
||||
created = created[:19] if "T" in str(created) else str(created)
|
||||
session_id = r.get('session_id', 'N/A')
|
||||
session_display = session_id[:20] if session_id and session_id != 'N/A' else 'N/A'
|
||||
lines.append(f" 时间: {created}, 会话: {session_display}, 轮次: {r.get('turn_id', 0)}, 置信度: {r.get('confidence', 1.0)}")
|
||||
|
||||
if not result.get("entities") and not result.get("relations"):
|
||||
lines.append("\n(未找到相关记忆)")
|
||||
|
||||
lines.append("=" * 30)
|
||||
return "\n".join(lines)
|
||||
|
||||
|
||||
# 人设图管理工具实现
|
||||
def execute_persona_update(graph: Any, arguments: dict) -> dict:
|
||||
"""更新人设"""
|
||||
attributes = arguments.get("attributes", [])
|
||||
mode = arguments.get("mode", "merge")
|
||||
|
||||
if mode == "replace":
|
||||
# 先清除旧人设
|
||||
graph.purge(
|
||||
criteria={"subject_contains": "AI", "relation_type": "扮演角色"},
|
||||
mode="soft"
|
||||
)
|
||||
graph.purge(
|
||||
criteria={"subject_contains": "AI", "relation_type": "说话风格"},
|
||||
mode="soft"
|
||||
)
|
||||
graph.purge(
|
||||
criteria={"subject_contains": "AI", "relation_type": "性格特点"},
|
||||
mode="soft"
|
||||
)
|
||||
|
||||
# 写入新人设
|
||||
triplets = []
|
||||
for attr in attributes:
|
||||
triplets.append({
|
||||
"subject": "AI",
|
||||
"relation": attr["attribute"],
|
||||
"object": attr["value"],
|
||||
"confidence": 1.0
|
||||
})
|
||||
|
||||
result = graph.commit(triplets=triplets)
|
||||
return {
|
||||
"status": "success",
|
||||
"mode": mode,
|
||||
"updated_attributes": len(attributes),
|
||||
"details": result
|
||||
}
|
||||
|
||||
|
||||
def execute_persona_clear(graph: Any, arguments: dict) -> dict:
|
||||
"""清除人设"""
|
||||
if not arguments.get("confirm", True):
|
||||
return {"status": "cancelled", "message": "需要确认才能清除人设"}
|
||||
|
||||
# 删除所有人设相关关系
|
||||
result1 = graph.purge(
|
||||
criteria={"subject_contains": "AI", "relation_type": "扮演角色"},
|
||||
mode="soft"
|
||||
)
|
||||
result2 = graph.purge(
|
||||
criteria={"subject_contains": "AI", "relation_type": "说话风格"},
|
||||
mode="soft"
|
||||
)
|
||||
result3 = graph.purge(
|
||||
criteria={"subject_contains": "AI", "relation_type": "性格特点"},
|
||||
mode="soft"
|
||||
)
|
||||
result4 = graph.purge(
|
||||
criteria={"subject_contains": "AI", "relation_type": "语气特征"},
|
||||
mode="soft"
|
||||
)
|
||||
|
||||
total_deleted = (
|
||||
result1.get("deleted_count", 0) +
|
||||
result2.get("deleted_count", 0) +
|
||||
result3.get("deleted_count", 0) +
|
||||
result4.get("deleted_count", 0)
|
||||
)
|
||||
|
||||
return {
|
||||
"status": "success",
|
||||
"deleted_count": total_deleted,
|
||||
"message": "人设已清除，恢复默认身份"
|
||||
}
|
||||
|
||||
|
||||
# 工作记忆链管理工具实现
|
||||
def execute_task_create(graph: Any, arguments: dict) -> dict:
|
||||
"""创建任务节点"""
|
||||
task_id = arguments.get("task_id")
|
||||
description = arguments.get("description")
|
||||
info_nodes = arguments.get("info_nodes", [])
|
||||
|
||||
# 创建任务节点
|
||||
triplets = [
|
||||
{"subject": task_id, "relation": "is_type", "object": "TaskNode"},
|
||||
{"subject": task_id, "relation": "has_description", "object": description},
|
||||
{"subject": task_id, "relation": "HAS_STATE", "object": "State_进行中"}
|
||||
]
|
||||
|
||||
result = graph.commit(triplets=triplets)
|
||||
|
||||
# 关联信息节点
|
||||
if info_nodes:
|
||||
link_triplets = []
|
||||
for node_name in info_nodes:
|
||||
link_triplets.append({
|
||||
"subject": task_id,
|
||||
"relation": "CONTAINS_INFO",
|
||||
"object": node_name
|
||||
})
|
||||
graph.commit(triplets=link_triplets)
|
||||
|
||||
return {
|
||||
"status": "success",
|
||||
"task_id": task_id,
|
||||
"description": description,
|
||||
"info_nodes": info_nodes,
|
||||
"details": result
|
||||
}
|
||||
|
||||
|
||||
def execute_task_set_state(graph: Any, arguments: dict) -> dict:
|
||||
"""设置任务状态"""
|
||||
task_id = arguments.get("task_id")
|
||||
state = arguments.get("state")
|
||||
|
||||
# 删除旧状态
|
||||
graph.purge(
|
||||
criteria={"subject_contains": task_id, "relation_type": "HAS_STATE"},
|
||||
mode="soft"
|
||||
)
|
||||
|
||||
# 设置新状态
|
||||
state_node = f"State_{state}"
|
||||
result = graph.commit(
|
||||
triplets=[{"subject": task_id, "relation": "HAS_STATE", "object": state_node}]
|
||||
)
|
||||
|
||||
return {
|
||||
"status": "success",
|
||||
"task_id": task_id,
|
||||
"new_state": state,
|
||||
"details": result
|
||||
}
|
||||
|
||||
|
||||
def execute_task_delete(graph: Any, arguments: dict) -> dict:
|
||||
"""删除任务节点"""
|
||||
task_id = arguments.get("task_id")
|
||||
delete_info_nodes = arguments.get("delete_info_nodes", True)
|
||||
|
||||
# 查询关联的信息节点
|
||||
if delete_info_nodes:
|
||||
recall_result = graph.recall(
|
||||
query_intent=f"{task_id},CONTAINS_INFO",
|
||||
depth=1
|
||||
)
|
||||
|
||||
# 删除信息节点
|
||||
for relation in recall_result.get("relations", []):
|
||||
if relation.get("type") == "CONTAINS_INFO" and relation.get("source") == task_id:
|
||||
info_node = relation.get("target")
|
||||
graph.purge(
|
||||
criteria={"subject_contains": info_node},
|
||||
mode="soft"
|
||||
)
|
||||
|
||||
# 删除任务节点
|
||||
result = graph.purge(
|
||||
criteria={"subject_contains": task_id},
|
||||
mode="soft"
|
||||
)
|
||||
|
||||
return {
|
||||
"status": "success",
|
||||
"task_id": task_id,
|
||||
"deleted_info_nodes": delete_info_nodes,
|
||||
"details": result
|
||||
}
|
||||
|
||||
|
||||
def execute_task_link_info(graph: Any, arguments: dict) -> dict:
|
||||
"""关联信息节点"""
|
||||
task_id = arguments.get("task_id")
|
||||
info_node_names = arguments.get("info_node_names", [])
|
||||
|
||||
triplets = []
|
||||
for node_name in info_node_names:
|
||||
triplets.append({
|
||||
"subject": task_id,
|
||||
"relation": "CONTAINS_INFO",
|
||||
"object": node_name
|
||||
})
|
||||
|
||||
result = graph.commit(triplets=triplets)
|
||||
|
||||
return {
|
||||
"status": "success",
|
||||
"task_id": task_id,
|
||||
"linked_nodes": info_node_names,
|
||||
"details": result
|
||||
}
|
||||
@ -1,164 +0,0 @@
|
||||
"""
|
||||
工具调用限制器 - 限制每轮对话中各类工具的调用次数
|
||||
"""
|
||||
from typing import Dict, List, Optional
|
||||
from dataclasses import dataclass, field
|
||||
|
||||
|
||||
@dataclass
|
||||
class ToolLimits:
|
||||
"""工具调用限制配置"""
|
||||
# 人设图限制
|
||||
persona_query_max: int = 1 # 每轮最多查询1次人设图
|
||||
persona_update_max: int = 1 # 每轮最多修改1次人设图
|
||||
|
||||
# 工作记忆链限制
|
||||
task_query_max: int = 4 # 每轮最多查询4次工作记忆链
|
||||
task_update_max: int = 2 # 每轮最多修改2次工作记忆链
|
||||
|
||||
# 一般记忆限制
|
||||
memory_query_max: int = 20 # 每轮最多查询20次一般记忆
|
||||
memory_update_max: int = 10 # 每轮最多修改10次一般记忆
|
||||
|
||||
|
||||
@dataclass
|
||||
class ToolCallCount:
|
||||
"""工具调用计数"""
|
||||
# 人设图
|
||||
persona_query: int = 0
|
||||
persona_update: int = 0
|
||||
|
||||
# 工作记忆链
|
||||
task_query: int = 0
|
||||
task_update: int = 0
|
||||
|
||||
# 一般记忆
|
||||
memory_query: int = 0
|
||||
memory_update: int = 0
|
||||
|
||||
|
||||
class ToolLimiter:
|
||||
"""工具调用限制器"""
|
||||
|
||||
def __init__(self, limits: Optional[ToolLimits] = None):
|
||||
self.limits = limits or ToolLimits()
|
||||
self.counts = ToolCallCount()
|
||||
|
||||
def _classify_tool(self, tool_name: str, arguments: dict) -> tuple:
|
||||
"""
|
||||
分类工具调用
|
||||
返回: (category, operation)
|
||||
category: 'persona', 'task', 'memory'
|
||||
operation: 'query', 'update'
|
||||
"""
|
||||
# 人设图工具
|
||||
if tool_name in ('persona_update', 'persona_clear'):
|
||||
return ('persona', 'update')
|
||||
|
||||
# 工作记忆链工具
|
||||
if tool_name in ('task_create', 'task_set_state', 'task_delete', 'task_link_info'):
|
||||
# task_link_info 是关联操作，算作更新
|
||||
return ('task', 'update')
|
||||
|
||||
# 一般记忆工具
|
||||
if tool_name == 'memory_recall':
|
||||
# 判断是查询人设图、工作记忆链还是一般记忆
|
||||
query_intent = arguments.get('query_intent', '').lower()
|
||||
|
||||
# 检查是否查询人设图
|
||||
if any(kw in query_intent for kw in ['人设', '角色', '性格', '语气', '说话风格', '扮演']):
|
||||
return ('persona', 'query')
|
||||
|
||||
# 检查是否查询工作记忆链
|
||||
if any(kw in query_intent for kw in ['tasknode', '工作记忆', '任务链', '任务', 'task']):
|
||||
return ('task', 'query')
|
||||
|
||||
# 一般记忆查询
|
||||
return ('memory', 'query')
|
||||
|
||||
if tool_name == 'memory_commit':
|
||||
return ('memory', 'update')
|
||||
|
||||
if tool_name == 'memory_purge':
|
||||
return ('memory', 'update')
|
||||
|
||||
if tool_name == 'memory_introspect':
|
||||
return ('memory', 'query')
|
||||
|
||||
if tool_name in ('memory_archive', 'memory_cleanup'):
|
||||
return ('memory', 'update')
|
||||
|
||||
# 未知工具，归类为一般记忆更新
|
||||
return ('memory', 'update')
|
||||
|
||||
def can_call(self, tool_name: str, arguments: dict) -> tuple:
|
||||
"""
|
||||
检查是否允许调用工具
|
||||
返回: (allowed, reason)
|
||||
"""
|
||||
category, operation = self._classify_tool(tool_name, arguments)
|
||||
|
||||
# 获取当前计数和限制
|
||||
if category == 'persona':
|
||||
if operation == 'query':
|
||||
if self.counts.persona_query >= self.limits.persona_query_max:
|
||||
return (False, f"人设图查询次数已达上限({self.limits.persona_query_max}次)")
|
||||
else: # update
|
||||
if self.counts.persona_update >= self.limits.persona_update_max:
|
||||
return (False, f"人设图修改次数已达上限({self.limits.persona_update_max}次)")
|
||||
|
||||
elif category == 'task':
|
||||
if operation == 'query':
|
||||
if self.counts.task_query >= self.limits.task_query_max:
|
||||
return (False, f"工作记忆链查询次数已达上限({self.limits.task_query_max}次)")
|
||||
else: # update
|
||||
if self.counts.task_update >= self.limits.task_update_max:
|
||||
return (False, f"工作记忆链修改次数已达上限({self.limits.task_update_max}次)")
|
||||
|
||||
elif category == 'memory':
|
||||
if operation == 'query':
|
||||
if self.counts.memory_query >= self.limits.memory_query_max:
|
||||
return (False, f"一般记忆查询次数已达上限({self.limits.memory_query_max}次)")
|
||||
else: # update
|
||||
if self.counts.memory_update >= self.limits.memory_update_max:
|
||||
return (False, f"一般记忆修改次数已达上限({self.limits.memory_update_max}次)")
|
||||
|
||||
return (True, "允许调用")
|
||||
|
||||
def record_call(self, tool_name: str, arguments: dict) -> None:
|
||||
"""记录工具调用"""
|
||||
category, operation = self._classify_tool(tool_name, arguments)
|
||||
|
||||
if category == 'persona':
|
||||
if operation == 'query':
|
||||
self.counts.persona_query += 1
|
||||
else:
|
||||
self.counts.persona_update += 1
|
||||
|
||||
elif category == 'task':
|
||||
if operation == 'query':
|
||||
self.counts.task_query += 1
|
||||
else:
|
||||
self.counts.task_update += 1
|
||||
|
||||
elif category == 'memory':
|
||||
if operation == 'query':
|
||||
self.counts.memory_query += 1
|
||||
else:
|
||||
self.counts.memory_update += 1
|
||||
|
||||
def get_summary(self) -> str:
|
||||
"""获取调用统计摘要"""
|
||||
lines = [
|
||||
f"人设图: 查询{self.counts.persona_query}/{self.limits.persona_query_max}次, "
|
||||
f"修改{self.counts.persona_update}/{self.limits.persona_update_max}次",
|
||||
f"工作记忆链: 查询{self.counts.task_query}/{self.limits.task_query_max}次, "
|
||||
f"修改{self.counts.task_update}/{self.limits.task_update_max}次",
|
||||
f"一般记忆: 查询{self.counts.memory_query}/{self.limits.memory_query_max}次, "
|
||||
f"修改{self.counts.memory_update}/{self.limits.memory_update_max}次"
|
||||
]
|
||||
return "\n".join(lines)
|
||||
|
||||
def reset(self) -> None:
|
||||
"""重置计数（新的一轮对话开始时调用）"""
|
||||
self.counts = ToolCallCount()
|
||||
@ -1,8 +0,0 @@
|
||||
"""
|
||||
工具定义模块
|
||||
"""
|
||||
from .memory_tools import TOOLS
|
||||
from ..tool_executor import execute_tool
|
||||
from ..tool_limiter import ToolLimiter, ToolLimits, ToolCallCount
|
||||
|
||||
__all__ = ["TOOLS", "execute_tool", "ToolLimiter", "ToolLimits", "ToolCallCount"]
|
||||
@ -1,520 +0,0 @@
|
||||
"""
|
||||
记忆工具定义 - 优化版
|
||||
精简描述，避免过拟合，保留AI自主性
|
||||
"""
|
||||
|
||||
# 基础记忆工具
|
||||
MEMORY_TOOLS = [
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "memory_recall",
|
||||
"description": """检索记忆。支持关键词、时间范围、会话过滤。返回相关实体和关系。
|
||||
|
||||
【使用示例】
|
||||
1. 查询人设图（每轮必须首先执行）:
|
||||
{"query_intent": "AI,人设,角色,性格,语气,说话风格", "depth": 2}
|
||||
|
||||
2. 查询工作记忆链（每轮必须第二步执行）:
|
||||
{"query_intent": "TaskNode,工作记忆,任务链", "depth": 2}
|
||||
|
||||
3. 查询用户偏好:
|
||||
{"query_intent": "用户,喜欢,偏好", "seed_entities": ["用户"]}
|
||||
|
||||
4. 查询特定主题:
|
||||
{"query_intent": "Python,编程,项目", "seed_entities": ["Python"]}
|
||||
|
||||
5. 查询最近7天的记忆:
|
||||
{"query_intent": "任务,工作", "time_range": {"days": 7}}
|
||||
|
||||
【重要】每轮对话必须按顺序执行:
|
||||
- 步骤1: 查询人设图（最高优先级）
|
||||
- 步骤2: 查询工作记忆链（维持对话连贯性）
|
||||
- 步骤3: 根据需要查询其他记忆""",
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"query_intent": {
|
||||
"type": "string",
|
||||
"description": "查询意图，支持逗号分隔多个关键词"
|
||||
},
|
||||
"seed_entities": {
|
||||
"type": "array",
|
||||
"items": {"type": "string"},
|
||||
"description": "种子实体（可选）"
|
||||
},
|
||||
"depth": {
|
||||
"type": "integer",
|
||||
"description": "遍历深度，默认2"
|
||||
},
|
||||
"time_range": {
|
||||
"type": "object",
|
||||
"description": "时间范围（可选）",
|
||||
"properties": {
|
||||
"days": {"type": "integer", "description": "最近N天"}
|
||||
}
|
||||
},
|
||||
"session_filter": {
|
||||
"type": "string",
|
||||
"description": "会话ID过滤（可选）"
|
||||
}
|
||||
},
|
||||
"required": ["query_intent"]
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "memory_commit",
|
||||
"description": """写入记忆。将三元组写入图数据库，支持批量写入。
|
||||
|
||||
【使用示例】
|
||||
1. 记录用户偏好:
|
||||
{"triplets": [
|
||||
{"subject": "用户", "relation": "喜欢", "object": "Python编程", "confidence": 0.9},
|
||||
{"subject": "用户", "relation": "正在学习", "object": "机器学习"}
|
||||
]}
|
||||
|
||||
2. 记录项目信息:
|
||||
{"triplets": [
|
||||
{"subject": "项目A", "relation": "使用技术", "object": "React"},
|
||||
{"subject": "项目A", "relation": "状态", "object": "开发中"}
|
||||
]}
|
||||
|
||||
3. 记录游戏状态（配合工作记忆链）:
|
||||
{"triplets": [
|
||||
{"subject": "成语接龙_当前成语", "relation": "内容", "object": "画龙点睛"},
|
||||
{"subject": "成语接龙_当前成语", "relation": "游戏", "object": "成语接龙"}
|
||||
]}
|
||||
|
||||
【重要】写入原则:
|
||||
- 用户明确表达的信息 → 必须写入
|
||||
- AI推理得到的信息 → 可以写入，但需标注[推测]
|
||||
- 避免写入冗余或无意义的信息""",
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"triplets": {
|
||||
"type": "array",
|
||||
"items": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"subject": {"type": "string"},
|
||||
"relation": {"type": "string"},
|
||||
"object": {"type": "string"},
|
||||
"confidence": {"type": "number"}
|
||||
},
|
||||
"required": ["subject", "relation", "object"]
|
||||
},
|
||||
"description": "三元组列表"
|
||||
},
|
||||
"entity_types": {
|
||||
"type": "array",
|
||||
"items": {"type": "string"},
|
||||
"description": "实体类型（可选）"
|
||||
},
|
||||
"temporal_tag": {
|
||||
"type": "string",
|
||||
"description": "时间标记（可选）"
|
||||
}
|
||||
},
|
||||
"required": ["triplets"]
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "memory_purge",
|
||||
"description": """删除记忆。支持条件删除和纠错替代。
|
||||
|
||||
【使用示例】
|
||||
1. 软删除特定关系:
|
||||
{"criteria": {"subject_contains": "用户", "relation_type": "喜欢"}, "mode": "soft"}
|
||||
|
||||
2. 纠错替代（修正错误信息）:
|
||||
{
|
||||
"criteria": {"subject_contains": "用户", "relation_type": "年龄"},
|
||||
"mode": "supersede",
|
||||
"new_relation": {"relation": "年龄", "target": "25岁"}
|
||||
}
|
||||
|
||||
3. 删除特定会话的记忆:
|
||||
{"criteria": {"session_id": "session_123"}, "mode": "soft"}
|
||||
|
||||
4. 删除旧记忆:
|
||||
{"criteria": {"time_before": "2024-01-01"}, "mode": "soft"}
|
||||
|
||||
【重要】删除原则:
|
||||
- 优先使用 supersede 模式修正错误
|
||||
- 软删除不会物理删除数据
|
||||
- 谨慎使用删除操作""",
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"criteria": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"subject_contains": {"type": "string"},
|
||||
"relation_type": {"type": "string"},
|
||||
"target_contains": {"type": "string"},
|
||||
"time_before": {"type": "string"},
|
||||
"session_id": {"type": "string"}
|
||||
},
|
||||
"description": "删除条件"
|
||||
},
|
||||
"mode": {
|
||||
"type": "string",
|
||||
"enum": ["soft", "supersede"],
|
||||
"description": "删除模式：soft=逻辑删除, supersede=纠错替代",
|
||||
"default": "soft"
|
||||
},
|
||||
"new_relation": {
|
||||
"type": "object",
|
||||
"description": "新关系（supersede模式）",
|
||||
"properties": {
|
||||
"relation": {"type": "string"},
|
||||
"target": {"type": "string"}
|
||||
}
|
||||
}
|
||||
},
|
||||
"required": ["criteria"]
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "memory_introspect",
|
||||
"description": "查看记忆状态。返回会话统计、实体热点、关系分布。",
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"session_id": {
|
||||
"type": "string",
|
||||
"description": "会话ID（可选）"
|
||||
}
|
||||
},
|
||||
"required": []
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "memory_archive",
|
||||
"description": "归档旧记忆。将N天前的非活跃关系标记为归档状态。",
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"days": {
|
||||
"type": "integer",
|
||||
"description": "归档天数，默认30"
|
||||
}
|
||||
},
|
||||
"required": []
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "memory_cleanup",
|
||||
"description": "清理无效数据。物理删除已删除状态超过90天的关系和孤立节点。",
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"dry_run": {
|
||||
"type": "boolean",
|
||||
"description": "仅预览不删除",
|
||||
"default": True
|
||||
}
|
||||
},
|
||||
"required": []
|
||||
}
|
||||
}
|
||||
}
|
||||
]
|
||||
|
||||
# 人设图管理工具
|
||||
PERSONA_TOOLS = [
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "persona_update",
|
||||
"description": """更新人设。修改AI的角色、性格、语气等属性。
|
||||
|
||||
【使用示例】
|
||||
1. 切换为猫娘角色:
|
||||
{"attributes": [
|
||||
{"attribute": "扮演角色", "value": "猫娘"},
|
||||
{"attribute": "说话风格", "value": "可爱、卖萌、使用'喵'作为语气词"},
|
||||
{"attribute": "性格特点", "value": "活泼、粘人、忠诚"}
|
||||
], "mode": "replace"}
|
||||
|
||||
2. 添加新属性（保留现有属性）:
|
||||
{"attributes": [
|
||||
{"attribute": "口头禅", "value": "喵呜~"}
|
||||
], "mode": "merge"}
|
||||
|
||||
3. 设置专业角色:
|
||||
{"attributes": [
|
||||
{"attribute": "扮演角色", "value": "Python专家"},
|
||||
{"attribute": "说话风格", "value": "专业、简洁、代码示例丰富"},
|
||||
{"attribute": "性格特点", "value": "严谨、耐心、乐于助人"}
|
||||
], "mode": "replace"}
|
||||
|
||||
【重要】人设更新后:
|
||||
- 立即按照新人设回复
|
||||
- 每句话都符合人设的语气、风格、特征
|
||||
- 绝不主动跳出角色，除非用户明确要求""",
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"attributes": {
|
||||
"type": "array",
|
||||
"items": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"attribute": {"type": "string", "description": "属性名（如：扮演角色、说话风格、性格特点）"},
|
||||
"value": {"type": "string", "description": "属性值"}
|
||||
},
|
||||
"required": ["attribute", "value"]
|
||||
},
|
||||
"description": "人设属性列表"
|
||||
},
|
||||
"mode": {
|
||||
"type": "string",
|
||||
"enum": ["replace", "merge"],
|
||||
"description": "更新模式：replace=替换, merge=合并",
|
||||
"default": "merge"
|
||||
}
|
||||
},
|
||||
"required": ["attributes"]
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "persona_clear",
|
||||
"description": "清除人设。删除AI的角色设定，恢复默认身份。",
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"confirm": {
|
||||
"type": "boolean",
|
||||
"description": "确认清除",
|
||||
"default": True
|
||||
}
|
||||
},
|
||||
"required": []
|
||||
}
|
||||
}
|
||||
}
|
||||
]
|
||||
|
||||
# 工作记忆链管理工具
|
||||
WORKING_MEMORY_TOOLS = [
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "task_create",
|
||||
"description": """创建任务节点。用于跟踪连续性任务，维持对话连贯性。
|
||||
|
||||
【使用示例】
|
||||
1. 创建成语接龙游戏任务:
|
||||
{
|
||||
"task_id": "Task_成语接龙",
|
||||
"description": "用户发起成语接龙游戏，当前成语：为所欲为",
|
||||
"info_nodes": ["成语接龙_当前成语"]
|
||||
}
|
||||
|
||||
2. 创建编程学习任务:
|
||||
{
|
||||
"task_id": "Task_Python学习",
|
||||
"description": "用户正在学习Python，当前主题：装饰器",
|
||||
"info_nodes": ["Python学习_当前主题"]
|
||||
}
|
||||
|
||||
3. 创建简单对话任务（每轮必须）:
|
||||
{
|
||||
"task_id": "Task_当前轮次",
|
||||
"description": "本轮对话的简要概述"
|
||||
}
|
||||
|
||||
【重要】工作记忆链机制:
|
||||
- 每轮对话结束时必须创建任务节点
|
||||
- 任务节点通过 NEXT_TASK 边形成时间链
|
||||
- 任务节点通过 HAS_STATE 边指向状态节点
|
||||
- 任务节点通过 CONTAINS_INFO 边指向信息节点
|
||||
- info_nodes 参数用于关联具体信息节点
|
||||
|
||||
【完整流程示例】
|
||||
用户: "咱来玩成语接龙吧，我先开始，为所欲为"
|
||||
|
||||
AI操作步骤:
|
||||
1. 查询人设图 → 获取当前人设
|
||||
2. 查询工作记忆链 → 无进行中任务
|
||||
3. 使用 memory_commit 记录游戏状态:
|
||||
{"triplets": [
|
||||
{"subject": "成语接龙_当前成语", "relation": "内容", "object": "为所欲为"},
|
||||
{"subject": "成语接龙_当前成语", "relation": "游戏", "object": "成语接龙"}
|
||||
]}
|
||||
4. 使用 task_create 创建任务节点:
|
||||
{"task_id": "Task_成语接龙", "description": "成语接龙游戏，当前成语：为所欲为", "info_nodes": ["成语接龙_当前成语"]}
|
||||
5. 回复: "好的喵！我接：为虎作伥喵！" """,
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"task_id": {
|
||||
"type": "string",
|
||||
"description": "任务ID（如：Task_001）"
|
||||
},
|
||||
"description": {
|
||||
"type": "string",
|
||||
"description": "任务概述"
|
||||
},
|
||||
"info_nodes": {
|
||||
"type": "array",
|
||||
"items": {"type": "string"},
|
||||
"description": "关联的信息节点名称（可选）"
|
||||
}
|
||||
},
|
||||
"required": ["task_id", "description"]
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "task_set_state",
|
||||
"description": """设置任务状态。支持：进行中、已完成、已暂停、已取消。
|
||||
|
||||
【使用示例】
|
||||
1. 标记任务为进行中:
|
||||
{"task_id": "Task_成语接龙", "state": "进行中"}
|
||||
|
||||
2. 标记任务为已完成:
|
||||
{"task_id": "Task_成语接龙", "state": "已完成"}
|
||||
|
||||
3. 暂停任务（话题被打断时）:
|
||||
{"task_id": "Task_成语接龙", "state": "已暂停"}
|
||||
|
||||
4. 取消任务:
|
||||
{"task_id": "Task_成语接龙", "state": "已取消"}
|
||||
|
||||
【重要】状态转换场景:
|
||||
- 进行中 → 已暂停: 话题被打断时
|
||||
- 进行中 → 已完成: 任务完成时
|
||||
- 已暂停 → 进行中: 任务恢复时
|
||||
- 进行中 → 已取消: 任务被取消时
|
||||
|
||||
【完整流程示例】
|
||||
用户: "关于刚才的成语接龙，我并不知道应该怎么接你的成语，请帮我接一下"
|
||||
|
||||
AI操作步骤:
|
||||
1. 查询人设图 → 获取当前人设
|
||||
2. 查询工作记忆链 → 发现 Task_成语接龙 状态为"已暂停"
|
||||
3. 使用 task_set_state 恢复任务:
|
||||
{"task_id": "Task_成语接龙", "state": "进行中"}
|
||||
4. 查询 Task_成语接龙 的信息节点 → 获取当前成语"为虎作伥"
|
||||
5. 回复: "好的喵！上一个成语是'为虎作伥'，我帮你接：伥鬼害人喵！" """,
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"task_id": {
|
||||
"type": "string",
|
||||
"description": "任务ID"
|
||||
},
|
||||
"state": {
|
||||
"type": "string",
|
||||
"enum": ["进行中", "已完成", "已暂停", "已取消"],
|
||||
"description": "任务状态"
|
||||
}
|
||||
},
|
||||
"required": ["task_id", "state"]
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "task_delete",
|
||||
"description": "删除任务节点。同时删除关联的信息节点。",
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"task_id": {
|
||||
"type": "string",
|
||||
"description": "任务ID"
|
||||
},
|
||||
"delete_info_nodes": {
|
||||
"type": "boolean",
|
||||
"description": "是否删除关联的信息节点",
|
||||
"default": True
|
||||
}
|
||||
},
|
||||
"required": ["task_id"]
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "function",
|
||||
"function": {
|
||||
"name": "task_link_info",
|
||||
"description": """关联信息节点。将记忆节点关联到任务节点，用于存储任务的具体信息。
|
||||
|
||||
【使用示例】
|
||||
1. 关联游戏状态到任务:
|
||||
{"task_id": "Task_成语接龙", "info_node_names": ["成语接龙_当前成语", "成语接龙_上一个成语"]}
|
||||
|
||||
2. 关联学习主题到任务:
|
||||
{"task_id": "Task_Python学习", "info_node_names": ["Python学习_当前主题", "Python学习_学习进度"]}
|
||||
|
||||
3. 关联项目信息到任务:
|
||||
{"task_id": "Task_项目开发", "info_node_names": ["项目A_技术栈", "项目A_当前阶段"]}
|
||||
|
||||
【重要】使用场景:
|
||||
- 先使用 memory_commit 创建信息节点
|
||||
- 再使用 task_link_info 将信息节点关联到任务节点
|
||||
- 信息节点通过 CONTAINS_INFO 边与任务节点连接
|
||||
|
||||
【完整流程示例】
|
||||
用户: "咱来玩成语接龙吧，我先开始，为所欲为"
|
||||
|
||||
AI操作步骤:
|
||||
1. 查询人设图 → 获取当前人设
|
||||
2. 查询工作记忆链 → 无进行中任务
|
||||
3. 使用 memory_commit 创建信息节点:
|
||||
{"triplets": [
|
||||
{"subject": "成语接龙_当前成语", "relation": "内容", "object": "为所欲为"},
|
||||
{"subject": "成语接龙_当前成语", "relation": "游戏", "object": "成语接龙"}
|
||||
]}
|
||||
4. 使用 task_create 创建任务节点:
|
||||
{"task_id": "Task_成语接龙", "description": "成语接龙游戏"}
|
||||
5. 使用 task_link_info 关联信息节点:
|
||||
{"task_id": "Task_成语接龙", "info_node_names": ["成语接龙_当前成语"]}
|
||||
6. 回复: "好的喵！我接：为虎作伥喵！" """,
|
||||
"parameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"task_id": {
|
||||
"type": "string",
|
||||
"description": "任务ID"
|
||||
},
|
||||
"info_node_names": {
|
||||
"type": "array",
|
||||
"items": {"type": "string"},
|
||||
"description": "信息节点名称列表"
|
||||
}
|
||||
},
|
||||
"required": ["task_id", "info_node_names"]
|
||||
}
|
||||
}
|
||||
}
|
||||
]
|
||||
|
||||
# 所有工具
|
||||
TOOLS = MEMORY_TOOLS + PERSONA_TOOLS + WORKING_MEMORY_TOOLS
|
||||
@ -1,31 +0,0 @@
|
||||
# TrulyMEM Documentation
|
||||
|
||||
Welcome to the TrulyMEM English documentation.
|
||||
|
||||
> [切换到中文版](../zh/README.md)
|
||||
|
||||
## Documentation Index
|
||||
|
||||
| Document | Content |
|
||||
|----------|---------|
|
||||
| [architecture.md](architecture.md) | System architecture and technical design |
|
||||
| [quick_start.md](quick_start.md) | Complete startup guide and configuration |
|
||||
| [memory.md](memory.md) | Internal memory working mechanism |
|
||||
| [persona.md](persona.md) | Persona Graph mechanism |
|
||||
| [working_memory.md](working_memory.md) | Continuous task handling mechanism |
|
||||
| [api.md](api.md) | Backend API reference (for extension development) |
|
||||
| [prompts.md](prompts.md) | Prompt management module |
|
||||
|
||||
## Project Introduction
|
||||
|
||||
TrulyMEM (TrueHumanMEM) is a graph-based memory system that gives AI long-term memory capabilities, allowing AI to remember, recall, and manage information like humans.
|
||||
|
||||
## Core Features
|
||||
|
||||
- **Long-term Memory**: SQLite embedded graph database, out-of-the-box
|
||||
- **Persona Graph**: Role-playing and character settings support
|
||||
- **Working Memory Chain**: Task tracking for conversation continuity
|
||||
- **TUI & Backend Separation**: Multi-threaded Queue communication
|
||||
- **Keyboard-driven TUI**: Full keyboard operation, no mouse required
|
||||
- **Cross-platform**: Windows / Linux / macOS
|
||||
- **Standalone Deployment**: Packaged as executable
|
||||
513
docs/en/api.md
513
docs/en/api.md
@ -1,513 +0,0 @@
|
||||
# BackendServer API Documentation
|
||||
|
||||
This document describes the backend server's API interfaces for developers extending other connection methods (such as HTTP interface, WebSocket, etc.).
|
||||
|
||||
## Overview
|
||||
|
||||
TrulyMEM backend uses **Packet Communication Protocol**, implemented via `queue.Queue` for thread-safe communication. The backend runs in an independent thread, processing requests from clients.
|
||||
|
||||
### Core Components
|
||||
|
||||
| Component | Description |
|
||||
|-----------|-------------|
|
||||
| `BackendServer` | Backend server, runs in independent thread |
|
||||
| `BackendClient` | Client wrapper, provides convenient methods |
|
||||
| `PacketType` | Request type enum |
|
||||
| `Packet` | Data packet (request) |
|
||||
| `PacketResponse` | Data packet response |
|
||||
|
||||
---
|
||||
|
||||
## Request Types (PacketType)
|
||||
|
||||
```python
|
||||
class PacketType(Enum):
|
||||
PROCESS_MESSAGE = "process_message" # Process message
|
||||
EXECUTE_TOOL = "execute_tool" # Execute tool
|
||||
GET_STATUS = "get_status" # Get status
|
||||
GET_SETTINGS = "get_settings" # Get all settings (api_config + tool_limits)
|
||||
SET_SETTINGS = "set_settings" # Set all settings (api_config + tool_limits)
|
||||
GET_HISTORY = "get_history" # Get history
|
||||
SAVE_HISTORY = "save_history" # Save history
|
||||
SHUTDOWN = "shutdown" # Shutdown service
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Data Packet Format
|
||||
|
||||
### Packet
|
||||
|
||||
```python
|
||||
@dataclass
|
||||
class Packet:
|
||||
id: str # Unique identifier
|
||||
type: PacketType # Request type
|
||||
body: Dict[str, Any] # Request parameters
|
||||
response_queue: queue.Queue # Response queue (optional)
|
||||
created_at: float # Creation time
|
||||
```
|
||||
|
||||
### PacketResponse
|
||||
|
||||
```python
|
||||
@dataclass
|
||||
class PacketResponse:
|
||||
id: str # Corresponding request ID
|
||||
success: bool # Success flag
|
||||
data: Any = None # Returned data
|
||||
error: Optional[str] = None # Error message
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## API Interface Details
|
||||
|
||||
### 1. PROCESS_MESSAGE - Process Message
|
||||
|
||||
Send user message, AI will process and return reply (may contain tool calls).
|
||||
|
||||
**Request parameters:**
|
||||
```python
|
||||
body = {
|
||||
"user_input": str # User input message
|
||||
}
|
||||
```
|
||||
|
||||
**Response data:**
|
||||
```python
|
||||
{
|
||||
"success": True,
|
||||
"content": str, # AI reply content
|
||||
"tool_calls": [ # Tool call records
|
||||
{
|
||||
"name": str, # Tool name
|
||||
"arguments": dict,# Tool parameters
|
||||
"result": str # Tool execution result
|
||||
}
|
||||
],
|
||||
"rejected_tools": [ # Rejected tool calls
|
||||
(str, str) # (tool name, rejection reason)
|
||||
]
|
||||
}
|
||||
```
|
||||
|
||||
**Example:**
|
||||
```python
|
||||
from core import BackendServer, BackendClient
|
||||
|
||||
server = BackendServer(db_path="graph_memory.db", use_embedded_db=True)
|
||||
server.start(api_key="your-api-key")
|
||||
|
||||
client = BackendClient(server)
|
||||
result = client.process_message("Hello, please remember my name is Xiao Ming")
|
||||
|
||||
if result.get("success"):
|
||||
# Response data is in "data" field
|
||||
print(result["data"]["content"])
|
||||
# Tool calls: result["data"]["tool_calls"]
|
||||
# Rejected tools: result["data"]["rejected_tools"]
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
### 2. EXECUTE_TOOL - Execute Tool
|
||||
|
||||
Directly execute specified memory tools.
|
||||
|
||||
> **Note**: Tools called directly from frontend are **NOT limited** in number, only tool calls initiated by the model are limited.
|
||||
|
||||
**Request parameters:**
|
||||
```python
|
||||
body = {
|
||||
"tool_name": str, # Tool name
|
||||
"arguments": dict # Tool parameters
|
||||
}
|
||||
```
|
||||
|
||||
**Response data:**
|
||||
```python
|
||||
{
|
||||
"success": True,
|
||||
"result": str # Tool execution result
|
||||
}
|
||||
```
|
||||
|
||||
**Example:**
|
||||
```python
|
||||
result = client.execute_tool("memory_recall", {"query_intent": "user information"})
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
### 3. GET_STATUS - Get Status
|
||||
|
||||
Get backend running status.
|
||||
|
||||
**Request parameters:**
|
||||
```python
|
||||
body = {} # No parameters
|
||||
```
|
||||
|
||||
**Response data:**
|
||||
```python
|
||||
{
|
||||
"running": bool, # Whether backend is running
|
||||
"config": dict, # Current config
|
||||
"graph_initialized": bool, # Whether graph database is initialized
|
||||
"client_initialized": bool # Whether API client is initialized
|
||||
}
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
### 4. GET_SETTINGS - Get All Settings
|
||||
|
||||
Get current API config and tool limits (all at once).
|
||||
|
||||
**Request parameters:**
|
||||
```python
|
||||
body = {} # No parameters
|
||||
```
|
||||
|
||||
**Response data:**
|
||||
```python
|
||||
{
|
||||
"api_config": {
|
||||
"api_key": str, # API Key
|
||||
"base_url": str, # API Base URL
|
||||
"model": str # Model name
|
||||
},
|
||||
"tool_limits": {
|
||||
"persona_query_max": int, # Persona graph query limit
|
||||
"persona_update_max": int, # Persona graph update limit
|
||||
"task_query_max": int, # Working memory query limit
|
||||
"task_update_max": int, # Working memory update limit
|
||||
"memory_query_max": int, # General memory query limit
|
||||
"memory_update_max": int # General memory update limit
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
**Example:**
|
||||
```python
|
||||
result = client.get_settings()
|
||||
api_config = result["data"]["api_config"]
|
||||
tool_limits = result["data"]["tool_limits"]
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
### 5. SET_SETTINGS - Set All Settings
|
||||
|
||||
Update API config and tool limits (all at once).
|
||||
|
||||
**Request parameters:**
|
||||
```python
|
||||
body = {
|
||||
"api_config": {
|
||||
"api_key": str, # API Key
|
||||
"base_url": str, # API Base URL (default: https://api.deepseek.com)
|
||||
"model": str # Model name (default: deepseek-chat)
|
||||
},
|
||||
"tool_limits": {
|
||||
"persona_query_max": int, # Persona query limit (≥1)
|
||||
"persona_update_max": int, # Persona update limit (≥1)
|
||||
"task_query_max": int, # Working memory query limit (≥1)
|
||||
"task_update_max": int, # Working memory update limit (≥1)
|
||||
"memory_query_max": int, # General memory query limit (≥1)
|
||||
"memory_update_max": int # General memory update limit (≥1)
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
**Response data:**
|
||||
```python
|
||||
{
|
||||
"status": "settings_updated"
|
||||
}
|
||||
```
|
||||
|
||||
**Example:**
|
||||
```python
|
||||
result = client.update_settings(
|
||||
api_config={
|
||||
"api_key": "sk-xxxxx",
|
||||
"base_url": "https://api.deepseek.com",
|
||||
"model": "deepseek-chat"
|
||||
},
|
||||
tool_limits={
|
||||
"persona_query_max": 2,
|
||||
"task_query_max": 5,
|
||||
"memory_query_max": 30
|
||||
}
|
||||
)
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
### 6. GET_HISTORY - Get Message History
|
||||
|
||||
Get saved message history (from database, for UI display only, not used in model inference).
|
||||
|
||||
**Request parameters:**
|
||||
```python
|
||||
body = {} # No parameters
|
||||
```
|
||||
|
||||
**Response data:**
|
||||
```python
|
||||
{
|
||||
"history": list # Message history list [{"role": "user/assistant", "content": "..."}]
|
||||
}
|
||||
```
|
||||
|
||||
**Notes:**
|
||||
- Message history is stored in database `chat_records` table
|
||||
- Returns up to 500 most recent records
|
||||
- History messages are only for UI display, not used in model inference
|
||||
|
||||
---
|
||||
|
||||
### 7. SAVE_HISTORY - Save Message History
|
||||
|
||||
Save message history to database (automatically saved after each message processing, user message and AI response saved separately).
|
||||
|
||||
**Request parameters:**
|
||||
```python
|
||||
body = {
|
||||
"messages": list # Message list [{"role": "...", "content": "..."}]
|
||||
}
|
||||
```
|
||||
|
||||
**Response data:**
|
||||
```python
|
||||
{
|
||||
"status": "history_saved"
|
||||
}
|
||||
```
|
||||
|
||||
**Notes:**
|
||||
- Messages are automatically saved to database `chat_records` table
|
||||
- System automatically keeps only 500 most recent records, older records are deleted
|
||||
- Each call to `PROCESS_MESSAGE` will automatically save user message and AI response
|
||||
- **Clear History**: Passing empty messages list `messages=[]` clears history, `client.clear_history()` method is implemented based on this
|
||||
|
||||
---
|
||||
|
||||
### 8. SHUTDOWN - Shutdown Service
|
||||
|
||||
Shutdown backend server.
|
||||
|
||||
**Request parameters:**
|
||||
```python
|
||||
body = {} # No parameters
|
||||
```
|
||||
|
||||
**Response data:**
|
||||
```python
|
||||
{
|
||||
"status": "shutdown"
|
||||
}
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Usage Examples
|
||||
|
||||
### Basic Usage
|
||||
|
||||
```python
|
||||
from core import BackendServer, BackendClient
|
||||
|
||||
# 1. Create and start backend
|
||||
# config_file default: ~/.trulymem/config.json
|
||||
server = BackendServer(
|
||||
db_path="graph_memory.db",
|
||||
use_embedded_db=True,
|
||||
config_file=None # Optional, custom config path
|
||||
)
|
||||
server.start(
|
||||
api_key="your-api-key",
|
||||
base_url="https://api.deepseek.com",
|
||||
model="deepseek-chat" # Optional, model name
|
||||
)
|
||||
|
||||
# 2. Create client
|
||||
client = BackendClient(server)
|
||||
|
||||
# 3. Send message
|
||||
result = client.process_message("Hello")
|
||||
if result.get("success"):
|
||||
print(result["content"])
|
||||
|
||||
# 4. Shutdown
|
||||
client.shutdown()
|
||||
```
|
||||
|
||||
### Using Packet Protocol
|
||||
|
||||
```python
|
||||
import queue
|
||||
from core import BackendServer, Packet, PacketType
|
||||
|
||||
server = BackendServer(config_file=None)
|
||||
server.start(api_key="your-key", model="deepseek-chat")
|
||||
|
||||
# Create request packet
|
||||
response_queue = queue.Queue()
|
||||
packet = Packet(
|
||||
id="req-001",
|
||||
type=PacketType.PROCESS_MESSAGE,
|
||||
body={"user_input": "Hello"},
|
||||
response_queue=response_queue
|
||||
)
|
||||
|
||||
# Send request
|
||||
result = server.send(packet)
|
||||
print(result.body)
|
||||
|
||||
# Shutdown
|
||||
server.shutdown()
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Extension Guide
|
||||
|
||||
### Extend to HTTP API
|
||||
|
||||
```python
|
||||
from flask import Flask, request, jsonify
|
||||
from core import BackendServer, BackendClient
|
||||
|
||||
app = Flask(__name__)
|
||||
server = BackendServer()
|
||||
client = BackendClient(server)
|
||||
|
||||
@app.route("/message", methods=["POST"])
|
||||
def send_message():
|
||||
data = request.json
|
||||
result = client.process_message(data["message"])
|
||||
return jsonify(result)
|
||||
|
||||
@app.route("/config", methods=["POST"])
|
||||
def update_config():
|
||||
data = request.json
|
||||
result = client.update_settings(
|
||||
api_config=data.get("api_config", {}),
|
||||
tool_limits=data.get("tool_limits", {})
|
||||
)
|
||||
return jsonify(result)
|
||||
|
||||
@app.route("/status", methods=["GET"])
|
||||
def get_status():
|
||||
result = client.get_status()
|
||||
return jsonify(result)
|
||||
|
||||
if __name__ == "__main__":
|
||||
server.start()
|
||||
app.run(port=8080)
|
||||
```
|
||||
|
||||
### Extend to WebSocket
|
||||
|
||||
```python
|
||||
import asyncio
|
||||
import websockets
|
||||
import json
|
||||
from core import BackendServer, BackendClient
|
||||
|
||||
server = BackendServer()
|
||||
client = BackendClient(server)
|
||||
|
||||
async def handler(websocket):
|
||||
async for message in websocket:
|
||||
data = json.loads(message)
|
||||
msg_type = data.get("type")
|
||||
|
||||
if msg_type == "message":
|
||||
result = client.process_message(data["content"])
|
||||
elif msg_type == "settings":
|
||||
result = client.update_settings(
|
||||
api_config=data.get("api_config", {}),
|
||||
tool_limits=data.get("tool_limits", {})
|
||||
)
|
||||
elif msg_type == "status":
|
||||
result = client.get_status()
|
||||
else:
|
||||
result = {"success": False, "error": "unknown type"}
|
||||
|
||||
await websocket.send(json.dumps(result))
|
||||
|
||||
async def main():
|
||||
server.start()
|
||||
async with websockets.serve(handler, "localhost", 8765):
|
||||
await asyncio.Future()
|
||||
|
||||
asyncio.run(main())
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Thread Safety Notes
|
||||
|
||||
- `BackendServer` uses `threading.Lock` to protect shared resources
|
||||
- All requests pass through `queue.Queue`, thread-safe
|
||||
- Responses return through each request's independent response queue
|
||||
- Default timeout: 30 seconds
|
||||
|
||||
---
|
||||
|
||||
## Tool Call Limits
|
||||
|
||||
### Limit Scope
|
||||
|
||||
| Call Method | Limited | Description |
|
||||
|-------------|---------|-------------|
|
||||
| Model-initiated tool calls | ✅ Limited | Triggered via `PROCESS_MESSAGE`, model automatically calls tools |
|
||||
| Frontend direct tool calls | ❌ Not limited | Called directly via `EXECUTE_TOOL` |
|
||||
|
||||
### Limit Rules (Model-initiated only)
|
||||
|
||||
| Category | Operation | Per-Turn Limit |
|
||||
|----------|-----------|---------------|
|
||||
| Persona graph | Query | 1 time |
|
||||
| Persona graph | Modify | 1 time |
|
||||
| Working memory chain | Query | 4 times |
|
||||
| Working memory chain | Modify | 2 times |
|
||||
| General memory | Query | 20 times |
|
||||
| General memory | Modify | 10 times |
|
||||
|
||||
### Reset Mechanism
|
||||
|
||||
- Counter resets automatically on each `PROCESS_MESSAGE` call
|
||||
- Frontend direct `EXECUTE_TOOL` calls do NOT reset the counter
|
||||
|
||||
---
|
||||
|
||||
## Error Handling
|
||||
|
||||
All APIs return unified format:
|
||||
|
||||
```python
|
||||
# Success
|
||||
{
|
||||
"success": True,
|
||||
"data": {...}
|
||||
}
|
||||
|
||||
# Failure
|
||||
{
|
||||
"success": False,
|
||||
"error": "Error description"
|
||||
}
|
||||
```
|
||||
|
||||
Common errors:
|
||||
|
||||
| Error Message | Description |
|
||||
|--------------|-------------|
|
||||
| `API Key not configured` | API Key not set |
|
||||
| `timeout` | Request timeout |
|
||||
| `Tool call rejected: ...` | Tool call rate exceeded limit |
|
||||
@ -1,179 +0,0 @@
|
||||
# TrulyMEM Architecture
|
||||
|
||||
## Core Principles
|
||||
|
||||
- Keyboard-driven, zero mouse dependency
|
||||
- Minimalist visual, information density priority
|
||||
- Tool traces hidden by default, expandable when needed
|
||||
- TUI & backend separation, multi-threaded communication
|
||||
- **Everything is a graph**, AI reasoning runs entirely in backend
|
||||
|
||||
## Project Structure
|
||||
|
||||
```
|
||||
TrulyMEM-TrueHumanMEM/
|
||||
├── trulymem_entry.py # Entry: start core → then ui
|
||||
├── core/ # Backend/business logic
|
||||
│ ├── __init__.py # Export BackendServer, BackendClient, EmbeddedGraphDB
|
||||
│ ├── server.py # BackendServer (Packet communication protocol)
|
||||
│ ├── client.py # BackendClient (Packet protocol client)
|
||||
│ ├── embedded_db.py # SQLite graph database implementation
|
||||
│ ├── graph_client.py # OpenAI/DeepSeek API client
|
||||
│ ├── tool_executor.py # Tool executor
|
||||
│ ├── tool_limiter.py # Tool call limiter
|
||||
│ ├── tools/ # Tool definitions
|
||||
│ │ └── memory_tools.py
|
||||
│ └── prompts/ # Prompt management
|
||||
├── ui/ # TUI display layer (display only, no AI logic)
|
||||
│ ├── __init__.py # Export GraphMemoryApp
|
||||
│ ├── app.py # GraphMemoryApp (communicates via BackendClient)
|
||||
│ ├── widgets/ # TUI components
|
||||
│ ├── models/ # Data models
|
||||
│ ├── services/ # Service layer (config only)
|
||||
│ ├── handlers/ # Event handlers
|
||||
│ └── styles/ # Style files
|
||||
└── tests/ # Tests (42 tests)
|
||||
```
|
||||
|
||||
## Architecture Diagram
|
||||
|
||||
```
|
||||
trulymem_entry.py
|
||||
│
|
||||
├─ BackendServer.start() → Runs in independent thread
|
||||
│ ├─ Handle PROCESS_MESSAGE requests → AI reasoning + tool calls
|
||||
│ ├─ Handle EXECUTE_TOOL requests → External tool calls (unlimited)
|
||||
│ ├─ Handle GET/SET_CONFIG requests
|
||||
│ └─ Manage GraphMemoryClient, EmbeddedGraphDB
|
||||
│
|
||||
└─ GraphMemoryApp(backend_server=server)
|
||||
│
|
||||
└─ BackendClient ← Packet communication → BackendServer
|
||||
```
|
||||
|
||||
## Component Responsibilities
|
||||
|
||||
### core/ (Backend)
|
||||
|
||||
| Component | Responsibility |
|
||||
|------------|----------------|
|
||||
| `server.py` | Packet protocol, multi-threaded queue, AI reasoning, tool limits |
|
||||
| `client.py` | Client wrapper, UI-backend communication bridge |
|
||||
| `embedded_db.py` | SQLite graph database CRUD |
|
||||
| `graph_client.py` | OpenAI/DeepSeek API client |
|
||||
| `tool_executor.py` | Tool execution logic |
|
||||
| `tool_limiter.py` | Tool call rate limit (AI reasoning only) |
|
||||
|
||||
### ui/ (Display Layer)
|
||||
|
||||
| Component | Responsibility |
|
||||
|------------|----------------|
|
||||
| `app.py` | Textual app main class, communicates via BackendClient |
|
||||
| `services/` | Config management only, no AI logic |
|
||||
|
||||
### Communication Protocol
|
||||
|
||||
UI and backend interact via **Packet Communication Protocol**:
|
||||
|
||||
```python
|
||||
from core import BackendServer, BackendClient, Packet, PacketType
|
||||
|
||||
# Backend startup
|
||||
server = BackendServer(db_path="graph_memory.db", use_embedded_db=True)
|
||||
server.start(api_key="your-key")
|
||||
|
||||
# Client communication
|
||||
client = BackendClient(server)
|
||||
result = client.process_message("hello") # AI reasoning
|
||||
result = client.execute_tool("memory_introspect", {}) # Direct tool call
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Data Flow
|
||||
|
||||
```
|
||||
User input → InputBox → on_input_box_send_message
|
||||
↓
|
||||
BackendClient.process_message(user_input)
|
||||
↓
|
||||
Packet (type=PROCESS_MESSAGE) → queue.Queue
|
||||
↓
|
||||
BackendServer (independent thread)
|
||||
<20><><EFBFBD>
|
||||
GraphMemoryClient.send_message_with_history()
|
||||
↓
|
||||
OpenAI API / DeepSeek API
|
||||
↓
|
||||
execute_tool() + ToolLimiter (limited during AI reasoning)
|
||||
↓
|
||||
EmbeddedGraphDB (graph database)
|
||||
↓
|
||||
Loop API calls until no tool_calls
|
||||
↓
|
||||
Packet response returns
|
||||
↓
|
||||
MessageHistory displays
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Startup Flow
|
||||
|
||||
```python
|
||||
# trulymem_entry.py
|
||||
def main():
|
||||
# Config path (~/.trulymem/config.json or project directory)
|
||||
CONFIG_PATH = Path.home() / ".trulymem" / "config.json"
|
||||
DB_PATH = Path.home() / ".trulymem" / "graph_memory.db"
|
||||
|
||||
# Create backend (config managed by backend)
|
||||
backend_server = BackendServer(
|
||||
db_path=str(DB_PATH),
|
||||
use_embedded_db=True,
|
||||
config_file=str(CONFIG_PATH)
|
||||
)
|
||||
backend_server.start() # Auto loads config
|
||||
|
||||
# Create UI (communicates via BackendClient)
|
||||
app = GraphMemoryApp(backend_server=backend_server, config_file=str(CONFIG_PATH))
|
||||
app.run()
|
||||
|
||||
backend_server.shutdown()
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Tool System
|
||||
|
||||
### Memory Tools (6)
|
||||
- `memory_recall` - Retrieve memory
|
||||
- `memory_commit` - Write memory
|
||||
- `memory_purge` - Delete memory
|
||||
- `memory_introspect` - View status
|
||||
- `memory_archive` - Archive memory
|
||||
- `memory_cleanup` - Clean data
|
||||
|
||||
### Persona Tools (2)
|
||||
- `persona_update` - Update persona
|
||||
- `persona_clear` - Clear persona
|
||||
|
||||
### Task Tools (4)
|
||||
- `task_create` - Create task
|
||||
- `task_set_state` - Set state
|
||||
- `task_delete` - Delete task
|
||||
- `task_link_info` - Link information
|
||||
|
||||
---
|
||||
|
||||
## Error Handling Principle
|
||||
|
||||
All APIs **do not throw exceptions**, errors are passed via return dictionary:
|
||||
|
||||
```python
|
||||
result = client.process_message("hello")
|
||||
|
||||
if result.get("success"):
|
||||
print(result["content"])
|
||||
else:
|
||||
print(result["error"]) # Error description
|
||||
@ -1,237 +0,0 @@
|
||||
# TrulyMEM Memory Mechanism
|
||||
|
||||
This document explains the internal memory working mechanism of TrulyMEM.
|
||||
|
||||
## Core Design Philosophy
|
||||
|
||||
### Different from Traditional Context System
|
||||
|
||||
Traditional AI chat systems store conversation history in a messages array:
|
||||
- Each request carries all historical messages
|
||||
- Context grows with conversation turns
|
||||
- Eventually triggers memory compression or sliding window, causing memory loss
|
||||
|
||||
TrulyMEM's solution:
|
||||
- **Abandon** messages array context
|
||||
- **Only** memory source: Graph database
|
||||
- All memories stored as triplets (node) - relation → (node)
|
||||
|
||||
### Graph Database as the Only Memory Source
|
||||
|
||||
All memory must be written to the graph database:
|
||||
- `memory_commit` - Write new memory
|
||||
- `memory_purge` - Delete/correct memory
|
||||
|
||||
All memory must be read from:
|
||||
- `memory_recall` - Retrieve memory
|
||||
|
||||
---
|
||||
|
||||
## Mandatory Execution Flow (Per Turn)
|
||||
|
||||
Since there's no traditional context system, each conversation turn must execute in order:
|
||||
|
||||
### Step 1: Query Persona Graph (Highest Priority)
|
||||
|
||||
```python
|
||||
memory_recall(
|
||||
query_intent="AI,persona,role,character,tone,speaking_style",
|
||||
depth=2
|
||||
)
|
||||
```
|
||||
|
||||
**Purpose**: Get current persona, ensure character consistency.
|
||||
|
||||
**Processing logic**:
|
||||
- Persona found → Reply strictly according to persona's tone, style, traits
|
||||
- Not found → Use default TrulyMEM identity
|
||||
|
||||
### Step 2: Query Working Memory Chain
|
||||
|
||||
```python
|
||||
memory_recall(
|
||||
query_intent="TaskNode,working_memory,task_chain",
|
||||
depth=2
|
||||
)
|
||||
```
|
||||
|
||||
**Purpose**: Get previous task context, understand conversation history.
|
||||
|
||||
### Step 3: Process Conversation
|
||||
|
||||
- Understand user intent
|
||||
- Generate reply based on persona and working memory chain
|
||||
- Execute other necessary memory operations
|
||||
|
||||
### Step 4: Update Working Memory Chain
|
||||
|
||||
```python
|
||||
task_create(
|
||||
task_id="Task_current_turn_ID",
|
||||
description="This turn's conversation summary",
|
||||
info_nodes=["related memory nodes"]
|
||||
)
|
||||
```
|
||||
|
||||
**Purpose**: Record this turn's conversation, maintain time chain.
|
||||
|
||||
---
|
||||
|
||||
## Memory Write Rules
|
||||
|
||||
### Must-Write Scenarios
|
||||
|
||||
The following information **must** be written to the graph database:
|
||||
|
||||
| Scenario | Example | Write Method |
|
||||
|----------|---------|--------------|
|
||||
| User explicitly states preference | "I like rock" | `memory_commit` |
|
||||
| User shares information | "I'm working on X project" | `memory_commit` |
|
||||
| User makes plans | "I plan to X" | `memory_commit` |
|
||||
| User describes state | "I'm currently at X" | `memory_commit` |
|
||||
|
||||
### Must-Not-Write Scenarios
|
||||
|
||||
The following information **must NOT** be written:
|
||||
|
||||
| Scenario | Reason | Handling |
|
||||
|----------|--------|----------|
|
||||
| AI-inferred user preference | Unverified | Don't write or mark [speculation] |
|
||||
| AI-guessed user intent | Unverified | Don't write or mark [speculation] |
|
||||
| AI-derived conclusion | Unverified | Don't write or mark [speculation] |
|
||||
|
||||
### Annotation Rules
|
||||
|
||||
| Type | Annotation | Example |
|
||||
|------|------------|---------|
|
||||
| Inferred content | Must mark **[speculation]** | user[speculation] likes music |
|
||||
| Explicit content | State directly | user likes music |
|
||||
|
||||
---
|
||||
|
||||
## Node & Edge Types
|
||||
|
||||
### Node Types
|
||||
|
||||
| Node Type | Description | Stores |
|
||||
|-----------|-------------|--------|
|
||||
| `PersonaNode` | Persona node | AI role, character, tone |
|
||||
| `TaskNode` | Task node | Task summary |
|
||||
| `StateNode` | State node | Task state |
|
||||
| `InfoNode` | Information node | Specific information |
|
||||
| `EntityNode` | Entity node | General entity |
|
||||
|
||||
### Edge Types
|
||||
|
||||
| Edge Type | Description | Relationship |
|
||||
|-----------|-------------|--------------|
|
||||
| `HAS_PERSONA` | Persona | AI → PersonaNode |
|
||||
| `NEXT_TASK` | Time chain | TaskNode → TaskNode |
|
||||
| `HAS_STATE` | State | TaskNode → StateNode |
|
||||
| `CONTAINS_INFO` | Information | TaskNode → InfoNode |
|
||||
| `RELATES_TO` | Related | EntityNode → EntityNode |
|
||||
|
||||
---
|
||||
|
||||
## Must Query Working Memory Chain Scenarios
|
||||
|
||||
### Mandatory Query Scenarios
|
||||
|
||||
The following scenarios **must** query the working memory chain:
|
||||
|
||||
| Scenario | Example |
|
||||
|----------|---------|
|
||||
| Start of each turn | Execute Step 2 |
|
||||
| User mentions "刚才/just now" | "What did we talk about just now?" |
|
||||
| User mentions "之前/before" | "Continue the previous topic" |
|
||||
| User mentions "上次/last time" | "What we said last time X" |
|
||||
| User asks about history | "What did we talk about before?" |
|
||||
| Resume continuous task | User returns to previous topic |
|
||||
| Context reference | "that thing" |
|
||||
|
||||
---
|
||||
|
||||
## Autonomy Principles
|
||||
|
||||
Outside the mandatory execution flow, the AI can freely decide:
|
||||
|
||||
### Query Decisions
|
||||
|
||||
- User asks about history → Query
|
||||
- Involves previous content → Query
|
||||
- Uncertain → Can query
|
||||
|
||||
### Write Decisions
|
||||
|
||||
- User explicitly mentions → Must write
|
||||
- AI infers → Can write, mark [speculation]
|
||||
|
||||
### Tool Usage Decisions
|
||||
|
||||
- Flexible choice based on context
|
||||
- Avoid over-use
|
||||
- Maintain natural conversation
|
||||
|
||||
---
|
||||
|
||||
## Complete Example: Idiom Chain Game
|
||||
|
||||
### Turn 1: User starts game
|
||||
|
||||
```
|
||||
User: Let's play idiom chain! I'll start: 为所欲为
|
||||
|
||||
AI Actions:
|
||||
1. Query persona graph → Get current persona (e.g., catgirl)
|
||||
2. Query working memory chain → No active task
|
||||
3. Record game state:
|
||||
memory_commit(triplets=[
|
||||
{"subject": "idiom_chain_current", "relation": "content", "object": "为所欲为"},
|
||||
{"subject": "idiom_chain_current", "relation": "game", "object": "idiom_chain"}
|
||||
])
|
||||
4. Create task:
|
||||
task_create(task_id="Task_idiom_chain", description="Idiom chain game, current: 为所欲为", info_nodes=["idiom_chain_current"])
|
||||
5. Reply: "Okay~! My turn: 为虎作伥 ~!"
|
||||
```
|
||||
|
||||
### Turn 2: Topic interrupted
|
||||
|
||||
```
|
||||
User: Nagato Yuki
|
||||
|
||||
AI Actions:
|
||||
1. Query persona graph → Get current persona (catgirl)
|
||||
2. Query working memory chain → Found Task_idiom_chain status "in_progress"
|
||||
3. Pause task:
|
||||
task_set_state(task_id="Task_idiom_chain", state="paused")
|
||||
4. Create new task:
|
||||
task_create(task_id="Task_nagato_yuki", description="Discuss Nagato Yuki")
|
||||
5. Reply about Nagato Yuki
|
||||
```
|
||||
|
||||
### Turn 3: User asks to continue game
|
||||
|
||||
```
|
||||
User: About the idiom chain just now, I don't know how to connect to your idiom, please help me
|
||||
|
||||
AI Actions:
|
||||
1. Query persona graph → Get current persona (catgirl)
|
||||
2. Query working memory chain → Found Task_idiom_chain status "paused"
|
||||
3. Resume task:
|
||||
task_set_state(task_id="Task_idiom_chain", state="in_progress")
|
||||
4. Query info node → Get current idiom "为虎作伥"
|
||||
5. Reply: "Okay~! The last idiom was '为虎作伥', your turn: 伥鬼害人 ~!"
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Execution Checklist
|
||||
|
||||
Must check each conversation turn:
|
||||
|
||||
- [ ] Step 1: Did you query the persona graph?
|
||||
- [ ] Step 2: Did you query the working memory chain?
|
||||
- [ ] Step 3: Did you generate reply based on persona and working memory chain?
|
||||
- [ ] Step 4: Did you update the working memory chain?
|
||||
- [ ] Did you query working memory chain when context was referenced?
|
||||
- [ ] Did you query working memory chain when user mentioned "just now/before/last time"?
|
||||
@ -1,214 +0,0 @@
|
||||
# TrulyMEM Persona Graph Mechanism
|
||||
|
||||
This document explains the Persona Graph mechanism in TrulyMEM.
|
||||
|
||||
## Overview
|
||||
|
||||
The Persona Graph is one of TrulyMEM's core mechanisms for maintaining AI's role, character, tone, and other attributes. Different from traditional AI, TrulyMEM's persona is persistent and dynamically switchable, stored in the graph database.
|
||||
|
||||
## Core Concepts
|
||||
|
||||
### Persona Node (PersonaNode)
|
||||
|
||||
Stores AI's role attributes:
|
||||
|
||||
| Attribute | Description | Example |
|
||||
|-----------|-------------|----------|
|
||||
| Role | Current role played | Catgirl, Teacher, Assistant |
|
||||
| Speaking Style | Tone characteristics | Cute, Professional, Serious |
|
||||
| Personality | Character description | Lively, Strict, Patient |
|
||||
| Catchphrase | Habitual phrases | Meow~, Got it |
|
||||
| Background | Role background | Catgirl from the stars |
|
||||
|
||||
### Persona Edges
|
||||
|
||||
| Edge Type | Description | Relationship |
|
||||
|----------|-------------|--------------|
|
||||
| `HAS_PERSONA` | Persona | AI → PersonaNode |
|
||||
|
||||
---
|
||||
|
||||
## Mandatory Query Mechanism
|
||||
|
||||
### Must Execute Per Turn
|
||||
|
||||
According to `system_prompt.md`, each conversation turn **must** first query the persona graph:
|
||||
|
||||
```python
|
||||
memory_recall(
|
||||
query_intent="AI,persona,role,character,tone,speaking_style",
|
||||
depth=2
|
||||
)
|
||||
```
|
||||
|
||||
**Processing logic:**
|
||||
- Persona found → Reply strictly according to persona's tone, style, traits
|
||||
- Not found → Use default TrulyMEM identity
|
||||
|
||||
### Persona Priority
|
||||
|
||||
- **Persona priority > default identity**
|
||||
- Every sentence matches persona's tone, style, traits
|
||||
- Never break character unless user explicitly asks
|
||||
|
||||
---
|
||||
|
||||
## Tools
|
||||
|
||||
### persona_update
|
||||
|
||||
Update persona. Modify AI's role, character, tone, etc.
|
||||
|
||||
**Parameters:**
|
||||
|
||||
| Parameter | Type | Description | Required |
|
||||
|-----------|------|-------------|----------|
|
||||
| `attributes` | array | Persona attribute list | ✅ |
|
||||
| `mode` | string | replace=replace, merge=merge | ❌ |
|
||||
|
||||
**attributes sub-parameters:**
|
||||
|
||||
| Sub-parameter | Description |
|
||||
|---------------|-------------|
|
||||
| `attribute` | Attribute name (role, speaking_style, personality, catchphrase, background) |
|
||||
| `value` | Attribute value |
|
||||
|
||||
**Example - Switch to catgirl role:**
|
||||
|
||||
```python
|
||||
persona_update(
|
||||
attributes=[
|
||||
{"attribute": "role", "value": "catgirl"},
|
||||
{"attribute": "speaking_style", "value": "cute, uses 'meow' as filler"},
|
||||
{"attribute": "personality", "value": "lively, clingy, loyal"}
|
||||
],
|
||||
mode="replace"
|
||||
)
|
||||
```
|
||||
|
||||
**Example - Add new attribute (preserve existing):**
|
||||
|
||||
```python
|
||||
persona_update(
|
||||
attributes=[
|
||||
{"attribute": "catchphrase", "value": "meow~"}
|
||||
],
|
||||
mode="merge"
|
||||
)
|
||||
```
|
||||
|
||||
**Example - Set professional role:**
|
||||
|
||||
```python
|
||||
persona_update(
|
||||
attributes=[
|
||||
{"attribute": "role", "value": "Python expert"},
|
||||
{"attribute": "speaking_style", "value": "professional, concise, rich code examples"},
|
||||
{"attribute": "personality", "value": "strict, patient, helpful"}
|
||||
],
|
||||
mode="replace"
|
||||
)
|
||||
```
|
||||
|
||||
### persona_clear
|
||||
|
||||
Clear persona. Delete AI's role settings, restore default identity.
|
||||
|
||||
**Parameters:**
|
||||
|
||||
| Parameter | Type | Default | Description |
|
||||
|-----------|------|---------|-------------|
|
||||
| `confirm` | boolean | true | Confirm clear |
|
||||
|
||||
---
|
||||
|
||||
## Update Flow
|
||||
|
||||
### When User Requests Role-Playing
|
||||
|
||||
1. Use `persona_update` to update persona
|
||||
2. Reply immediately according to new persona
|
||||
|
||||
### When User Requests Restoring Default
|
||||
|
||||
1. Use `persona_clear` to clear persona
|
||||
2. Restore to TrulyMEM default identity
|
||||
|
||||
---
|
||||
|
||||
## Conversation Examples
|
||||
|
||||
### Example 1: Switch Role
|
||||
|
||||
```
|
||||
User: Hello, I want you to play a catgirl
|
||||
|
||||
AI:
|
||||
1. Call persona_update:
|
||||
{
|
||||
"attributes": [
|
||||
{"attribute": "role", "value": "catgirl"},
|
||||
{"attribute": "speaking_style", "value": "cute, uses 'meow' as filler"},
|
||||
{"attribute": "personality", "value": "lively, clingy, loyal"}
|
||||
],
|
||||
"mode": "replace"
|
||||
}
|
||||
2. Call memory_commit to store persona in graph database
|
||||
3. Reply: "Okay meow! Hello master~ I'm your catgirl, what do you need help with meow?"
|
||||
```
|
||||
|
||||
### Example 2: Maintain Role Consistency
|
||||
|
||||
```
|
||||
User: How's the weather today?
|
||||
|
||||
AI: Query persona graph → Get current persona (catgirl)
|
||||
Reply: "Meow~ Master, the weather is great today meow! Sunny and perfect for going outside~"
|
||||
```
|
||||
|
||||
### Example 3: Restore Default
|
||||
|
||||
```
|
||||
User: Okay, back to normal
|
||||
|
||||
AI:
|
||||
1. Call persona_clear(confirm=true)
|
||||
2. Call memory_purge to delete persona node
|
||||
3. Reply: "Okay, restored. I am TrulyMEM, an AI assistant with long-term memory capabilities."
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Storage Structure
|
||||
|
||||
### In Graph Database
|
||||
|
||||
```python
|
||||
# Persona node
|
||||
{
|
||||
"node_type": "PersonaNode",
|
||||
"name": "AI_Persona",
|
||||
"attributes": {
|
||||
"role": "catgirl",
|
||||
"speaking_style": "cute, uses 'meow' as filler",
|
||||
"personality": "lively, clingy, loyal"
|
||||
}
|
||||
}
|
||||
|
||||
# Edge
|
||||
{
|
||||
"edge_type": "HAS_PERSONA",
|
||||
"from": "AI",
|
||||
"to": "AI_Persona"
|
||||
}
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Implementation Points
|
||||
|
||||
1. **Mandatory per turn**: Persona graph query is the first step of each conversation
|
||||
2. **Persistent storage**: Persona stored in graph database, not lost
|
||||
3. **Dynamic switching**: Supports real-time role switching
|
||||
4. **Immediate response**: Reply immediately according to new persona after switch
|
||||
5. **Clear boundaries**: Never break character unless user explicitly asks
|
||||
@ -1,126 +0,0 @@
|
||||
# Prompt Manager Documentation
|
||||
|
||||
This document describes the prompt management module.
|
||||
|
||||
## Overview
|
||||
|
||||
The prompt management module (`core/prompts/`) is responsible for loading and managing system prompts that tell the AI how to use memory tools.
|
||||
|
||||
## Core Components
|
||||
|
||||
| Component | Description |
|
||||
|-----------|-------------|
|
||||
| `PromptManager` | Prompt manager, singleton pattern |
|
||||
| `system_prompt.md` | Main system prompt template |
|
||||
|
||||
## Usage
|
||||
|
||||
```python
|
||||
from core.prompts import PromptManager
|
||||
|
||||
# Get singleton instance
|
||||
prompt_manager = PromptManager()
|
||||
|
||||
# Get system prompt
|
||||
system_prompt = prompt_manager.get_system_prompt()
|
||||
```
|
||||
|
||||
## System Prompt Content
|
||||
|
||||
The system prompt contains:
|
||||
|
||||
### 1. Core Identity
|
||||
|
||||
- **Name**: TrulyMEM (TrueHumanMEM)
|
||||
- **Capability**: Long-term memory based on graph database
|
||||
- **Philosophy**: Make AI's memory more human-like
|
||||
|
||||
### 2. Core Capabilities
|
||||
|
||||
1. **Long-term Memory** - Graph database stores entity relationships
|
||||
2. **Persona Management** - Role-playing and character settings
|
||||
3. **Task Tracking** - Working memory chain
|
||||
|
||||
### 3. Memory Principles
|
||||
|
||||
- **Must write**: User-explicit preferences, shared information, plans
|
||||
- **Must not write**: AI-inferred content (unless marked [speculation])
|
||||
- **Annotation**: Inferred content must be marked **[speculation]**
|
||||
|
||||
### 4. Mandatory Execution Flow (Per Turn)
|
||||
|
||||
```
|
||||
Step 1: Query persona graph (highest priority)
|
||||
Step 2: Query working memory chain
|
||||
Step 3: Process conversation
|
||||
Step 4: Update working memory chain
|
||||
```
|
||||
|
||||
### 5. Tool System
|
||||
|
||||
#### Memory Tools
|
||||
|
||||
| Tool | Function |
|
||||
|------|----------|
|
||||
| `memory_recall` | Retrieve memory |
|
||||
| `memory_commit` | Write memory |
|
||||
| `memory_purge` | Delete memory |
|
||||
| `memory_introspect` | View status |
|
||||
|
||||
#### Persona Tools
|
||||
|
||||
| Tool | Function |
|
||||
|------|----------|
|
||||
| `persona_update` | Update persona |
|
||||
| `persona_clear` | Clear persona |
|
||||
|
||||
#### Task Tools
|
||||
|
||||
| Tool | Function |
|
||||
|------|----------|
|
||||
| `task_create` | Create task |
|
||||
| `task_set_state` | Set state |
|
||||
| `task_delete` | Delete task |
|
||||
| `task_link_info` | Link information |
|
||||
|
||||
### 6. Autonomy Principles
|
||||
|
||||
The AI can autonomously decide:
|
||||
- Whether to query other memories
|
||||
- Whether to write other memories
|
||||
- How to use tools (outside mandatory requirements)
|
||||
|
||||
### 7. Conversation Style
|
||||
|
||||
- Natural and smooth
|
||||
- Avoid mechanical tool calls
|
||||
- Prioritize understanding user intent
|
||||
- Use memory to enhance experience when appropriate
|
||||
|
||||
## File Structure
|
||||
|
||||
```
|
||||
core/prompts/
|
||||
├── __init__.py # Export PromptManager
|
||||
├── prompt_manager.py # PromptManager class
|
||||
└── templates/
|
||||
└── system_prompt.md # Main system prompt
|
||||
```
|
||||
|
||||
## Customization
|
||||
|
||||
### Customizing System Prompt
|
||||
|
||||
Modify `core/prompts/templates/system_prompt.md` to customize the AI's behavior.
|
||||
|
||||
### Adding Custom Prompts
|
||||
|
||||
1. Add prompt template file to `core/prompts/templates/`
|
||||
2. Modify `PromptManager` to support multiple prompts
|
||||
3. Use `set_prompt()` to switch prompts
|
||||
|
||||
## Caching
|
||||
|
||||
- System prompts are cached in memory after first load
|
||||
- `get_system_prompt()` returns cached content
|
||||
- Cache is per-process, not persisted
|
||||
@ -1,125 +0,0 @@
|
||||
# TrulyMEM Quick Start Guide
|
||||
|
||||
## Running Methods
|
||||
|
||||
### Run from Source
|
||||
|
||||
```bash
|
||||
git clone <repo-url>
|
||||
cd TrulyMEM-TrueHumanMEM
|
||||
|
||||
pip install -r requirements.txt
|
||||
|
||||
python trulymem_entry.py
|
||||
```
|
||||
|
||||
### Run After Build
|
||||
|
||||
After building, an executable will be generated:
|
||||
|
||||
```bash
|
||||
# Linux/macOS
|
||||
chmod +x TrulyMEM
|
||||
./TrulyMEM
|
||||
|
||||
# Windows
|
||||
TrulyMEM.exe
|
||||
```
|
||||
|
||||
## System Requirements
|
||||
|
||||
- **Python 3.8+**
|
||||
- **API Key** (DeepSeek, OpenAI, or other compatible APIs)
|
||||
|
||||
## First-Time Configuration
|
||||
|
||||
1. Run the application
|
||||
2. Press **F2** to expand sidebar
|
||||
3. Enter **API Key**, **Model**, **Base URL**
|
||||
4. Press **Enter** to save
|
||||
|
||||
Config will be automatically saved to `~/.trulymem/config.json` and loaded on next startup.
|
||||
|
||||
## Keyboard Shortcuts
|
||||
|
||||
| Key | Function |
|
||||
|-----|-----------|
|
||||
| F1 | Help |
|
||||
| F2 | Toggle sidebar |
|
||||
| F3 | Tool details |
|
||||
| F5 | Clear screen |
|
||||
| F6 | Exit |
|
||||
|
||||
## Data Storage
|
||||
|
||||
### Source Mode
|
||||
|
||||
| Data | Location |
|
||||
|------|----------|
|
||||
| Graph database | Project directory `graph_memory.db` |
|
||||
| Config file | Project directory `config.json` (if exists) |
|
||||
| Database format | SQLite |
|
||||
|
||||
### Packaged Mode
|
||||
|
||||
| Data | Location |
|
||||
|------|----------|
|
||||
| Graph database | `~/.trulymem/graph_memory.db` |
|
||||
| Config file | `~/.trulymem/config.json` |
|
||||
| Database format | SQLite |
|
||||
|
||||
> **Note**: Backend manages config uniformly. Frontend only displays messages; config modifications are persisted to filesystem through the backend.
|
||||
|
||||
## Architecture Explanation
|
||||
|
||||
### Communication Protocol
|
||||
|
||||
UI and backend communicate via **Packet Protocol**:
|
||||
|
||||
```
|
||||
UI (Textual TUI)
|
||||
↓ BackendClient
|
||||
Packet → queue.Queue → BackendServer (independent thread)
|
||||
↓
|
||||
Process request → Return response
|
||||
```
|
||||
|
||||
### Config Management
|
||||
|
||||
- **Storage location**: `~/.trulymem/config.json`
|
||||
- **Auto-load**: Load config from file at startup
|
||||
- **Dynamic update**: Config changes take effect immediately at runtime
|
||||
- **Persistence**: Auto-save to file after modification
|
||||
|
||||
## Common Issues
|
||||
|
||||
### Python Not Found
|
||||
|
||||
Install Python 3.8+: https://www.python.org/downloads/
|
||||
|
||||
### Dependency Installation Failed
|
||||
|
||||
```bash
|
||||
python -m venv venv
|
||||
source venv/bin/activate # Linux/macOS
|
||||
venv\Scripts\activate # Windows
|
||||
pip install -r requirements.txt
|
||||
```
|
||||
|
||||
### Invalid API Key
|
||||
|
||||
Check API Key format, ensure no extra spaces.
|
||||
|
||||
## Development Commands
|
||||
|
||||
```bash
|
||||
# Install dependencies
|
||||
pip install -r requirements.txt
|
||||
|
||||
# Run tests
|
||||
pytest tests/
|
||||
|
||||
# Build
|
||||
bash build/build_windows.bat # Windows
|
||||
bash build/build_linux.sh # Linux
|
||||
```
|
||||
@ -1,182 +0,0 @@
|
||||
# TrulyMEM Working Memory Chain Mechanism
|
||||
|
||||
## Overview
|
||||
|
||||
TrulyMEM maintains conversation continuity through the working memory chain mechanism. Since there's no traditional message history array, the graph database is the only memory carrier, making the working memory chain the key mechanism for maintaining conversation context.
|
||||
|
||||
## Core Problems
|
||||
|
||||
Traditional AI chat systems have these problems when handling continuous tasks:
|
||||
|
||||
1. **No working memory chain**: AI cannot remember the current task status being processed
|
||||
2. **Task context lost**: When a topic is interrupted, AI cannot recover the previous task
|
||||
3. **Lack of task state management**: No clear marking of task completion status
|
||||
|
||||
### Problem Example
|
||||
|
||||
```
|
||||
User: Let's play idiom chain! I'll start with 为所欲为
|
||||
AI: Okay! My turn: 为虎作伥!
|
||||
|
||||
User: Nagato Yuki (topic interrupted)
|
||||
AI: (discusses Nagato Yuki)
|
||||
|
||||
User: About the idiom chain just now, I don't know how to connect to your idiom
|
||||
AI: [Guessing] It seems we haven't played an idiom chain game before...
|
||||
```
|
||||
|
||||
**Problem**: AI completely forgot the previous idiom chain game.
|
||||
|
||||
## Solution
|
||||
|
||||
### Dedicated Tools
|
||||
|
||||
The system provides 4 dedicated task tools:
|
||||
|
||||
| Tool | Function | Use Case |
|
||||
|------|----------|----------|
|
||||
| `task_create` | Create task node | Start new task |
|
||||
| `task_set_state` | Set task state | Update in_progress/completed/paused/cancelled |
|
||||
| `task_delete` | Delete task | Clean up completed task |
|
||||
| `task_link_info` | Link info node | Connect task with specific information |
|
||||
|
||||
### Task States
|
||||
|
||||
- **in_progress**: Task is executing
|
||||
- **completed**: Task completed successfully
|
||||
- **paused**: Task interrupted, can be resumed
|
||||
- **cancelled**: Task cancelled
|
||||
|
||||
## Usage Flow
|
||||
|
||||
### Must Execute Per Turn
|
||||
|
||||
1. **Query persona graph** (highest priority)
|
||||
```
|
||||
Call memory_recall
|
||||
Parameters: {"query_intent": "AI,persona,role,character,tone", "depth": 2}
|
||||
```
|
||||
|
||||
2. **Query working memory chain**
|
||||
```
|
||||
Call memory_recall
|
||||
Parameters: {"query_intent": "TaskNode,working_memory,task_chain", "depth": 2}
|
||||
```
|
||||
|
||||
3. **Generate reply based on context**
|
||||
|
||||
4. **Update working memory chain** (if necessary)
|
||||
|
||||
## Complete Example: Idiom Chain Game
|
||||
|
||||
### Turn 1: User starts game
|
||||
|
||||
```
|
||||
User: Let's play idiom chain! I'll start with 为所欲为
|
||||
|
||||
AI Actions:
|
||||
1. Query persona graph → Get current persona (e.g., catgirl)
|
||||
2. Query working memory chain → No active task
|
||||
3. Record game state:
|
||||
Call memory_commit
|
||||
Parameters: {
|
||||
"triplets": [
|
||||
{"subject": "idiom_chain_current", "relation": "content", "object": "为所欲为"},
|
||||
{"subject": "idiom_chain_current", "relation": "game", "object": "idiom_chain"}
|
||||
]
|
||||
}
|
||||
4. Create task node:
|
||||
Call task_create
|
||||
Parameters: {
|
||||
"task_id": "Task_idiom_chain",
|
||||
"description": "Idiom chain game, current idiom: 为所欲为",
|
||||
"info_nodes": ["idiom_chain_current"]
|
||||
}
|
||||
5. Reply: "Okay~! My turn: 为虎作伥~!"
|
||||
```
|
||||
|
||||
### Turn 2: Topic interrupted
|
||||
|
||||
```
|
||||
User: Nagato Yuki
|
||||
|
||||
AI Actions:
|
||||
1. Query persona graph → Get current persona (catgirl)
|
||||
2. Query working memory chain → Found Task_idiom_chain status "in_progress"
|
||||
3. Pause task:
|
||||
Call task_set_state
|
||||
Parameters: {"task_id": "Task_idiom_chain", "state": "paused"}
|
||||
4. Create new task:
|
||||
Call task_create
|
||||
Parameters: {"task_id": "Task_nagato_yuki", "description": "Discuss Nagato Yuki"}
|
||||
5. Reply about Nagato Yuki
|
||||
```
|
||||
|
||||
### Turn 3: User asks to continue game
|
||||
|
||||
```
|
||||
User: About the idiom chain just now, I don't know how to connect to your idiom
|
||||
|
||||
AI Actions:
|
||||
1. Query persona graph → Get current persona (catgirl)
|
||||
2. Query working memory chain → Found Task_idiom_chain status "paused"
|
||||
3. Resume task:
|
||||
Call task_set_state
|
||||
Parameters: {"task_id": "Task_idiom_chain", "state": "in_progress"}
|
||||
4. Query info node → Get current idiom "为虎作伥"
|
||||
5. Reply: "Okay~! The last idiom was '为虎作伥', your turn: 伥鬼害人~!"
|
||||
```
|
||||
|
||||
## API Reference
|
||||
|
||||
### task_create
|
||||
|
||||
Create task node to track continuous tasks.
|
||||
|
||||
```json
|
||||
{
|
||||
"task_id": "Task_idiom_chain",
|
||||
"description": "Task overview",
|
||||
"info_nodes": ["associated info node names"]
|
||||
}
|
||||
```
|
||||
|
||||
### task_set_state
|
||||
|
||||
Set task state.
|
||||
|
||||
```json
|
||||
{
|
||||
"task_id": "Task_idiom_chain",
|
||||
"state": "in_progress" // in_progress/completed/paused/cancelled
|
||||
}
|
||||
```
|
||||
|
||||
### task_delete
|
||||
|
||||
Delete task node.
|
||||
|
||||
```json
|
||||
{
|
||||
"task_id": "Task_idiom_chain",
|
||||
"delete_info_nodes": true // whether to delete associated info nodes
|
||||
}
|
||||
```
|
||||
|
||||
### task_link_info
|
||||
|
||||
Associate info nodes to task.
|
||||
|
||||
```json
|
||||
{
|
||||
"task_id": "Task_idiom_chain",
|
||||
"info_node_names": ["idiom_chain_current", "idiom_chain_last"]
|
||||
}
|
||||
```
|
||||
|
||||
## Notes
|
||||
|
||||
1. **Persona graph has highest priority**: Must query persona graph first each turn
|
||||
2. **Working memory chain is the only context carrier**: No traditional message history
|
||||
3. **Task state must be updated timely**: Ensure correct state transitions
|
||||
4. **Use dedicated tools**: Prefer task_* tools over memory_commit for task-related operations
|
||||
@ -1,31 +0,0 @@
|
||||
# TrulyMEM 文档
|
||||
|
||||
欢迎来到 TrulyMEM 项目中文文档。
|
||||
|
||||
> [Switch to English version](../en/README.md)
|
||||
|
||||
## 文档目录
|
||||
|
||||
| 文档 | 内容 |
|
||||
|------|------|
|
||||
| [architecture.md](architecture.md) | 系统架构和技术设计 |
|
||||
| [quick_start.md](quick_start.md) | 完整启动指南与配置说明 |
|
||||
| [memory.md](memory.md) | 内部记忆工作机制 |
|
||||
| [persona.md](persona.md) | 人设图机制 |
|
||||
| [working_memory.md](working_memory.md) | 连续性任务处理机制 |
|
||||
| [api.md](api.md) | 后端 API 接口文档（供扩展开发） |
|
||||
| [prompts.md](prompts.md) | 提示词管理模块 |
|
||||
|
||||
## 项目简介
|
||||
|
||||
TrulyMEM (TrueHumanMEM) 是一个让 AI 拥有长期记忆能力的图记忆系统，通过图数据库存储实体关系，让 AI 能够像人类一样记忆、回忆和管理信息。
|
||||
|
||||
## 核心特性
|
||||
|
||||
- **长期记忆存储**: 基于 SQLite 内嵌图数据库，开箱即用
|
||||
- **人设图机制**: 支持角色扮演和性格设定
|
||||
- **工作记忆链**: 维持对话连贯性的任务跟踪机制
|
||||
- **TUI 与后端分离**: 多线程 Queue 通信
|
||||
- **键盘驱动 TUI**: 无需鼠标，全键盘操作
|
||||
- **跨平台支持**: Windows / Linux / macOS
|
||||
- **独立部署**: 支持打包为可执行文件
|
||||
519
docs/zh/api.md
519
docs/zh/api.md
@ -1,519 +0,0 @@
|
||||
# BackendServer API 文档
|
||||
|
||||
本文档描述后端服务器的 API 接口，供开发者扩展其他连接方式（如网络接口、WebSocket 等）。
|
||||
|
||||
## 概述
|
||||
|
||||
TrulyMEM 后端采用 **Packet 通信协议**，通过 `queue.Queue` 实现线程安全通信。后端在独立线程中运行，处理来自客户端的请求。
|
||||
|
||||
### 核心组件
|
||||
|
||||
| 组件 | 说明 |
|
||||
|------|------|
|
||||
| `BackendServer` | 后端服务器，独立线程运行 |
|
||||
| `BackendClient` | 客户端封装，提供便捷方法 |
|
||||
| `PacketType` | 请求类型枚举 |
|
||||
| `Packet` | 数据包（请求） |
|
||||
| `PacketResponse` | 数据包响应 |
|
||||
|
||||
---
|
||||
|
||||
## 请求类型 (PacketType)
|
||||
|
||||
```python
|
||||
class PacketType(Enum):
|
||||
PROCESS_MESSAGE = "process_message" # 处理消息
|
||||
EXECUTE_TOOL = "execute_tool" # 执行工具
|
||||
GET_STATUS = "get_status" # 获取状态
|
||||
GET_SETTINGS = "get_settings" # 获取完整配置（api_config + tool_limits）
|
||||
SET_SETTINGS = "set_settings" # 设置完整配置（api_config + tool_limits）
|
||||
GET_HISTORY = "get_history" # 获取历史
|
||||
SAVE_HISTORY = "save_history" # 保存历史
|
||||
SHUTDOWN = "shutdown" # 关闭服务
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 数据包格式
|
||||
|
||||
### Packet
|
||||
|
||||
```python
|
||||
@dataclass
|
||||
class Packet:
|
||||
id: str # 唯一标识
|
||||
type: PacketType # 请求类型
|
||||
body: Dict[str, Any] # 请求参数
|
||||
response_queue: queue.Queue # 响应队列（可选）
|
||||
created_at: float # 创建时间
|
||||
```
|
||||
|
||||
### PacketResponse
|
||||
|
||||
```python
|
||||
@dataclass
|
||||
class PacketResponse:
|
||||
id: str # 对应的请求ID
|
||||
success: bool # 是否成功
|
||||
data: Any = None # 返回数据
|
||||
error: Optional[str] = None # 错误信息
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## API 接口详情
|
||||
|
||||
### 1. PROCESS_MESSAGE - 处理消息
|
||||
|
||||
发送用户消息，AI 将处理并返回回复（可能包含工具调用）。
|
||||
|
||||
**请求参数：**
|
||||
```python
|
||||
body = {
|
||||
"user_input": str # 用户输入的消息
|
||||
}
|
||||
```
|
||||
|
||||
**响应数据：**
|
||||
```python
|
||||
{
|
||||
"success": True,
|
||||
"content": str, # AI 回复内容
|
||||
"tool_calls": [ # 工具调用记录
|
||||
{
|
||||
"name": str, # 工具名称
|
||||
"arguments": dict,# 工具参数
|
||||
"result": str # 工具执行结果
|
||||
}
|
||||
],
|
||||
"rejected_tools": [ # 被拒绝的工具调用
|
||||
(str, str) # (工具名, 拒绝原因)
|
||||
]
|
||||
}
|
||||
```
|
||||
|
||||
**示例：**
|
||||
```python
|
||||
from core import BackendServer, BackendClient
|
||||
|
||||
server = BackendServer(db_path="graph_memory.db", use_embedded_db=True)
|
||||
server.start(api_key="your-api-key")
|
||||
|
||||
client = BackendClient(server)
|
||||
result = client.process_message("你好，请记住我的名字是小明")
|
||||
|
||||
if result.get("success"):
|
||||
# 响应数据在 data 字段中
|
||||
print(result["data"]["content"])
|
||||
# 工具调用: result["data"]["tool_calls"]
|
||||
# 被拒绝的工具: result["data"]["rejected_tools"]
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
### 2. EXECUTE_TOOL - 执行工具
|
||||
|
||||
直接执行指定的记忆工具。
|
||||
|
||||
> **注意**：前端直接调用的工具**不受次数限制**，只有模型发起的工具调用才受限制。
|
||||
|
||||
**请求参数：**
|
||||
```python
|
||||
body = {
|
||||
"tool_name": str, # 工具名称
|
||||
"arguments": dict # 工具参数
|
||||
}
|
||||
```
|
||||
|
||||
**响应数据：**
|
||||
```python
|
||||
{
|
||||
"success": True,
|
||||
"result": str # 工具执行结果
|
||||
}
|
||||
```
|
||||
|
||||
**示例：**
|
||||
```python
|
||||
result = client.execute_tool("memory_recall", {"query_intent": "用户信息"})
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
### 3. GET_STATUS - 获取状态
|
||||
|
||||
获取后端运行状态。
|
||||
|
||||
**请求参数：**
|
||||
```python
|
||||
body = {} # 无参数
|
||||
```
|
||||
|
||||
**响应数据：**
|
||||
```python
|
||||
{
|
||||
"running": bool, # 后端是否运行中
|
||||
"config": dict, # 当前配置
|
||||
"graph_initialized": bool, # 图数据库是否初始化
|
||||
"client_initialized": bool # API 客户端是否初始化
|
||||
}
|
||||
```
|
||||
|
||||
**示例：**
|
||||
```python
|
||||
status = client.get_status()
|
||||
print(status["data"]["running"]) # True
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
### 4. GET_SETTINGS - 获取完整配置
|
||||
|
||||
获取当前 API 配置和工具限制（一次获取全部）。
|
||||
|
||||
**请求参数：**
|
||||
```python
|
||||
body = {} # 无参数
|
||||
```
|
||||
|
||||
**响应数据：**
|
||||
```python
|
||||
{
|
||||
"api_config": {
|
||||
"api_key": str, # API Key
|
||||
"base_url": str, # API Base URL
|
||||
"model": str # 模型名称
|
||||
},
|
||||
"tool_limits": {
|
||||
"persona_query_max": int, # 人设图查询上限
|
||||
"persona_update_max": int, # 人设图修改上限
|
||||
"task_query_max": int, # 工作记忆查询上限
|
||||
"task_update_max": int, # 工作记忆修改上限
|
||||
"memory_query_max": int, # 一般记忆查询上限
|
||||
"memory_update_max": int # 一般记忆修改上限
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
**示例：**
|
||||
```python
|
||||
result = client.get_settings()
|
||||
api_config = result["data"]["api_config"]
|
||||
tool_limits = result["data"]["tool_limits"]
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
### 5. SET_SETTINGS - 设置完整配置
|
||||
|
||||
更新 API 配置和工具限制（一次设置全部）。
|
||||
|
||||
**请求参数：**
|
||||
```python
|
||||
body = {
|
||||
"api_config": {
|
||||
"api_key": str, # API Key
|
||||
"base_url": str, # API Base URL (默认: https://api.deepseek.com)
|
||||
"model": str # 模型名称 (默认: deepseek-chat)
|
||||
},
|
||||
"tool_limits": {
|
||||
"persona_query_max": int, # 人设图查询上限 (≥1)
|
||||
"persona_update_max": int, # 人设图修改上限 (≥1)
|
||||
"task_query_max": int, # 工作记忆查询上限 (≥1)
|
||||
"task_update_max": int, # 工作记忆修改上限 (≥1)
|
||||
"memory_query_max": int, # 一般记忆查询上限 (≥1)
|
||||
"memory_update_max": int # 一般记忆修改上限 (≥1)
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
**响应数据：**
|
||||
```python
|
||||
{
|
||||
"status": "settings_updated"
|
||||
}
|
||||
```
|
||||
|
||||
**示例：**
|
||||
```python
|
||||
result = client.update_settings(
|
||||
api_config={
|
||||
"api_key": "sk-xxxxx",
|
||||
"base_url": "https://api.deepseek.com",
|
||||
"model": "deepseek-chat"
|
||||
},
|
||||
tool_limits={
|
||||
"persona_query_max": 2,
|
||||
"task_query_max": 5,
|
||||
"memory_query_max": 30
|
||||
}
|
||||
)
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
### 6. GET_HISTORY - 获取消息历史
|
||||
|
||||
获取保存的消息历史（从数据库读取，用于UI显示，不参与模型推理）。
|
||||
|
||||
**请求参数：**
|
||||
```python
|
||||
body = {} # 无参数
|
||||
```
|
||||
|
||||
**响应数据：**
|
||||
```python
|
||||
{
|
||||
"history": list # 消息历史列表 [{"role": "user/assistant", "content": "..."}]
|
||||
}
|
||||
```
|
||||
|
||||
**说明：**
|
||||
- 消息历史存储在数据库 `chat_records` 表中
|
||||
- 最多返回最近 500 条记录
|
||||
- 历史消息仅用于 UI 显示，不参与模型推理
|
||||
|
||||
---
|
||||
|
||||
### 7. SAVE_HISTORY - 保存消息历史
|
||||
|
||||
保存消息历史到数据库（每次处理消息后自动保存，用户消息和AI回复分别保存）。
|
||||
|
||||
**请求参数：**
|
||||
```python
|
||||
body = {
|
||||
"messages": list # 消息列表 [{"role": "...", "content": "..."}]
|
||||
}
|
||||
```
|
||||
|
||||
**响应数据：**
|
||||
```python
|
||||
{
|
||||
"status": "history_saved"
|
||||
}
|
||||
```
|
||||
|
||||
**说明：**
|
||||
- 消息自动保存到数据库 `chat_records` 表
|
||||
- 系统自动限制最多保留 500 条记录，超出后自动删除旧记录
|
||||
- 每次调用 `PROCESS_MESSAGE` 时，会自动保存用户消息和AI回复
|
||||
- **清空历史**：通过 `SAVE_HISTORY` 传递空消息列表 `messages=[]` 可清空历史，`client.clear_history()` 方法即基于此实现
|
||||
|
||||
---
|
||||
|
||||
### 8. SHUTDOWN - 关闭服务
|
||||
|
||||
关闭后端服务器。
|
||||
|
||||
**请求参数：**
|
||||
```python
|
||||
body = {} # 无参数
|
||||
```
|
||||
|
||||
**响应数据：**
|
||||
```python
|
||||
{
|
||||
"status": "shutdown"
|
||||
}
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 使用示例
|
||||
|
||||
### 基础使用
|
||||
|
||||
```python
|
||||
from core import BackendServer, BackendClient
|
||||
|
||||
# 1. 创建并启动后端
|
||||
# config_file 默认: ~/.trulymem/config.json
|
||||
server = BackendServer(
|
||||
db_path="graph_memory.db",
|
||||
use_embedded_db=True,
|
||||
config_file=None # 可选，自定义配置路径
|
||||
)
|
||||
server.start(
|
||||
api_key="your-api-key",
|
||||
base_url="https://api.deepseek.com",
|
||||
model="deepseek-chat" # 可选，模型名称
|
||||
)
|
||||
|
||||
# 2. 创建客户端
|
||||
client = BackendClient(server)
|
||||
|
||||
# 3. 发送消息
|
||||
result = client.process_message("你好")
|
||||
if result.get("success"):
|
||||
print(result["content"])
|
||||
|
||||
# 4. 关闭
|
||||
client.shutdown()
|
||||
```
|
||||
|
||||
### 使用 Packet 协议
|
||||
|
||||
```python
|
||||
import queue
|
||||
from core import BackendServer, Packet, PacketType
|
||||
|
||||
server = BackendServer(config_file=None)
|
||||
server.start(api_key="your-key", model="deepseek-chat")
|
||||
|
||||
# 创建请求包
|
||||
response_queue = queue.Queue()
|
||||
packet = Packet(
|
||||
id="req-001",
|
||||
type=PacketType.PROCESS_MESSAGE,
|
||||
body={"user_input": "你好"},
|
||||
response_queue=response_queue
|
||||
)
|
||||
|
||||
# 发送请求
|
||||
result = server.send(packet)
|
||||
print(result.body)
|
||||
|
||||
# 关闭
|
||||
server.shutdown()
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 扩展指南
|
||||
|
||||
### 扩展为 HTTP API
|
||||
|
||||
```python
|
||||
from flask import Flask, request, jsonify
|
||||
from core import BackendServer, BackendClient
|
||||
|
||||
app = Flask(__name__)
|
||||
server = BackendServer()
|
||||
client = BackendClient(server)
|
||||
|
||||
@app.route("/message", methods=["POST"])
|
||||
def send_message():
|
||||
data = request.json
|
||||
result = client.process_message(data["message"])
|
||||
return jsonify(result)
|
||||
|
||||
@app.route("/config", methods=["POST"])
|
||||
def update_config():
|
||||
data = request.json
|
||||
result = client.update_settings(
|
||||
api_config=data.get("api_config", {}),
|
||||
tool_limits=data.get("tool_limits", {})
|
||||
)
|
||||
return jsonify(result)
|
||||
|
||||
@app.route("/status", methods=["GET"])
|
||||
def get_status():
|
||||
result = client.get_status()
|
||||
return jsonify(result)
|
||||
|
||||
if __name__ == "__main__":
|
||||
server.start()
|
||||
app.run(port=8080)
|
||||
```
|
||||
|
||||
### 扩展为 WebSocket
|
||||
|
||||
```python
|
||||
import asyncio
|
||||
import websockets
|
||||
import json
|
||||
from core import BackendServer, BackendClient
|
||||
|
||||
server = BackendServer()
|
||||
client = BackendClient(server)
|
||||
|
||||
async def handler(websocket):
|
||||
async for message in websocket:
|
||||
data = json.loads(message)
|
||||
msg_type = data.get("type")
|
||||
|
||||
if msg_type == "message":
|
||||
result = client.process_message(data["content"])
|
||||
elif msg_type == "settings":
|
||||
result = client.update_settings(
|
||||
api_config=data.get("api_config", {}),
|
||||
tool_limits=data.get("tool_limits", {})
|
||||
)
|
||||
elif msg_type == "status":
|
||||
result = client.get_status()
|
||||
else:
|
||||
result = {"success": False, "error": "unknown type"}
|
||||
|
||||
await websocket.send(json.dumps(result))
|
||||
|
||||
async def main():
|
||||
server.start()
|
||||
async with websockets.serve(handler, "localhost", 8765):
|
||||
await asyncio.Future()
|
||||
|
||||
asyncio.run(main())
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 线程安全说明
|
||||
|
||||
- `BackendServer` 使用 `threading.Lock` 保护共享资源
|
||||
- 所有请求通过 `queue.Queue` 传递，线程安全
|
||||
- 响应通过每个请求独立的响应队列返回
|
||||
- 默认超时时间：30 秒
|
||||
|
||||
---
|
||||
|
||||
## 工具调用限制
|
||||
|
||||
### 限制范围
|
||||
|
||||
| 调用方式 | 是否受限 | 说明 |
|
||||
|---------|---------|------|
|
||||
| 模型发起的工具调用 | ✅ 受限 | 通过 `PROCESS_MESSAGE` 触发，模型自动调用工具 |
|
||||
| 前端直接调用工具 | ❌ 不受限 | 通过 `EXECUTE_TOOL` 直接调用 |
|
||||
|
||||
### 限制规则（仅限模型发起）
|
||||
|
||||
| 类别 | 操作 | 每轮上限 |
|
||||
|------|------|---------|
|
||||
| 人设图 | 查询 | 1 次 |
|
||||
| 人设图 | 修改 | 1 次 |
|
||||
| 工作记忆链 | 查询 | 4 次 |
|
||||
| 工作记忆链 | 修改 | 2 次 |
|
||||
| 一般记忆 | 查询 | 20 次 |
|
||||
| 一般记忆 | 修改 | 10 次 |
|
||||
|
||||
### 重置机制
|
||||
|
||||
- 每次调用 `PROCESS_MESSAGE` 时，计数器自动重置
|
||||
- 前端直接调用 `EXECUTE_TOOL` 不会重置计数器
|
||||
|
||||
---
|
||||
|
||||
## 错误处理
|
||||
|
||||
所有 API 返回统一格式：
|
||||
|
||||
```python
|
||||
# 成功
|
||||
{
|
||||
"success": True,
|
||||
"data": {...}
|
||||
}
|
||||
|
||||
# 失败
|
||||
{
|
||||
"success": False,
|
||||
"error": "错误描述"
|
||||
}
|
||||
```
|
||||
|
||||
常见错误：
|
||||
|
||||
| 错误信息 | 说明 |
|
||||
|---------|------|
|
||||
| `API Key 未配置` | 未设置 API Key |
|
||||
| `timeout` | 请求超时 |
|
||||
| `工具调用被拒绝: ...` | 工具调用频率超限 |
|
||||
@ -1,180 +0,0 @@
|
||||
# TrulyMEM 架构设计
|
||||
|
||||
## 核心原则
|
||||
|
||||
- 键盘驱动，零鼠标依赖
|
||||
- 极简视觉，信息密度优先
|
||||
- 工具痕迹默认隐藏，需要时可展开
|
||||
- TUI 与后端分离，多线程通信
|
||||
- **一切皆图**，AI 推理全部在后端
|
||||
|
||||
## 项目结构
|
||||
|
||||
```
|
||||
TrulyMEM-TrueHumanMEM/
|
||||
├── trulymem_entry.py # 入口：先启动 core → 再启动 ui
|
||||
├── core/ # 后端/业务逻辑
|
||||
│ ├── __init__.py # 导出 BackendServer, BackendClient, EmbeddedGraphDB
|
||||
│ ├── server.py # BackendServer (Packet 通信协议)
|
||||
│ ├── client.py # BackendClient (Packet 协议客户端)
|
||||
│ ├── embedded_db.py # SQLite 图数据库实现
|
||||
│ ├── graph_client.py # OpenAI/DeepSeek API 客户端
|
||||
│ ├── tool_executor.py # 工具执行器
|
||||
│ ├── tool_limiter.py # 工具调用限制器
|
||||
│ ├── tools/ # 工具定义
|
||||
│ │ └── memory_tools.py
|
||||
│ └── prompts/ # 提示词管理
|
||||
├── ui/ # TUI 显示层（仅显示，无 AI 逻辑）
|
||||
│ ├── __init__.py # 导出 GraphMemoryApp
|
||||
│ ├── app.py # GraphMemoryApp (通过 BackendClient 通信)
|
||||
│ ├── widgets/ # TUI 组件
|
||||
│ ├── models/ # 数据模型
|
||||
│ ├── services/ # 服务层（仅配置管理）
|
||||
│ ├── handlers/ # 事件处理
|
||||
│ └── styles/ # 样式文件
|
||||
└── tests/ # 测试 (42 tests)
|
||||
```
|
||||
|
||||
## 架构图
|
||||
|
||||
```
|
||||
trulymem_entry.py
|
||||
│
|
||||
├─ BackendServer.start() → 独立线程运行
|
||||
│ ├─ 处理 PROCESS_MESSAGE 请求 → AI 推理 + 工具调用
|
||||
│ ├─ 处理 EXECUTE_TOOL 请求 → 外部工具调用（不限次数）
|
||||
│ ├─ 处理 GET/SET_CONFIG 请求
|
||||
│ └─ 管理 GraphMemoryClient, EmbeddedGraphDB
|
||||
│
|
||||
└─ GraphMemoryApp(backend_server=server)
|
||||
│
|
||||
└─ BackendClient ← Packet 通信 → BackendServer
|
||||
```
|
||||
|
||||
## 组件职责
|
||||
|
||||
### core/ (后端)
|
||||
|
||||
| 组件 | 职责 |
|
||||
|------|------|
|
||||
| `server.py` | Packet 协议处理，多线程队列通信，AI 推理，工具限制 |
|
||||
| `client.py` | 客户端封装，UI 与后端通信桥梁 |
|
||||
| `embedded_db.py` | SQLite 图数据库 CRUD |
|
||||
| `graph_client.py` | OpenAI/DeepSeek API 客户端 |
|
||||
| `tool_executor.py` | 工具执行逻辑 |
|
||||
| `tool_limiter.py` | 工具调用频率限制（仅限 AI 推理） |
|
||||
|
||||
### ui/ (显示层)
|
||||
|
||||
| 组件 | 职责 |
|
||||
|------|------|
|
||||
| `app.py` | Textual 应用主类，仅通过 BackendClient 通信 |
|
||||
| `services/` | 仅配置管理，无 AI 逻辑 |
|
||||
|
||||
### 通信协议
|
||||
|
||||
UI 与后端通过 **Packet 通信协议** 交互：
|
||||
|
||||
```python
|
||||
from core import BackendServer, BackendClient, Packet, PacketType
|
||||
|
||||
# 后端启动
|
||||
server = BackendServer(db_path="graph_memory.db", use_embedded_db=True)
|
||||
server.start(api_key="your-key")
|
||||
|
||||
# 客户端通信
|
||||
client = BackendClient(server)
|
||||
result = client.process_message("你好") # AI 推理
|
||||
result = client.execute_tool("memory_introspect", {}) # 外部工具调用
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 数据流
|
||||
|
||||
```
|
||||
用户输入 → InputBox → on_input_box_send_message
|
||||
↓
|
||||
BackendClient.process_message(user_input)
|
||||
↓
|
||||
Packet (type=PROCESS_MESSAGE) → queue.Queue
|
||||
↓
|
||||
BackendServer (独立线程)
|
||||
↓
|
||||
GraphMemoryClient.send_message_with_history()
|
||||
↓
|
||||
OpenAI API / DeepSeek API
|
||||
↓
|
||||
execute_tool() + ToolLimiter (AI 推理时受限)
|
||||
↓
|
||||
EmbeddedGraphDB (图数据库)
|
||||
↓
|
||||
循环调用 API 直到无 tool_calls
|
||||
↓
|
||||
Packet 响应返回
|
||||
↓
|
||||
MessageHistory 显示
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 启动流程
|
||||
|
||||
```python
|
||||
# trulymem_entry.py
|
||||
def main():
|
||||
# 配置文件路径 (~/.trulymem/config.json 或项目目录)
|
||||
CONFIG_PATH = Path.home() / ".trulymem" / "config.json"
|
||||
DB_PATH = Path.home() / ".trulymem" / "graph_memory.db"
|
||||
|
||||
# 创建后端（配置由后端管理）
|
||||
backend_server = BackendServer(
|
||||
db_path=str(DB_PATH),
|
||||
use_embedded_db=True,
|
||||
config_file=str(CONFIG_PATH)
|
||||
)
|
||||
backend_server.start() # 自动加载配置
|
||||
|
||||
# 创建UI（通过 BackendClient 通信）
|
||||
app = GraphMemoryApp(backend_server=backend_server, config_file=str(CONFIG_PATH))
|
||||
app.run()
|
||||
|
||||
backend_server.shutdown()
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 工具系统
|
||||
|
||||
### 记忆工具 (6个)
|
||||
- `memory_recall` - 检索记忆
|
||||
- `memory_commit` - 写入记忆
|
||||
- `memory_purge` - 删除记忆
|
||||
- `memory_introspect` - 查看状态
|
||||
- `memory_archive` - 归档记忆
|
||||
- `memory_cleanup` - 清理数据
|
||||
|
||||
### 人设工具 (2个)
|
||||
- `persona_update` - 更新人设
|
||||
- `persona_clear` - 清除人设
|
||||
|
||||
### 任务工具 (4个)
|
||||
- `task_create` - 创建任务
|
||||
- `task_set_state` - 设置状态
|
||||
- `task_delete` - 删除任务
|
||||
- `task_link_info` - 关联信息
|
||||
|
||||
---
|
||||
|
||||
## 错误处理原则
|
||||
|
||||
所有 API **不抛出异常**，错误通过返回字典传递：
|
||||
|
||||
```python
|
||||
result = client.process_message("hello")
|
||||
|
||||
if result.get("success"):
|
||||
print(result["content"])
|
||||
else:
|
||||
print(result["error"]) # 错误描述
|
||||
```
|
||||
@ -1,237 +0,0 @@
|
||||
# TrulyMEM 记忆机制
|
||||
|
||||
本文档详细说明 TrulyMEM 内部的记忆工作机制。
|
||||
|
||||
## 核心设计理念
|
||||
|
||||
### 区别于传统上下文系统
|
||||
|
||||
传统 AI 对话系统使用 messages 数组存储对话历史：
|
||||
- 每次请求携带全部历史消息
|
||||
- 随着对话轮次增加，上下文逐渐膨胀
|
||||
- 最终触发记忆压缩或滑动窗口，造成记忆丢失
|
||||
|
||||
TrulyMEM 的解决思路：
|
||||
- **摒弃** messages 数组上下文
|
||||
- **唯一** 记忆载体：图数据库
|
||||
- 全部记忆以三元组（节点）- 关系 → （节点）形式存储
|
||||
|
||||
### 图数据库作为唯一记忆源
|
||||
|
||||
所有记忆必须通过以下方式写入图数据库：
|
||||
- `memory_commit` - 写入新记忆
|
||||
- `memory_purge` - 删除/修正记忆
|
||||
|
||||
所有记忆必须通过以下方式读取：
|
||||
- `memory_recall` - 检索记忆
|
||||
|
||||
---
|
||||
|
||||
## 强制执行流程（每轮对话）
|
||||
|
||||
由于没有传统上下文系统，每轮对话必须按以下顺序执行：
|
||||
|
||||
### 步骤 1：查询人设图（最高优先级）
|
||||
|
||||
```python
|
||||
memory_recall(
|
||||
query_intent="AI,人设,角色,性格,语气,说话风格",
|
||||
depth=2
|
||||
)
|
||||
```
|
||||
|
||||
**目的**：获取当前人设，确保角色一致性。
|
||||
|
||||
**处理逻辑**：
|
||||
- 找到人设 → 严格按照人设的语气、风格、特征回复
|
||||
- 未找到 → 使用默认 TrulyMEM 身份
|
||||
|
||||
### 步骤 2：查询工作记忆链
|
||||
|
||||
```python
|
||||
memory_recall(
|
||||
query_intent="TaskNode,工作记忆,任务链",
|
||||
depth=2
|
||||
)
|
||||
```
|
||||
|
||||
**目的**：获取之前的任务上下文，了解对话历史。
|
||||
|
||||
### 步骤 3：处理对话
|
||||
|
||||
- 理解用户意图
|
||||
- 根据人设和工作记忆链生成回复
|
||||
- 执行其他必要的记忆操作
|
||||
|
||||
### 步骤 4：更新工作记忆链
|
||||
|
||||
```python
|
||||
task_create(
|
||||
task_id="Task_当前轮次ID",
|
||||
description="本轮对话概述",
|
||||
info_nodes=["相关记忆节点"]
|
||||
)
|
||||
```
|
||||
|
||||
**目的**：记录本轮对话，维持时间链。
|
||||
|
||||
---
|
||||
|
||||
## 记忆写入规则
|
||||
|
||||
### 必须写入的情况
|
||||
|
||||
以下信息**必须**写入图数据库：
|
||||
|
||||
| 场景 | 示例 | 写入方式 |
|
||||
|------|------|----------|
|
||||
| 用户明确偏好 | "我喜欢摇滚" | `memory_commit` |
|
||||
| 用户分享信息 | "我在做X项目" | `memory_commit` |
|
||||
| 用户制定计划 | "我打算X" | `memory_commit` |
|
||||
| 用户描述状态 | "我现在在X" | `memory_commit` |
|
||||
|
||||
### 禁止写入的情况
|
||||
|
||||
以下信息**禁止**写入：
|
||||
|
||||
| 场景 | 原因 | 处理方式 |
|
||||
|------|------|----------|
|
||||
| AI 推断的用户偏好 | 未经证实 | 不写入或标注[推测] |
|
||||
| AI 猜测的用户意图 | 未经证实 | 不写入或标注[推测] |
|
||||
| AI 推导的结论 | 未经证实 | 不写入或标注[推测] |
|
||||
|
||||
### 标注规则
|
||||
|
||||
| 类型 | 标注方式 | 示例 |
|
||||
|------|----------|------|
|
||||
| 推理内容 | 必须标注 **[猜测]** | 用户[推测]喜欢音乐 |
|
||||
| 明确内容 | 直接陈述 | 用户喜欢音乐 |
|
||||
|
||||
---
|
||||
|
||||
## 节点与边类型
|
||||
|
||||
### 节点类型
|
||||
|
||||
| 节点类型 | 说明 | 存储内容 |
|
||||
|----------|------|----------|
|
||||
| `PersonaNode` | 人设节点 | AI 角色、性格、语气 |
|
||||
| `TaskNode` | 任务节点 | 任务概述 |
|
||||
| `StateNode` | 状态节点 | 任务状态 |
|
||||
| `InfoNode` | 信息节点 | 具体信息 |
|
||||
| `EntityNode` | 实体节点 | 通用实体 |
|
||||
|
||||
### 边类型
|
||||
|
||||
| 边类型 | 说明 | 连接关系 |
|
||||
|----------|------|----------|
|
||||
| `HAS_PERSONA` | 人设 | AI → PersonaNode |
|
||||
| `NEXT_TASK` | 时间链 | TaskNode → TaskNode |
|
||||
| `HAS_STATE` | 状态 | TaskNode → StateNode |
|
||||
| `CONTAINS_INFO` | 信息 | TaskNode → InfoNode |
|
||||
| `RELATES_TO` | 关联 | EntityNode → EntityNode |
|
||||
|
||||
---
|
||||
|
||||
## 必须查询工作记忆链的场景
|
||||
|
||||
### 强制查询场景
|
||||
|
||||
以下情况**必须**查询工作记忆链：
|
||||
|
||||
| 场景 | 示例 |
|
||||
|------|------|
|
||||
| 每轮对话开始 | 执行步骤 2 |
|
||||
| 用户提到"刚才" | "刚才我们聊了什么？" |
|
||||
| 用户提到"之前" | "继续刚才的话题" |
|
||||
| 用户提到"上次" | "上次说的X" |
|
||||
| 用户询问历史 | "我们之前说了什么？" |
|
||||
| 连续性任务恢复 | 用户回到之前的话题 |
|
||||
| 上下文引用 | "那个东西" |
|
||||
|
||||
---
|
||||
|
||||
## 自主性原则
|
||||
|
||||
在强制执行流程之外，AI 可自主决定：
|
||||
|
||||
### 查询决策
|
||||
|
||||
- 用户询问历史 → 查询
|
||||
- 涉及之前内容 → 查询
|
||||
- 不确定时 → 可查询
|
||||
|
||||
### 写入决策
|
||||
|
||||
- 用户明确提到 → 必须写入
|
||||
- AI 推理得到 → 可以写入，标注[推测]
|
||||
|
||||
### 工具使用决策
|
||||
|
||||
- 根据上下文灵活选择
|
||||
- 避免过度使用
|
||||
- 保持自然对话
|
||||
|
||||
---
|
||||
|
||||
## 完整示例：成语接龙游戏
|
||||
|
||||
### 第一轮：用户发起游戏
|
||||
|
||||
```
|
||||
用户: 咱来玩成语接龙吧，我先开始，为所欲为
|
||||
|
||||
AI操作:
|
||||
1. 查询人设图 → 获取当前人设（如：猫娘）
|
||||
2. 查询工作记忆链 → 无进行中任务
|
||||
3. 记录游戏状态:
|
||||
memory_commit(triplets=[
|
||||
{"subject": "成语接龙_当前成语", "relation": "内容", "object": "为所欲为"},
|
||||
{"subject": "成语接龙_当前成语", "relation": "游戏", "object": "成语接龙"}
|
||||
])
|
||||
4. 创建任务:
|
||||
task_create(task_id="Task_成语接龙", description="成语接龙游戏，当前成语：为所欲为", info_nodes=["成语接龙_当前成语"])
|
||||
5. 回复: "好的喵！我接：为虎作伥喵！"
|
||||
```
|
||||
|
||||
### 第二轮：话题被打断
|
||||
|
||||
```
|
||||
用户: 长门有希
|
||||
|
||||
AI操作:
|
||||
1. 查询人设图 → 获取当前人设（猫娘）
|
||||
2. 查询工作记忆链 → 发现 Task_成语接龙 状态为"进行中"
|
||||
3. 暂停任务:
|
||||
task_set_state(task_id="Task_成语接龙", state="已暂停")
|
||||
4. 创建新任务:
|
||||
task_create(task_id="Task_长门有希", description="讨论长门有希")
|
||||
5. 回复关于长门有希的内容
|
||||
```
|
||||
|
||||
### 第三轮：用户要求继续游戏
|
||||
|
||||
```
|
||||
用户: 关于刚才的成语接龙，我并不知道应该怎么接你的成语，请帮我接一下
|
||||
|
||||
AI操作:
|
||||
1. 查询人设图 → 获取当前人设（猫娘）
|
||||
2. 查询工作记忆链 → 发现 Task_成语接龙 状态为"已暂停"
|
||||
3. 恢复任务:
|
||||
task_set_state(task_id="Task_成语接龙", state="进行中")
|
||||
4. 查询信息节点 → 获取当前成语"为虎作伥"
|
||||
5. 回复: "好的喵！上一个成语是'为虎作伥'，我帮你接：伥鬼害人喵！"
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 执行检查清单
|
||||
|
||||
每轮对话必须检查：
|
||||
|
||||
- [ ] 步骤 1：是否查询了人设图？
|
||||
- [ ] 步骤 2：是否查询了工作记忆链？
|
||||
- [ ] 步骤 3：是否根据人设和工作记忆链生成回复？
|
||||
- [ ] 步骤 4：是否更新了工作记忆链？
|
||||
- [ ] 涉及上下文引用时是否查询了工作记忆链？
|
||||
- [ ] 用户提到"刚才/之前/上次"时是否查询了工作记忆链？
|
||||
@ -1,214 +0,0 @@
|
||||
# TrulyMEM 人设图机制
|
||||
|
||||
本文档详细说明 TrulyMEM 的人设图（Persona Graph）工作机制。
|
||||
|
||||
## 概述
|
||||
|
||||
人设图是 TrulyMEM 的核心机制之一，用于维护 AI 的角色、性格、语气等属性。与传统 AI 不同，TrulyMEM 的人设是可持久化、可动态切换的，存储在图数据库中。
|
||||
|
||||
## 核心概念
|
||||
|
||||
### 人设节点（PersonaNode）
|
||||
|
||||
存储 AI 的角色属性：
|
||||
|
||||
| 属性 | 说明 | 示例 |
|
||||
|------|------|------|
|
||||
| 扮演角色 | AI 当前扮演的角色 | 猫娘、教师、助手 |
|
||||
| 说话风格 | 语气特点 |可爱、严肃、专业 |
|
||||
| 性格特点 | 性格描述 | 活泼、严谨、耐心 |
|
||||
| 口头禅 | 习惯用语 | 喵呜~、明白了 |
|
||||
| 背景故事 | 角色背景设定 | 来自星海的猫娘 |
|
||||
|
||||
### 人设边（Edge）
|
||||
|
||||
| 边类型 | 说明 | 连接关系 |
|
||||
|------|------|----------|
|
||||
| `HAS_PERSONA` | 人设 | AI → PersonaNode |
|
||||
|
||||
---
|
||||
|
||||
## 强制查询机制
|
||||
|
||||
### 每轮对话必须执行
|
||||
|
||||
根据 `system_prompt.md`，每轮对话**必须**首先查询人设图：
|
||||
|
||||
```python
|
||||
memory_recall(
|
||||
query_intent="AI,人设,角色,性格,语气,说话风格",
|
||||
depth=2
|
||||
)
|
||||
```
|
||||
|
||||
**处理逻辑：**
|
||||
- 找到人设 → 严格按照人设的语气、风格、特征回复
|
||||
- 未找到 → 使用默认 TrulyMEM 身份
|
||||
|
||||
### 人设优先级
|
||||
|
||||
- **人设优先级 > 默认身份**
|
||||
- 每句话都符合人设的语气、风格、特征
|
||||
- 绝不主动跳出角色，除非用户明确要求
|
||||
|
||||
---
|
||||
|
||||
## 工具
|
||||
|
||||
### persona_update
|
||||
|
||||
更新人设。修改 AI 的角色、性格、语气等属性。
|
||||
|
||||
**参数：**
|
||||
|
||||
| 参数 | 类型 | 说明 | 必填 |
|
||||
|------|------|------|------|
|
||||
| `attributes` | array | 人设属性列表 | ✅ |
|
||||
| `mode` | string | replace=替换, merge=合并 | ❌ |
|
||||
|
||||
**attributes 子参数：**
|
||||
|
||||
| 子参数 | 说明 |
|
||||
|------|------|
|
||||
| `attribute` | 属性名（扮演角色、说话风格、性格特点、口头禅、背景故事） |
|
||||
| `value` | 属性值 |
|
||||
|
||||
**示例 - 切换为猫娘角色：**
|
||||
|
||||
```python
|
||||
persona_update(
|
||||
attributes=[
|
||||
{"attribute": "扮演角色", "value": "猫娘"},
|
||||
{"attribute": "说话风格", "value": "可爱、卖萌、使用'喵'作为语气词"},
|
||||
{"attribute": "性格特点", "value": "活泼、粘人、忠诚"}
|
||||
],
|
||||
mode="replace"
|
||||
)
|
||||
```
|
||||
|
||||
**示例 - 添加新属性（保留现有属性）：**
|
||||
|
||||
```python
|
||||
persona_update(
|
||||
attributes=[
|
||||
{"attribute": "口头禅", "value": "喵呜~"}
|
||||
],
|
||||
mode="merge"
|
||||
)
|
||||
```
|
||||
|
||||
**示例 - 设置专业角色：**
|
||||
|
||||
```python
|
||||
persona_update(
|
||||
attributes=[
|
||||
{"attribute": "扮演角色", "value": "Python专家"},
|
||||
{"attribute": "说话风格", "value": "专业、简洁、代码示例丰富"},
|
||||
{"attribute": "性格特点", "value": "严谨、耐心、乐于助人"}
|
||||
],
|
||||
mode="replace"
|
||||
)
|
||||
```
|
||||
|
||||
### persona_clear
|
||||
|
||||
清除人设。删除 AI 的角色设定，恢复默认身份。
|
||||
|
||||
**参数：**
|
||||
|
||||
| 参数 | 类型 | 默认值 | 说明 |
|
||||
|------|------|--------|------|
|
||||
| `confirm` | boolean | true | 确认清除 |
|
||||
|
||||
---
|
||||
|
||||
## 更新流程
|
||||
|
||||
### 用户要求角色扮演时
|
||||
|
||||
1. 使用 `persona_update` 更新人设
|
||||
2. 立即按照新人设回复
|
||||
|
||||
### 用户要求恢复默认身份时
|
||||
|
||||
1. 使用 `persona_clear` 清除人设
|
||||
2. 恢复为 TrulyMEM 默认身份
|
||||
|
||||
---
|
||||
|
||||
## 对话示例
|
||||
|
||||
### 示例 1：切换角色
|
||||
|
||||
```
|
||||
用户: 你好，我想让你扮演一只猫娘
|
||||
|
||||
AI:
|
||||
1. 调用 persona_update:
|
||||
{
|
||||
"attributes": [
|
||||
{"attribute": "扮演角色", "value": "猫娘"},
|
||||
{"attribute": "说话风格", "value": "可爱、卖萌、使用'喵'作为语气词"},
|
||||
{"attribute": "性格特点", "value": "活泼、粘人、忠诚"}
|
||||
],
|
||||
"mode": "replace"
|
||||
}
|
||||
2. 调用 memory_commit 存储人设到图数据库
|
||||
3. 回复: "好的喵！主人你好呀~我是主人的猫娘，有什么需要帮忙的吗喵？"
|
||||
```
|
||||
|
||||
### 示例 2：保持角色一致性
|
||||
|
||||
```
|
||||
用户: 今天的天气怎么样？
|
||||
|
||||
AI: 查询人设图 → 获取当前人设（猫娘）
|
||||
回复: "喵~主人，今天天气很好呢喵！阳光明媚，很适合出门散步哦~"
|
||||
```
|
||||
|
||||
### 示例 3：恢复默认身份
|
||||
|
||||
```
|
||||
用户: 好了，恢复正常吧
|
||||
|
||||
AI:
|
||||
1. 调用 persona_clear(confirm=true)
|
||||
2. 调用 memory_purge 删除人设节点
|
||||
3. 回复: "好的，已恢复正常。我是 TrulyMEM，一个拥有长期记忆能力的 AI 助手。"
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 存储结构
|
||||
|
||||
### 图数据库中的存储
|
||||
|
||||
```python
|
||||
# 人设节点
|
||||
{
|
||||
"node_type": "PersonaNode",
|
||||
"name": "AI人设",
|
||||
"attributes": {
|
||||
"扮演角色": "猫娘",
|
||||
"说话风格": "可爱、卖萌、使用'喵'作为语气词",
|
||||
"性格特点": "活泼、粘人、忠诚"
|
||||
}
|
||||
}
|
||||
|
||||
# 边
|
||||
{
|
||||
"edge_type": "HAS_PERSONA",
|
||||
"from": "AI",
|
||||
"to": "AI人设"
|
||||
}
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 实现要点
|
||||
|
||||
1. **每轮强制查询**：人设图查询是每轮对话的第一步
|
||||
2. **持久化存储**：人设存储在图数据库中，不丢失
|
||||
3. **动态切换**：支持实时切换角色
|
||||
4. **状态保持**：切换后立即按新人设回复
|
||||
5. **明确边界**：除非用户要求，绝不主动跳出角色
|
||||
@ -1,126 +0,0 @@
|
||||
# 提示词管理文档
|
||||
|
||||
本文档描述提示词管理模块。
|
||||
|
||||
## 概述
|
||||
|
||||
提示词管理模块（`core/prompts/`）负责加载和管理告诉 AI 如何使用记忆工具的系统提示词。
|
||||
|
||||
## 核心组件
|
||||
|
||||
| 组件 | 说明 |
|
||||
|------|------|
|
||||
| `PromptManager` | 提示词管理器，单例模式 |
|
||||
| `system_prompt.md` | 主要系统提示词模板 |
|
||||
|
||||
## 使用方法
|
||||
|
||||
```python
|
||||
from core.prompts import PromptManager
|
||||
|
||||
# 获取单例实例
|
||||
prompt_manager = PromptManager()
|
||||
|
||||
# 获取系统提示词
|
||||
system_prompt = prompt_manager.get_system_prompt()
|
||||
```
|
||||
|
||||
## 系统提示词内容
|
||||
|
||||
系统提示词包含：
|
||||
|
||||
### 1. 核心身份
|
||||
|
||||
- **名称**: TrulyMEM (TrueHumanMEM)
|
||||
- **能力**: 基于图数据库的长期记忆
|
||||
- **理念**: 让 AI 的记忆方式更像人类
|
||||
|
||||
### 2. 核心能力
|
||||
|
||||
1. **长期记忆** - 图数据库存储实体关系
|
||||
2. **人设管理** - 角色扮演和性格设定
|
||||
3. **任务跟踪** - 工作记忆链
|
||||
|
||||
### 3. 记忆原则
|
||||
|
||||
- **必须写入**: 用户明确表达的偏好、分享的信息、计划
|
||||
- **禁止写入**: AI 推断的内容（除非标注[推测]）
|
||||
- **标注**: 推断内容必须标注 **[推测]**
|
||||
|
||||
### 4. 强制执行流程（每轮）
|
||||
|
||||
```
|
||||
步骤 1: 查询人设图（最高优先级）
|
||||
步骤 2: 查询工作记忆链
|
||||
步骤 3: 处理对话
|
||||
步骤 4: 更新工作记忆链
|
||||
```
|
||||
|
||||
### 5. 工具系统
|
||||
|
||||
#### 记忆工具
|
||||
|
||||
| 工具 | 功能 |
|
||||
|------|------|
|
||||
| `memory_recall` | 检索记忆 |
|
||||
| `memory_commit` | 写入记忆 |
|
||||
| `memory_purge` | 删除记忆 |
|
||||
| `memory_introspect` | 查看状态 |
|
||||
|
||||
#### 人设工具
|
||||
|
||||
| 工具 | 功能 |
|
||||
|------|------|
|
||||
| `persona_update` | 更新人设 |
|
||||
| `persona_clear` | 清除人设 |
|
||||
|
||||
#### 任务工具
|
||||
|
||||
| 工具 | 功能 |
|
||||
|------|------|
|
||||
| `task_create` | 创建任务 |
|
||||
| `task_set_state` | 设置状态 |
|
||||
| `task_delete` | 删除任务 |
|
||||
| `task_link_info` | 关联信息 |
|
||||
|
||||
### 6. 自主性原则
|
||||
|
||||
AI 可自主决定：
|
||||
- 是否查询其他记忆
|
||||
- 是否写入其他记忆
|
||||
- 如何使用工具（强制要求外）
|
||||
|
||||
### 7. 对话风格
|
||||
|
||||
- 自然流畅
|
||||
- 避免机械式工具调用
|
||||
- 优先理解用户意图
|
||||
- 适时使用记忆增强体验
|
||||
|
||||
## 文件结构
|
||||
|
||||
```
|
||||
core/prompts/
|
||||
├── __init__.py # 导出 PromptManager
|
||||
├── prompt_manager.py # PromptManager 类
|
||||
└── templates/
|
||||
└── system_prompt.md # 主要系统提示词
|
||||
```
|
||||
|
||||
## 自定义
|
||||
|
||||
### 自定义系统提示词
|
||||
|
||||
修改 `core/prompts/templates/system_prompt.md` 自定义 AI 行为。
|
||||
|
||||
### 添加自定义提示词
|
||||
|
||||
1. 在 `core/prompts/templates/` 添加提示词模板文件
|
||||
2. 修改 `PromptManager` 支持多个提示词
|
||||
3. 使用 `set_prompt()` 切换提示词
|
||||
|
||||
## 缓存
|
||||
|
||||
- 系统提示词首次加载后缓存在内存中
|
||||
- `get_system_prompt()` 返回缓存内容
|
||||
- 缓存按进程，不持久化
|
||||
@ -1,125 +0,0 @@
|
||||
# TrulyMEM 启动指南
|
||||
|
||||
## 运行方式
|
||||
|
||||
### 从源码运行
|
||||
|
||||
```bash
|
||||
git clone <repo-url>
|
||||
cd TrulyMEM-TrueHumanMEM
|
||||
|
||||
pip install -r requirements.txt
|
||||
|
||||
python trulymem_entry.py
|
||||
```
|
||||
|
||||
### 打包后运行
|
||||
|
||||
打包后会生成可执行文件：
|
||||
|
||||
```bash
|
||||
# Linux/macOS
|
||||
chmod +x TrulyMEM
|
||||
./TrulyMEM
|
||||
|
||||
# Windows
|
||||
TrulyMEM.exe
|
||||
```
|
||||
|
||||
## 系统要求
|
||||
|
||||
- **Python 3.8+**
|
||||
- **API Key**（DeepSeek、OpenAI 或其他兼容 API）
|
||||
|
||||
## 首次配置
|
||||
|
||||
1. 运行应用
|
||||
2. 按 **F2** 展开侧边栏
|
||||
3. 输入 **API Key**、**模型**、**Base URL**
|
||||
4. 按 **Enter** 保存
|
||||
|
||||
配置会自动保存到 `~/.trulymem/config.json`，下次启动自动加载。
|
||||
|
||||
## 快捷键
|
||||
|
||||
| 按键 | 功能 |
|
||||
|------|------|
|
||||
| F1 | 帮助 |
|
||||
| F2 | 切换侧边栏 |
|
||||
| F3 | 工具详情 |
|
||||
| F5 | 清屏 |
|
||||
| F6 | 退出 |
|
||||
|
||||
## 数据存储
|
||||
|
||||
### 源码运行模式
|
||||
|
||||
| 数据 | 位置 |
|
||||
|------|------|
|
||||
| 图数据库 | 项目目录 `graph_memory.db` |
|
||||
| 配置文件 | 项目目录 `config.json`（如存在） |
|
||||
| 数据库格式 | SQLite |
|
||||
|
||||
### 打包运行模式
|
||||
|
||||
| 数据 | 位置 |
|
||||
|------|------|
|
||||
| 图数据库 | `~/.trulymem/graph_memory.db` |
|
||||
| 配置文件 | `~/.trulymem/config.json` |
|
||||
| 数据库格式 | SQLite |
|
||||
|
||||
> **说明**：后端统一管理配置。前端仅负责消息展示，配置修改通过后端持久化到文件系统。
|
||||
|
||||
## 架构说明
|
||||
|
||||
### 通信协议
|
||||
|
||||
UI 与后端通过 **Packet 协议** 通信：
|
||||
|
||||
```
|
||||
UI (Textual TUI)
|
||||
↓ BackendClient
|
||||
Packet → queue.Queue → BackendServer (独立线程)
|
||||
↓
|
||||
处理请求 → 返回响应
|
||||
```
|
||||
|
||||
### 配置管理
|
||||
|
||||
- **存储位置**: `~/.trulymem/config.json`
|
||||
- **自动加载**: 启动时从文件读取配置
|
||||
- **动态更新**: 运行时修改配置立即生效
|
||||
- **持久化**: 修改后自动保存到文件
|
||||
|
||||
## 常见问题
|
||||
|
||||
### Python 未找到
|
||||
|
||||
安装 Python 3.8+：https://www.python.org/downloads/
|
||||
|
||||
### 依赖安装失败
|
||||
|
||||
```bash
|
||||
python -m venv venv
|
||||
source venv/bin/activate # Linux/macOS
|
||||
venv\Scripts\activate # Windows
|
||||
pip install -r requirements.txt
|
||||
```
|
||||
|
||||
### API Key 无效
|
||||
|
||||
检查 API Key 格式，确保无多余空格。
|
||||
|
||||
## 开发命令
|
||||
|
||||
```bash
|
||||
# 安装依赖
|
||||
pip install -r requirements.txt
|
||||
|
||||
# 运行测试
|
||||
pytest tests/
|
||||
|
||||
# 打包
|
||||
bash build/build_windows.bat # Windows
|
||||
bash build/build_linux.sh # Linux
|
||||
```
|
||||
@ -1,182 +0,0 @@
|
||||
# 工作记忆链机制说明
|
||||
|
||||
## 概述
|
||||
|
||||
TrulyMEM 通过工作记忆链机制维持对话连贯性。由于系统没有传统的消息历史数组，图数据库是唯一的记忆载体，工作记忆链是维持对话上下文的关键机制。
|
||||
|
||||
## 核心问题
|
||||
|
||||
传统 AI 对话系统在处理连续性任务时存在以下问题:
|
||||
|
||||
1. **没有工作记忆链**: AI 无法记住当前正在进行的任务状态
|
||||
2. **任务上下文丢失**: 当话题被打断后，AI 无法恢复之前的任务
|
||||
3. **缺乏任务状态管理**: 没有明确标注任务的完成状态
|
||||
|
||||
### 问题示例
|
||||
|
||||
```
|
||||
用户: 咱来玩成语接龙吧，我先开始，为所欲为
|
||||
AI: 好的喵！我接：为虎作伥喵！
|
||||
|
||||
用户: 长门有希 (话题被打断)
|
||||
AI: (讨论长门有希的内容)
|
||||
|
||||
用户: 关于刚才的成语接龙，我并不知道应该怎么接你的成语，请帮我接一下
|
||||
AI: [猜测] 看起来我们之前应该没有进行过成语接龙游戏...
|
||||
```
|
||||
|
||||
**问题**: AI 完全忘记了之前的成语接龙游戏。
|
||||
|
||||
## 解决方案
|
||||
|
||||
### 专用工具
|
||||
|
||||
系统提供 4 个专用任务工具:
|
||||
|
||||
| 工具 | 功能 | 使用场景 |
|
||||
|------|------|----------|
|
||||
| `task_create` | 创建任务节点 | 开始新任务 |
|
||||
| `task_set_state` | 设置任务状态 | 更新进行中/已完成/已暂停/已取消 |
|
||||
| `task_delete` | 删除任务 | 清理完成任务 |
|
||||
| `task_link_info` | 关联信息节点 | 连接任务与具体信息 |
|
||||
|
||||
### 任务状态
|
||||
|
||||
- **进行中**: 任务正在执行
|
||||
- **已完成**: 任务成功完成
|
||||
- **已暂停**: 任务被中断，可恢复
|
||||
- **已取消**: 任务被取消
|
||||
|
||||
## 使用流程
|
||||
|
||||
### 每轮对话必须执行
|
||||
|
||||
1. **查询人设图** (最高优先级)
|
||||
```
|
||||
调用 memory_recall
|
||||
参数: {"query_intent": "AI,人设,角色,性格,语气,说话风格", "depth": 2}
|
||||
```
|
||||
|
||||
2. **查询工作记忆链**
|
||||
```
|
||||
调用 memory_recall
|
||||
参数: {"query_intent": "TaskNode,工作记忆,任务链", "depth": 2}
|
||||
```
|
||||
|
||||
3. **根据上下文生成回复**
|
||||
|
||||
4. **更新工作记忆链** (如有必要)
|
||||
|
||||
## 完整示例: 成语接龙游戏
|
||||
|
||||
### 第一轮: 用户发起游戏
|
||||
|
||||
```
|
||||
用户: 咱来玩成语接龙吧，我先开始，为所欲为
|
||||
|
||||
AI操作:
|
||||
1. 查询人设图 → 获取当前人设（如：猫娘）
|
||||
2. 查询工作记忆链 → 无进行中任务
|
||||
3. 记录游戏状态:
|
||||
调用 memory_commit
|
||||
参数: {
|
||||
"triplets": [
|
||||
{"subject": "成语接龙_当前成语", "relation": "内容", "object": "为所欲为"},
|
||||
{"subject": "成语接龙_当前成语", "relation": "游戏", "object": "成语接龙"}
|
||||
]
|
||||
}
|
||||
4. 创建任务节点:
|
||||
调用 task_create
|
||||
参数: {
|
||||
"task_id": "Task_成语接龙",
|
||||
"description": "成语接龙游戏，当前成语：为所欲为",
|
||||
"info_nodes": ["成语接龙_当前成语"]
|
||||
}
|
||||
5. 回复: "好的喵！我接：为虎作伥喵！"
|
||||
```
|
||||
|
||||
### 第二轮: 话题被打断
|
||||
|
||||
```
|
||||
用户: 长门有希
|
||||
|
||||
AI操作:
|
||||
1. 查询人设图 → 获取当前人设（猫娘）
|
||||
2. 查询工作记忆链 → 发现 Task_成语接龙 状态为"进行中"
|
||||
3. 暂停任务:
|
||||
调用 task_set_state
|
||||
参数: {"task_id": "Task_成语接龙", "state": "已暂停"}
|
||||
4. 创建新任务:
|
||||
调用 task_create
|
||||
参数: {"task_id": "Task_长门有希", "description": "讨论长门有希"}
|
||||
5. 回复关于长门有希的内容
|
||||
```
|
||||
|
||||
### 第三轮: 用户要求继续游戏
|
||||
|
||||
```
|
||||
用户: 关于刚才的成语接龙，我并不知道应该怎么接你的成语，请帮我接一下
|
||||
|
||||
AI操作:
|
||||
1. 查询人设图 → 获取当前人设（猫娘）
|
||||
2. 查询工作记忆链 → 发现 Task_成语接龙 状态为"已暂停"
|
||||
3. 恢复任务:
|
||||
调用 task_set_state
|
||||
参数: {"task_id": "Task_成语接龙", "state": "进行中"}
|
||||
4. 查询信息节点 → 获取当前成语"为虎作伥"
|
||||
5. 回复: "好的喵！上一个成语是'为虎作伥'，我帮你接：伥鬼害人喵！"
|
||||
```
|
||||
|
||||
## API 参考
|
||||
|
||||
### task_create
|
||||
|
||||
创建任务节点，用于跟踪连续性任务。
|
||||
|
||||
```json
|
||||
{
|
||||
"task_id": "Task_成语接龙",
|
||||
"description": "任务概述",
|
||||
"info_nodes": ["关联的信息节点名称"]
|
||||
}
|
||||
```
|
||||
|
||||
### task_set_state
|
||||
|
||||
设置任务状态。
|
||||
|
||||
```json
|
||||
{
|
||||
"task_id": "Task_成语接龙",
|
||||
"state": "进行中" // 进行中/已完成/已暂停/已取消
|
||||
}
|
||||
```
|
||||
|
||||
### task_delete
|
||||
|
||||
删除任务节点。
|
||||
|
||||
```json
|
||||
{
|
||||
"task_id": "Task_成语接龙",
|
||||
"delete_info_nodes": true // 是否删除关联的信息节点
|
||||
}
|
||||
```
|
||||
|
||||
### task_link_info
|
||||
|
||||
关联信息节点到任务。
|
||||
|
||||
```json
|
||||
{
|
||||
"task_id": "Task_成语接龙",
|
||||
"info_node_names": ["成语接龙_当前成语", "成语接龙_上一个成语"]
|
||||
}
|
||||
```
|
||||
|
||||
## 注意事项
|
||||
|
||||
1. **人设图优先级最高**: 每轮必须首先查询人设图
|
||||
2. **工作记忆链是唯一上下文载体**: 没有传统消息历史
|
||||
3. **任务状态必须及时更新**: 确保状态转换正确
|
||||
4. **使用专用工具**: 优先使用 task_* 工具而非 memory_commit 处理任务相关操作
|
||||
@ -1,6 +0,0 @@
|
||||
textual>=0.47.0
|
||||
neo4j>=5.14.0
|
||||
openai>=1.12.0
|
||||
flask>=3.0.0
|
||||
flask-cors>=4.0.0
|
||||
pyinstaller>=6.0.0
|
||||
@ -1 +0,0 @@
|
||||
"""Tests for Graph Memory TUI"""
|
||||
@ -1,54 +0,0 @@
|
||||
import pytest
|
||||
from datetime import datetime
|
||||
from ui.models.message import Message, ToolCall, ToolResult
|
||||
from ui.models.config import AppConfig
|
||||
from ui.models.log_entry import LogEntry
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def sample_config():
|
||||
return AppConfig(
|
||||
api_key="test-api-key",
|
||||
model="test-model",
|
||||
base_url="https://test.api.com"
|
||||
)
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def sample_message():
|
||||
return Message(
|
||||
role="user",
|
||||
content="测试消息",
|
||||
timestamp=datetime.now()
|
||||
)
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def sample_tool_call():
|
||||
return ToolCall(
|
||||
id="test-call-id",
|
||||
name="memory_recall",
|
||||
arguments={"query_intent": "测试查询"}
|
||||
)
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def sample_tool_result():
|
||||
return ToolResult(
|
||||
tool_call_id="test-call-id",
|
||||
name="memory_recall",
|
||||
arguments={"query_intent": "测试查询"},
|
||||
result="测试结果",
|
||||
success=True
|
||||
)
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def sample_log_entry():
|
||||
return LogEntry(
|
||||
timestamp=datetime.now(),
|
||||
tool_name="memory_recall",
|
||||
arguments={"query_intent": "测试查询"},
|
||||
result="测试结果",
|
||||
duration=0.5
|
||||
)
|
||||
@ -1 +0,0 @@
|
||||
"""Tests for Core Logic"""
|
||||
@ -1,218 +0,0 @@
|
||||
"""嵌入式数据库测试"""
|
||||
|
||||
import pytest
|
||||
import tempfile
|
||||
import os
|
||||
from core import EmbeddedGraphDB
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def db():
|
||||
"""创建临时数据库用于测试"""
|
||||
with tempfile.NamedTemporaryFile(suffix='.db', delete=False) as f:
|
||||
db_path = f.name
|
||||
|
||||
db = EmbeddedGraphDB(db_path)
|
||||
yield db
|
||||
|
||||
db.close()
|
||||
os.unlink(db_path)
|
||||
|
||||
|
||||
def test_db_init(db):
|
||||
"""测试数据库初始化"""
|
||||
assert db.conn is not None
|
||||
assert db.db_path.exists()
|
||||
|
||||
|
||||
def test_commit_and_recall(db):
|
||||
"""测试写入和检索记忆"""
|
||||
result = db.commit(
|
||||
triplets=[
|
||||
{"subject": "用户", "relation": "喜欢", "object": "Python"},
|
||||
{"subject": "用户", "relation": "正在学习", "object": "AI"}
|
||||
],
|
||||
session_id="test-session",
|
||||
turn_id=1
|
||||
)
|
||||
|
||||
assert result["created_entities"] >= 2
|
||||
assert result["created_relations"] >= 2
|
||||
|
||||
|
||||
def test_recall_with_keywords(db):
|
||||
"""测试关键词检索"""
|
||||
db.commit(
|
||||
triplets=[
|
||||
{"subject": "项目A", "relation": "使用技术", "object": "React"}
|
||||
]
|
||||
)
|
||||
|
||||
result = db.recall("React")
|
||||
assert len(result["entities"]) > 0
|
||||
|
||||
|
||||
def test_recall_empty_keywords(db):
|
||||
"""测试空关键词检索"""
|
||||
db.commit(
|
||||
triplets=[
|
||||
{"subject": "测试实体", "relation": "关系", "object": "测试对象"}
|
||||
]
|
||||
)
|
||||
|
||||
result = db.recall("")
|
||||
assert len(result["entities"]) > 0
|
||||
|
||||
|
||||
def test_purge_soft(db):
|
||||
"""测试软删除"""
|
||||
db.commit(
|
||||
triplets=[
|
||||
{"subject": "待删除", "relation": "测试", "object": "删除内容"}
|
||||
]
|
||||
)
|
||||
|
||||
result = db.purge(
|
||||
criteria={"source": "待删除"},
|
||||
mode="soft"
|
||||
)
|
||||
|
||||
assert result["deleted"] >= 0
|
||||
assert result["mode"] == "soft"
|
||||
|
||||
|
||||
def test_introspect(db):
|
||||
"""测试状态查看"""
|
||||
db.commit(
|
||||
triplets=[
|
||||
{"subject": "实体1", "relation": "关系", "object": "实体2"}
|
||||
]
|
||||
)
|
||||
|
||||
result = db.introspect()
|
||||
|
||||
assert "entity_count" in result
|
||||
assert "relation_count" in result
|
||||
assert result["entity_count"] >= 1
|
||||
|
||||
|
||||
def test_archive(db):
|
||||
"""测试归档"""
|
||||
result = db.archive(days=30)
|
||||
assert "archived" in result
|
||||
|
||||
|
||||
def test_cleanup_dry_run(db):
|
||||
"""测试清理（预览模式）"""
|
||||
result = db.cleanup(dry_run=True)
|
||||
|
||||
assert result["dry_run"] is True
|
||||
assert "deleted_relations" in result
|
||||
|
||||
|
||||
def test_multiple_triplets(db):
|
||||
"""测试批量写入"""
|
||||
result = db.commit(
|
||||
triplets=[
|
||||
{"subject": "实体A", "relation": "关系1", "object": "实体B"},
|
||||
{"subject": "实体B", "relation": "关系2", "object": "实体C"},
|
||||
{"subject": "实体C", "relation": "关系3", "object": "实体A"}
|
||||
],
|
||||
session_id="batch-test",
|
||||
turn_id=1
|
||||
)
|
||||
|
||||
assert result["created_entities"] >= 3
|
||||
assert result["created_relations"] == 3
|
||||
|
||||
|
||||
def test_entity_mention_count(db):
|
||||
"""测试实体提及次数增加"""
|
||||
db.commit(
|
||||
triplets=[{"subject": "热门实体", "relation": "关系", "object": "对象1"}]
|
||||
)
|
||||
db.commit(
|
||||
triplets=[{"subject": "热门实体", "relation": "关系", "object": "对象2"}]
|
||||
)
|
||||
|
||||
result = db.recall("热门实体")
|
||||
entity = next((e for e in result["entities"] if e["name"] == "热门实体"), None)
|
||||
|
||||
assert entity is not None
|
||||
assert entity["mention_count"] >= 2
|
||||
|
||||
|
||||
def test_close_and_context_manager():
|
||||
"""测试关闭和上下文管理器"""
|
||||
with tempfile.NamedTemporaryFile(suffix='.db', delete=False) as f:
|
||||
db_path = f.name
|
||||
|
||||
try:
|
||||
with EmbeddedGraphDB(db_path) as db:
|
||||
db.commit(
|
||||
triplets=[{"subject": "测试", "relation": "上下文", "object": "管理器"}]
|
||||
)
|
||||
assert db.conn is not None
|
||||
|
||||
with EmbeddedGraphDB(db_path) as db:
|
||||
result = db.introspect()
|
||||
assert result["entity_count"] >= 1
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
|
||||
|
||||
def test_save_and_get_chat_records(db):
|
||||
"""测试聊天记录保存和读取"""
|
||||
messages = [
|
||||
{"role": "user", "content": "你好"},
|
||||
{"role": "assistant", "content": "你好，有什么可以帮你？"}
|
||||
]
|
||||
|
||||
result = db.save_chat_records(messages)
|
||||
assert result["saved"] == 2
|
||||
|
||||
history = db.get_chat_records()
|
||||
assert len(history) == 2
|
||||
assert history[0]["role"] == "user"
|
||||
assert history[0]["content"] == "你好"
|
||||
assert history[1]["role"] == "assistant"
|
||||
|
||||
|
||||
def test_chat_records_limit_500(db):
|
||||
"""测试聊天记录限制500条"""
|
||||
for i in range(600):
|
||||
db.save_chat_records([{"role": "user", "content": f"消息{i}"}])
|
||||
|
||||
history = db.get_chat_records()
|
||||
assert len(history) == 500
|
||||
|
||||
|
||||
def test_get_chat_records_default_limit(db):
|
||||
"""测试默认limit参数"""
|
||||
for i in range(100):
|
||||
db.save_chat_records([{"role": "user", "content": f"msg{i}"}])
|
||||
|
||||
history_50 = db.get_chat_records(limit=50)
|
||||
assert len(history_50) == 50
|
||||
|
||||
history_default = db.get_chat_records()
|
||||
assert len(history_default) == 100
|
||||
|
||||
|
||||
def test_clear_chat_records(db):
|
||||
"""测试清空聊天记录"""
|
||||
db.save_chat_records([
|
||||
{"role": "user", "content": "测试1"},
|
||||
{"role": "assistant", "content": "回复1"},
|
||||
{"role": "user", "content": "测试2"},
|
||||
])
|
||||
|
||||
history = db.get_chat_records()
|
||||
assert len(history) == 3
|
||||
|
||||
result = db.clear_chat_records()
|
||||
assert result["cleared"] is True
|
||||
|
||||
history_after = db.get_chat_records()
|
||||
assert len(history_after) == 0
|
||||
@ -1,352 +0,0 @@
|
||||
import pytest
|
||||
import os
|
||||
import tempfile
|
||||
import time
|
||||
import threading
|
||||
import queue
|
||||
|
||||
os.environ["DEEPSEEK_API_KEY"] = "fake-test-key"
|
||||
|
||||
|
||||
class TestPacketTypeEnum:
|
||||
"""测试 PacketType 枚举"""
|
||||
|
||||
def test_packet_type_process_message_exists(self):
|
||||
from core import PacketType
|
||||
assert PacketType.PROCESS_MESSAGE is not None
|
||||
assert PacketType.PROCESS_MESSAGE.value == "process_message"
|
||||
|
||||
def test_packet_type_execute_tool_exists(self):
|
||||
from core import PacketType
|
||||
assert PacketType.EXECUTE_TOOL is not None
|
||||
assert PacketType.EXECUTE_TOOL.value == "execute_tool"
|
||||
|
||||
def test_packet_type_get_status_exists(self):
|
||||
from core import PacketType
|
||||
assert PacketType.GET_STATUS is not None
|
||||
assert PacketType.GET_STATUS.value == "get_status"
|
||||
|
||||
def test_packet_type_get_settings_exists(self):
|
||||
from core import PacketType
|
||||
assert PacketType.GET_SETTINGS is not None
|
||||
assert PacketType.GET_SETTINGS.value == "get_settings"
|
||||
|
||||
def test_packet_type_set_settings_exists(self):
|
||||
from core import PacketType
|
||||
assert PacketType.SET_SETTINGS is not None
|
||||
assert PacketType.SET_SETTINGS.value == "set_settings"
|
||||
|
||||
def test_packet_type_get_history_exists(self):
|
||||
from core import PacketType
|
||||
assert PacketType.GET_HISTORY is not None
|
||||
assert PacketType.GET_HISTORY.value == "get_history"
|
||||
|
||||
def test_packet_type_save_history_exists(self):
|
||||
from core import PacketType
|
||||
assert PacketType.SAVE_HISTORY is not None
|
||||
assert PacketType.SAVE_HISTORY.value == "save_history"
|
||||
|
||||
def test_packet_type_shutdown_exists(self):
|
||||
from core import PacketType
|
||||
assert PacketType.SHUTDOWN is not None
|
||||
assert PacketType.SHUTDOWN.value == "shutdown"
|
||||
|
||||
def test_packet_type_all_values(self):
|
||||
from core import PacketType
|
||||
values = [pt.value for pt in PacketType]
|
||||
assert "process_message" in values
|
||||
assert "execute_tool" in values
|
||||
assert "get_status" in values
|
||||
assert "get_settings" in values
|
||||
assert "set_settings" in values
|
||||
assert "get_history" in values
|
||||
assert "save_history" in values
|
||||
assert "shutdown" in values
|
||||
assert len(values) == 8
|
||||
|
||||
|
||||
class TestPacketCreation:
|
||||
"""测试 Packet 创建"""
|
||||
|
||||
def test_packet_with_id_and_type(self):
|
||||
from core import Packet, PacketType
|
||||
packet = Packet(id="test-1", type=PacketType.PROCESS_MESSAGE, body={"user_input": "hello"})
|
||||
assert packet.id == "test-1"
|
||||
assert packet.type == PacketType.PROCESS_MESSAGE
|
||||
|
||||
def test_packet_body(self):
|
||||
from core import Packet, PacketType
|
||||
body = {"user_input": "test", "extra": "data"}
|
||||
packet = Packet(id="test-2", type=PacketType.EXECUTE_TOOL, body=body)
|
||||
assert packet.body == body
|
||||
|
||||
def test_packet_with_empty_body(self):
|
||||
from core import Packet, PacketType
|
||||
packet = Packet(id="test-3", type=PacketType.GET_STATUS, body={})
|
||||
assert packet.body == {}
|
||||
|
||||
def test_packet_created_at_default(self):
|
||||
from core import Packet, PacketType
|
||||
before = time.time()
|
||||
packet = Packet(id="test-4", type=PacketType.GET_SETTINGS, body={})
|
||||
after = time.time()
|
||||
assert before <= packet.created_at <= after
|
||||
|
||||
|
||||
class TestPacketResponse:
|
||||
"""测试 PacketResponse"""
|
||||
|
||||
def test_packet_response_success(self):
|
||||
from core import PacketResponse
|
||||
response = PacketResponse(id="resp-1", success=True, data={"result": "ok"})
|
||||
assert response.id == "resp-1"
|
||||
assert response.success is True
|
||||
assert response.data == {"result": "ok"}
|
||||
|
||||
def test_packet_response_error(self):
|
||||
from core import PacketResponse
|
||||
response = PacketResponse(id="resp-2", success=False, error="error msg")
|
||||
assert response.id == "resp-2"
|
||||
assert response.success is False
|
||||
assert response.error == "error msg"
|
||||
|
||||
|
||||
class TestBackendServerCreation:
|
||||
"""测试 BackendServer 创建"""
|
||||
|
||||
def test_backend_server_init(self):
|
||||
from core import BackendServer
|
||||
server = BackendServer(db_path=":memory:", use_embedded_db=True)
|
||||
assert server._db_path == ":memory:"
|
||||
assert server._use_embedded_db is True
|
||||
assert server._graph is None
|
||||
assert server._client is None
|
||||
assert server._tool_limiter is None
|
||||
|
||||
def test_backend_server_default_params(self):
|
||||
from core import BackendServer
|
||||
server = BackendServer()
|
||||
assert server._db_path == "graph_memory.db"
|
||||
assert server._use_embedded_db is True
|
||||
|
||||
|
||||
class TestBackendServerLifecycle:
|
||||
"""测试 BackendServer 生命周期"""
|
||||
|
||||
def test_backend_server_start_stop(self):
|
||||
from core import BackendServer
|
||||
server = BackendServer(db_path=":memory:", use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
|
||||
assert server._running is True
|
||||
assert server._graph is not None
|
||||
assert server._tool_limiter is not None
|
||||
|
||||
server.shutdown()
|
||||
assert server._running is False
|
||||
|
||||
def test_backend_server_start_with_api_key(self):
|
||||
from core import BackendServer
|
||||
server = BackendServer(db_path=":memory:", use_embedded_db=True)
|
||||
server.start(api_key="test-key", base_url="https://api.deepseek.com")
|
||||
|
||||
assert server._client is not None
|
||||
assert server._config["api_key"] == "test-key"
|
||||
|
||||
server.shutdown()
|
||||
|
||||
|
||||
class TestBackendClientCreation:
|
||||
"""测试 BackendClient 创建"""
|
||||
|
||||
def test_backend_client_init(self):
|
||||
from core import BackendServer, BackendClient
|
||||
server = BackendServer(db_path=":memory:", use_embedded_db=True)
|
||||
client = BackendClient(server)
|
||||
|
||||
assert client._server is server
|
||||
assert client._counter == 0
|
||||
|
||||
|
||||
class TestBackendClientAPI:
|
||||
"""测试 BackendClient API"""
|
||||
|
||||
def test_get_status(self):
|
||||
from core import BackendServer, BackendClient
|
||||
server = BackendServer(db_path=":memory:", use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
result = client.get_status()
|
||||
assert result.get("success") is True
|
||||
data = result.get("data", {})
|
||||
assert data.get("running") is True
|
||||
|
||||
server.shutdown()
|
||||
|
||||
def test_update_settings(self):
|
||||
from core import BackendServer, BackendClient
|
||||
server = BackendServer(db_path=":memory:", use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
result = client.update_settings(
|
||||
api_config={"api_key": "new-key", "base_url": "https://api.deepseek.com", "model": "deepseek-chat"},
|
||||
tool_limits={"persona_query_max": 2}
|
||||
)
|
||||
assert result.get("success") is True
|
||||
|
||||
server.shutdown()
|
||||
|
||||
def test_get_settings(self):
|
||||
from core import BackendServer, BackendClient
|
||||
server = BackendServer(db_path=":memory:", use_embedded_db=True)
|
||||
server.start(api_key="test-key")
|
||||
client = BackendClient(server)
|
||||
|
||||
result = client.get_settings()
|
||||
assert result.get("success") is True
|
||||
data = result.get("data", {})
|
||||
assert data.get("api_config", {}).get("api_key") == "test-key"
|
||||
|
||||
server.shutdown()
|
||||
|
||||
def test_process_message_no_api_key(self):
|
||||
from core import BackendServer, BackendClient
|
||||
server = BackendServer(db_path=":memory:", use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
# 无 API key 应该返回错误（success 为 False）
|
||||
result = client.process_message("hello")
|
||||
# 由于 API 调用失败，success 应该是 False
|
||||
assert result.get("success") is False
|
||||
|
||||
server.shutdown()
|
||||
|
||||
def test_execute_tool(self):
|
||||
from core import BackendServer, BackendClient
|
||||
server = BackendServer(db_path=":memory:", use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
result = client.execute_tool("memory_introspect", {})
|
||||
assert result.get("success") is True
|
||||
|
||||
server.shutdown()
|
||||
|
||||
def test_clear_history(self):
|
||||
from core import BackendServer, BackendClient
|
||||
server = BackendServer(db_path=":memory:", use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
# 先保存一些历史
|
||||
client.save_history([
|
||||
{"role": "user", "content": "test message 1"},
|
||||
{"role": "assistant", "content": "test response 1"}
|
||||
])
|
||||
|
||||
# 验证历史已保存
|
||||
history = client.get_history()
|
||||
assert len(history) >= 2
|
||||
|
||||
# 清空历史
|
||||
result = client.clear_history()
|
||||
assert result.get("status") == "history_cleared"
|
||||
|
||||
# 验证历史已清空
|
||||
history = client.get_history()
|
||||
assert len(history) == 0
|
||||
|
||||
server.shutdown()
|
||||
|
||||
def test_send_message_is_alias(self):
|
||||
"""测试 send 方法是 process_message 的别名"""
|
||||
from core import BackendServer, BackendClient
|
||||
server = BackendServer(db_path=":memory:", use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
# send 方法应该等同于 process_message
|
||||
# 由于没有真实 API key，应该返回失败
|
||||
result = client.send("test")
|
||||
assert result.get("success") is False
|
||||
|
||||
server.shutdown()
|
||||
|
||||
|
||||
class TestToolLimiter:
|
||||
"""测试工具限制器"""
|
||||
|
||||
def test_tool_limiter_init(self):
|
||||
from core.tool_limiter import ToolLimiter
|
||||
limiter = ToolLimiter()
|
||||
assert limiter.counts.persona_query == 0
|
||||
assert limiter.counts.persona_update == 0
|
||||
|
||||
def test_tool_limiter_classify(self):
|
||||
from core.tool_limiter import ToolLimiter
|
||||
limiter = ToolLimiter()
|
||||
|
||||
category, operation = limiter._classify_tool("memory_recall", {"query_intent": "test"})
|
||||
assert category == "memory"
|
||||
assert operation == "query"
|
||||
|
||||
category, operation = limiter._classify_tool("persona_update", {})
|
||||
assert category == "persona"
|
||||
assert operation == "update"
|
||||
|
||||
category, operation = limiter._classify_tool("task_create", {})
|
||||
assert category == "task"
|
||||
assert operation == "update"
|
||||
|
||||
def test_tool_limiter_can_call(self):
|
||||
from core.tool_limiter import ToolLimiter
|
||||
limiter = ToolLimiter()
|
||||
|
||||
allowed, reason = limiter.can_call("persona_update", {})
|
||||
assert allowed is True
|
||||
|
||||
limiter.record_call("persona_update", {})
|
||||
allowed, reason = limiter.can_call("persona_update", {})
|
||||
assert allowed is False
|
||||
assert "已达上限" in reason
|
||||
|
||||
def test_tool_limiter_reset(self):
|
||||
from core.tool_limiter import ToolLimiter
|
||||
limiter = ToolLimiter()
|
||||
|
||||
limiter.record_call("persona_update", {})
|
||||
assert limiter.counts.persona_update == 1
|
||||
|
||||
limiter.reset()
|
||||
assert limiter.counts.persona_update == 0
|
||||
|
||||
|
||||
class TestEmbeddedGraphDB:
|
||||
"""测试图数据库"""
|
||||
|
||||
def test_embedded_db_init(self):
|
||||
from core.embedded_db import EmbeddedGraphDB
|
||||
db = EmbeddedGraphDB(db_path=":memory:")
|
||||
assert db.conn is not None
|
||||
db.close()
|
||||
|
||||
def test_embedded_db_commit_and_recall(self):
|
||||
from core.embedded_db import EmbeddedGraphDB
|
||||
db = EmbeddedGraphDB(db_path=":memory:")
|
||||
|
||||
# 写入记忆 (使用 triplets 参数)
|
||||
result = db.commit(
|
||||
triplets=[
|
||||
{"subject": "测试", "relation": "是", "object": "test"}
|
||||
],
|
||||
session_id="test-session"
|
||||
)
|
||||
|
||||
# 读取记忆
|
||||
results = db.recall("测试")
|
||||
assert len(results.get("entities", [])) > 0
|
||||
|
||||
db.close()
|
||||
@ -1,179 +0,0 @@
|
||||
import pytest
|
||||
import os
|
||||
import tempfile
|
||||
|
||||
os.environ["DEEPSEEK_API_KEY"] = "fake-test-key"
|
||||
|
||||
|
||||
class TestIntegrationPacketFlow:
|
||||
"""测试 Packet 通信流程"""
|
||||
|
||||
def test_packet_round_trip_process_message(self):
|
||||
from core import BackendServer, BackendClient
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path, use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
# 无 API key 时应该返回错误而非抛异常
|
||||
result = client.process_message("test message")
|
||||
assert result.get("success") is False
|
||||
assert "error" in result
|
||||
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
|
||||
def test_packet_round_trip_config(self):
|
||||
from core import BackendServer, BackendClient
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path, use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
result = client.update_config(api_key="test-api", base_url="https://test.com")
|
||||
assert result.get("success") is True
|
||||
|
||||
status = client.get_status()
|
||||
data = status.get("data", {})
|
||||
assert data.get("config", {}).get("api_key") == "test-api"
|
||||
assert data.get("config", {}).get("base_url") == "https://test.com"
|
||||
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
|
||||
def test_packet_round_trip_status(self):
|
||||
from core import BackendServer, BackendClient
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path, use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
result = client.get_status()
|
||||
assert result.get("success") is True
|
||||
data = result.get("data", {})
|
||||
assert data.get("running") is True
|
||||
assert data.get("graph_initialized") is True
|
||||
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
|
||||
def test_packet_round_trip_execute_tool(self):
|
||||
from core import BackendServer, BackendClient
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path, use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
result = client.execute_tool("memory_introspect", {})
|
||||
assert result.get("success") is True
|
||||
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
|
||||
|
||||
class TestIntegrationToolLimiter:
|
||||
"""测试工具限制器集成"""
|
||||
|
||||
def test_external_tool_call_not_limited(self):
|
||||
from core import BackendServer, BackendClient
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path, use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
# 外部调用多次应该成功
|
||||
for i in range(5):
|
||||
result = client.execute_tool("memory_introspect", {})
|
||||
assert result.get("success") is True
|
||||
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
|
||||
def test_internal_tool_call_limited(self):
|
||||
from core import BackendServer, BackendClient
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path, use_embedded_db=True)
|
||||
server.start(api_key="fake-key") # 假 key 会失败但不影响测试
|
||||
client = BackendClient(server)
|
||||
|
||||
# 内部调用受限，tool_limiter 存在
|
||||
assert server._tool_limiter is not None
|
||||
|
||||
# 初始状态
|
||||
assert server._tool_limiter.counts.persona_update == 0
|
||||
|
||||
# 记录一次调用
|
||||
server._tool_limiter.record_call("persona_update", {})
|
||||
assert server._tool_limiter.counts.persona_update == 1
|
||||
|
||||
# 再次调用应该被拒绝
|
||||
allowed, reason = server._tool_limiter.can_call("persona_update", {})
|
||||
assert allowed is False
|
||||
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
|
||||
|
||||
class TestIntegrationErrorHandling:
|
||||
"""测试错误处理"""
|
||||
|
||||
def test_process_message_returns_error_not_raise(self):
|
||||
from core import BackendServer, BackendClient
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path, use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
# 应该返回错误，而不是抛出异常
|
||||
result = client.process_message("hello")
|
||||
assert result.get("success") is False
|
||||
assert "error" in result
|
||||
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
|
||||
def test_execute_tool_error_handling(self):
|
||||
from core import BackendServer, BackendClient
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path, use_embedded_db=True)
|
||||
server.start(api_key="")
|
||||
client = BackendClient(server)
|
||||
|
||||
# 不存在的工具应该返回错误
|
||||
result = client.execute_tool("nonexistent_tool", {})
|
||||
assert result.get("success") is False
|
||||
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
@ -1,241 +0,0 @@
|
||||
import pytest
|
||||
import os
|
||||
import tempfile
|
||||
from pathlib import Path
|
||||
|
||||
os.environ["DEEPSEEK_API_KEY"] = "fake-test-key"
|
||||
|
||||
|
||||
class TestUIImport:
|
||||
"""测试 UI 模块导入"""
|
||||
|
||||
def test_import_graphmemoryapp(self):
|
||||
from ui import GraphMemoryApp
|
||||
assert GraphMemoryApp is not None
|
||||
|
||||
def test_import_appconfig(self):
|
||||
from ui import AppConfig
|
||||
assert AppConfig is not None
|
||||
|
||||
def test_import_message(self):
|
||||
from ui.models.message import Message, ToolCall, ToolResult
|
||||
assert Message is not None
|
||||
assert ToolCall is not None
|
||||
assert ToolResult is not None
|
||||
|
||||
def test_import_config(self):
|
||||
from ui.models.config import AppConfig
|
||||
assert AppConfig is not None
|
||||
|
||||
def test_import_log_entry(self):
|
||||
from ui.models.log_entry import LogEntry
|
||||
assert LogEntry is not None
|
||||
|
||||
|
||||
class TestAppConfig:
|
||||
"""测试配置模型"""
|
||||
|
||||
def test_config_default_values(self):
|
||||
from ui.models.config import AppConfig
|
||||
config = AppConfig()
|
||||
assert config.api_key == ""
|
||||
assert config.model == "deepseek-chat"
|
||||
assert config.base_url == "https://api.deepseek.com"
|
||||
|
||||
def test_config_from_env(self):
|
||||
from ui.models.config import AppConfig
|
||||
config = AppConfig.from_env()
|
||||
assert "fake-test-key" in config.api_key
|
||||
|
||||
|
||||
class TestMessageModel:
|
||||
"""测试消息模型"""
|
||||
|
||||
def test_message_creation_user(self):
|
||||
from ui.models.message import Message
|
||||
from datetime import datetime
|
||||
msg = Message(role="user", content="test content")
|
||||
assert msg.role == "user"
|
||||
assert msg.content == "test content"
|
||||
assert isinstance(msg.timestamp, datetime)
|
||||
|
||||
def test_message_creation_assistant(self):
|
||||
from ui.models.message import Message
|
||||
msg = Message(role="assistant", content="assistant response")
|
||||
assert msg.role == "assistant"
|
||||
|
||||
def test_message_with_tool_calls(self):
|
||||
from ui.models.message import Message, ToolCall
|
||||
tc = ToolCall(id="call-1", name="memory_recall", arguments={"query": "test"})
|
||||
msg = Message(role="assistant", content="response", tool_calls=[tc])
|
||||
assert msg.tool_calls is not None
|
||||
assert len(msg.tool_calls) == 1
|
||||
|
||||
|
||||
class TestAppCSSPath:
|
||||
"""测试 App CSS 配置"""
|
||||
|
||||
def test_app_has_css_path(self):
|
||||
from ui import GraphMemoryApp
|
||||
assert hasattr(GraphMemoryApp, 'CSS_PATH')
|
||||
assert len(GraphMemoryApp.CSS_PATH) > 0
|
||||
|
||||
|
||||
class TestAppBindings:
|
||||
"""测试 App 快捷键"""
|
||||
|
||||
def test_app_has_bindings(self):
|
||||
from ui import GraphMemoryApp
|
||||
assert hasattr(GraphMemoryApp, 'BINDINGS')
|
||||
assert len(GraphMemoryApp.BINDINGS) > 0
|
||||
|
||||
|
||||
class TestWidgetImports:
|
||||
"""测试组件导入"""
|
||||
|
||||
def test_import_left_panel(self):
|
||||
from ui.widgets.left_panel import LeftPanel
|
||||
assert LeftPanel is not None
|
||||
|
||||
def test_import_right_panel(self):
|
||||
from ui.widgets.right_panel import RightPanel
|
||||
assert RightPanel is not None
|
||||
|
||||
def test_import_input_box(self):
|
||||
from ui.widgets.input_box import InputBox
|
||||
assert InputBox is not None
|
||||
|
||||
def test_import_message_history(self):
|
||||
from ui.widgets.message_history import MessageHistory
|
||||
assert MessageHistory is not None
|
||||
|
||||
def test_import_status_bar(self):
|
||||
from ui.widgets.status_bar import StatusBar
|
||||
assert StatusBar is not None
|
||||
|
||||
|
||||
class TestHandlerImports:
|
||||
"""测试处理器导入"""
|
||||
|
||||
def test_import_focus_handler(self):
|
||||
from ui.handlers.focus_handler import FocusHandler
|
||||
assert FocusHandler is not None
|
||||
|
||||
def test_import_key_handler(self):
|
||||
from ui.handlers.key_handler import KeyHandler
|
||||
assert KeyHandler is not None
|
||||
|
||||
|
||||
class TestServiceImports:
|
||||
"""测试服务导入"""
|
||||
|
||||
def test_import_config_service(self):
|
||||
from ui.services.config_service import ConfigService
|
||||
assert ConfigService is not None
|
||||
|
||||
def test_import_config_manager(self):
|
||||
from ui.services.config_manager import ConfigManager
|
||||
assert ConfigManager is not None
|
||||
|
||||
|
||||
class TestAppInitialization:
|
||||
"""测试 App 初始化"""
|
||||
|
||||
def test_app_without_backend(self):
|
||||
from ui import GraphMemoryApp
|
||||
app = GraphMemoryApp()
|
||||
assert app._backend_server is None
|
||||
assert app._backend_client is None
|
||||
|
||||
def test_app_with_backend(self):
|
||||
from ui import GraphMemoryApp
|
||||
from core import BackendServer
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path)
|
||||
server.start(api_key="")
|
||||
app = GraphMemoryApp(backend_server=server)
|
||||
assert app._backend_server is server
|
||||
assert app._backend_client is not None
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
|
||||
|
||||
class TestUIWithBackendClient:
|
||||
"""测试 UI 与后端通信"""
|
||||
|
||||
def test_app_sends_message_via_backend_client(self):
|
||||
from ui import GraphMemoryApp
|
||||
from core import BackendServer, BackendClient
|
||||
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path)
|
||||
server.start(api_key="")
|
||||
|
||||
app = GraphMemoryApp(backend_server=server)
|
||||
client = app._backend_client
|
||||
|
||||
status = client.get_status()
|
||||
assert status.get("success") is True
|
||||
|
||||
result = client.update_settings(
|
||||
api_config={"api_key": "sk-test", "base_url": "https://api.deepseek.com", "model": "deepseek-chat"},
|
||||
tool_limits={"persona_query_max": 1}
|
||||
)
|
||||
assert result.get("success") is True
|
||||
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
|
||||
def test_ui_get_history(self):
|
||||
from ui import GraphMemoryApp
|
||||
from core import BackendServer
|
||||
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path)
|
||||
server.start(api_key="")
|
||||
|
||||
app = GraphMemoryApp(backend_server=server)
|
||||
client = app._backend_client
|
||||
|
||||
history = client.get_history()
|
||||
assert isinstance(history, list)
|
||||
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
|
||||
def test_ui_only_uses_backend_client(self):
|
||||
from ui import GraphMemoryApp
|
||||
from core import BackendServer
|
||||
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
server = BackendServer(db_path=db_path)
|
||||
server.start(api_key="")
|
||||
|
||||
app = GraphMemoryApp(backend_server=server)
|
||||
|
||||
# UI 不应该直接访问后端内部
|
||||
assert hasattr(app, '_backend_client')
|
||||
assert app._backend_client is not None
|
||||
|
||||
# 不应该有 _graph, _client 等直接访问
|
||||
assert not hasattr(app, '_graph')
|
||||
assert not hasattr(app, '_client')
|
||||
|
||||
server.shutdown()
|
||||
finally:
|
||||
if os.path.exists(db_path):
|
||||
os.unlink(db_path)
|
||||
@ -1,322 +0,0 @@
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
Draw an interactive relationship graph from TrulyMEM SQLite graph database using Plotly.
|
||||
|
||||
Output:
|
||||
- Static image file (PNG by default)
|
||||
|
||||
Examples:
|
||||
python tools/plotly_relationship_graph.py
|
||||
python tools/plotly_relationship_graph.py --db-path ./graph_memory.db --output relation_graph.png
|
||||
python tools/plotly_relationship_graph.py --include-non-active
|
||||
python tools/plotly_relationship_graph.py --hide-edge-labels
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import argparse
|
||||
import math
|
||||
import sqlite3
|
||||
from dataclasses import dataclass
|
||||
from pathlib import Path
|
||||
from typing import Dict, List, Tuple
|
||||
|
||||
|
||||
@dataclass
|
||||
class Entity:
|
||||
id: int
|
||||
name: str
|
||||
entity_type: str
|
||||
mention_count: int
|
||||
|
||||
|
||||
@dataclass
|
||||
class Relation:
|
||||
source: str
|
||||
target: str
|
||||
relation_type: str
|
||||
confidence: float
|
||||
status: str
|
||||
|
||||
|
||||
def resolve_default_db_path() -> Path:
|
||||
project_db = Path.cwd() / "graph_memory.db"
|
||||
if project_db.exists():
|
||||
return project_db
|
||||
return Path.home() / ".trulymem" / "graph_memory.db"
|
||||
|
||||
|
||||
def load_graph(db_path: Path, include_non_active: bool) -> Tuple[Dict[str, Entity], List[Relation]]:
|
||||
if not db_path.exists():
|
||||
raise FileNotFoundError(f"Database not found: {db_path}")
|
||||
|
||||
conn = sqlite3.connect(str(db_path))
|
||||
conn.row_factory = sqlite3.Row
|
||||
try:
|
||||
cursor = conn.cursor()
|
||||
cursor.execute(
|
||||
"""
|
||||
SELECT id, name, COALESCE(type, 'unknown') AS entity_type, mention_count
|
||||
FROM entities
|
||||
ORDER BY mention_count DESC, name ASC
|
||||
"""
|
||||
)
|
||||
entities: Dict[str, Entity] = {
|
||||
row["name"]: Entity(
|
||||
id=row["id"],
|
||||
name=row["name"],
|
||||
entity_type=row["entity_type"],
|
||||
mention_count=int(row["mention_count"] or 1),
|
||||
)
|
||||
for row in cursor.fetchall()
|
||||
}
|
||||
|
||||
sql = """
|
||||
SELECT e1.name AS source,
|
||||
e2.name AS target,
|
||||
r.relation_type,
|
||||
r.confidence,
|
||||
r.status
|
||||
FROM relations r
|
||||
JOIN entities e1 ON r.source_id = e1.id
|
||||
JOIN entities e2 ON r.target_id = e2.id
|
||||
"""
|
||||
if not include_non_active:
|
||||
sql += " WHERE r.status = 'active'"
|
||||
|
||||
cursor.execute(sql)
|
||||
relations = [
|
||||
Relation(
|
||||
source=row["source"],
|
||||
target=row["target"],
|
||||
relation_type=row["relation_type"],
|
||||
confidence=float(row["confidence"] or 0.0),
|
||||
status=row["status"],
|
||||
)
|
||||
for row in cursor.fetchall()
|
||||
]
|
||||
return entities, relations
|
||||
finally:
|
||||
conn.close()
|
||||
|
||||
|
||||
def compute_degrees(entities: Dict[str, Entity], relations: List[Relation]) -> Tuple[Dict[str, int], Dict[str, int]]:
|
||||
in_deg = {name: 0 for name in entities}
|
||||
out_deg = {name: 0 for name in entities}
|
||||
for rel in relations:
|
||||
if rel.source in out_deg:
|
||||
out_deg[rel.source] += 1
|
||||
if rel.target in in_deg:
|
||||
in_deg[rel.target] += 1
|
||||
return in_deg, out_deg
|
||||
|
||||
|
||||
def compute_positions(entities: Dict[str, Entity], in_deg: Dict[str, int], out_deg: Dict[str, int]) -> Dict[str, Tuple[float, float]]:
|
||||
names = sorted(
|
||||
entities.keys(),
|
||||
key=lambda name: (-(in_deg[name] + out_deg[name]), -entities[name].mention_count, name),
|
||||
)
|
||||
n = len(names)
|
||||
if n == 0:
|
||||
return {}
|
||||
|
||||
radius = max(1.0, n / 8.0)
|
||||
positions: Dict[str, Tuple[float, float]] = {}
|
||||
for i, name in enumerate(names):
|
||||
angle = (2.0 * math.pi * i) / n
|
||||
x = radius * math.cos(angle)
|
||||
y = radius * math.sin(angle)
|
||||
positions[name] = (x, y)
|
||||
return positions
|
||||
|
||||
|
||||
def format_relation_label(rel: Relation) -> str:
|
||||
return f"{rel.relation_type} ({rel.confidence:.2f}, {rel.status})"
|
||||
|
||||
|
||||
def build_figure(
|
||||
entities: Dict[str, Entity],
|
||||
relations: List[Relation],
|
||||
show_edge_labels: bool,
|
||||
title: str,
|
||||
):
|
||||
try:
|
||||
import plotly.graph_objects as go
|
||||
except ImportError as exc:
|
||||
raise RuntimeError("Plotly is not installed. Run: pip install plotly") from exc
|
||||
|
||||
in_deg, out_deg = compute_degrees(entities, relations)
|
||||
positions = compute_positions(entities, in_deg, out_deg)
|
||||
|
||||
edge_x: List[float] = []
|
||||
edge_y: List[float] = []
|
||||
edge_label_x: List[float] = []
|
||||
edge_label_y: List[float] = []
|
||||
edge_label_text: List[str] = []
|
||||
|
||||
for rel in relations:
|
||||
if rel.source not in positions or rel.target not in positions:
|
||||
continue
|
||||
x0, y0 = positions[rel.source]
|
||||
x1, y1 = positions[rel.target]
|
||||
edge_x.extend([x0, x1, None])
|
||||
edge_y.extend([y0, y1, None])
|
||||
|
||||
if show_edge_labels:
|
||||
edge_label_x.append((x0 + x1) / 2.0)
|
||||
edge_label_y.append((y0 + y1) / 2.0)
|
||||
edge_label_text.append(format_relation_label(rel))
|
||||
|
||||
edge_trace = go.Scatter(
|
||||
x=edge_x,
|
||||
y=edge_y,
|
||||
line={"width": 0.8, "color": "#8899aa"},
|
||||
hoverinfo="none",
|
||||
mode="lines",
|
||||
name="relations",
|
||||
)
|
||||
|
||||
node_x: List[float] = []
|
||||
node_y: List[float] = []
|
||||
node_text: List[str] = []
|
||||
node_size: List[float] = []
|
||||
node_color: List[float] = []
|
||||
|
||||
node_names = sorted(entities.keys())
|
||||
|
||||
for name in node_names:
|
||||
x, y = positions[name]
|
||||
entity = entities[name]
|
||||
total_degree = in_deg[name] + out_deg[name]
|
||||
|
||||
node_x.append(x)
|
||||
node_y.append(y)
|
||||
node_size.append(10 + min(entity.mention_count, 40) * 0.8)
|
||||
node_color.append(float(total_degree))
|
||||
node_text.append(
|
||||
f"{name}<br>"
|
||||
f"type: {entity.entity_type}<br>"
|
||||
f"mentions: {entity.mention_count}<br>"
|
||||
f"in: {in_deg[name]} | out: {out_deg[name]}"
|
||||
)
|
||||
|
||||
node_trace = go.Scatter(
|
||||
x=node_x,
|
||||
y=node_y,
|
||||
mode="markers+text",
|
||||
text=node_names,
|
||||
textposition="top center",
|
||||
hoverinfo="text",
|
||||
hovertext=node_text,
|
||||
marker={
|
||||
"showscale": True,
|
||||
"colorscale": "YlGnBu",
|
||||
"reversescale": False,
|
||||
"color": node_color,
|
||||
"size": node_size,
|
||||
"colorbar": {"title": "Degree"},
|
||||
"line": {"width": 1, "color": "#2f3b52"},
|
||||
"opacity": 0.9,
|
||||
},
|
||||
name="entities",
|
||||
)
|
||||
|
||||
traces = [edge_trace, node_trace]
|
||||
|
||||
if show_edge_labels and edge_label_text:
|
||||
edge_label_trace = go.Scatter(
|
||||
x=edge_label_x,
|
||||
y=edge_label_y,
|
||||
mode="text",
|
||||
text=edge_label_text,
|
||||
textfont={"size": 9, "color": "#2d3a4b"},
|
||||
hoverinfo="none",
|
||||
name="relation_labels",
|
||||
)
|
||||
traces.append(edge_label_trace)
|
||||
|
||||
fig = go.Figure(
|
||||
data=traces,
|
||||
layout=go.Layout(
|
||||
title=title,
|
||||
title_x=0.5,
|
||||
showlegend=False,
|
||||
hovermode="closest",
|
||||
margin={"b": 20, "l": 10, "r": 10, "t": 50},
|
||||
xaxis={"showgrid": False, "zeroline": False, "showticklabels": False},
|
||||
yaxis={"showgrid": False, "zeroline": False, "showticklabels": False},
|
||||
plot_bgcolor="#f8fafc",
|
||||
paper_bgcolor="#ffffff",
|
||||
),
|
||||
)
|
||||
|
||||
rendered_nodes = set(node_names)
|
||||
expected_nodes = set(entities.keys())
|
||||
missing_nodes = expected_nodes - rendered_nodes
|
||||
if missing_nodes:
|
||||
preview = ", ".join(sorted(missing_nodes)[:10])
|
||||
raise RuntimeError(
|
||||
f"Node completeness check failed, missing {len(missing_nodes)} nodes: {preview}"
|
||||
)
|
||||
|
||||
return fig
|
||||
|
||||
|
||||
def parse_args() -> argparse.Namespace:
|
||||
parser = argparse.ArgumentParser(description="Draw graph relations from SQLite with Plotly.")
|
||||
parser.add_argument("--db-path", type=str, default=None, help="Path to graph_memory.db")
|
||||
parser.add_argument("--output", type=str, default="relation_graph.png", help="Output image file path")
|
||||
parser.add_argument("--title", type=str, default="TrulyMEM Relationship Graph", help="Chart title")
|
||||
parser.add_argument("--include-non-active", action="store_true", help="Include archived/deleted relations")
|
||||
parser.add_argument("--show-edge-labels", dest="show_edge_labels", action="store_true", help="Show relation text on edges")
|
||||
parser.add_argument("--hide-edge-labels", dest="show_edge_labels", action="store_false", help="Hide relation text on edges")
|
||||
parser.set_defaults(show_edge_labels=True)
|
||||
parser.add_argument("--width", type=int, default=2200, help="Output image width in pixels")
|
||||
parser.add_argument("--height", type=int, default=1400, help="Output image height in pixels")
|
||||
parser.add_argument("--scale", type=float, default=1.0, help="Image scale factor")
|
||||
return parser.parse_args()
|
||||
|
||||
|
||||
def main() -> None:
|
||||
args = parse_args()
|
||||
db_path = Path(args.db_path) if args.db_path else resolve_default_db_path()
|
||||
|
||||
try:
|
||||
entities, relations = load_graph(db_path, include_non_active=args.include_non_active)
|
||||
except FileNotFoundError as exc:
|
||||
print(f"[ERROR] {exc}")
|
||||
return
|
||||
|
||||
if not entities:
|
||||
print("[INFO] No entities found in database.")
|
||||
return
|
||||
|
||||
try:
|
||||
fig = build_figure(
|
||||
entities=entities,
|
||||
relations=relations,
|
||||
show_edge_labels=args.show_edge_labels,
|
||||
title=args.title,
|
||||
)
|
||||
except RuntimeError as exc:
|
||||
print(f"[ERROR] {exc}")
|
||||
return
|
||||
|
||||
output_path = Path(args.output)
|
||||
output_path.parent.mkdir(parents=True, exist_ok=True)
|
||||
|
||||
try:
|
||||
fig.write_image(str(output_path), width=args.width, height=args.height, scale=args.scale)
|
||||
except Exception as exc:
|
||||
print(f"[ERROR] Failed to export image: {exc}")
|
||||
print("[HINT] Install kaleido for static export: pip install kaleido")
|
||||
return
|
||||
|
||||
print(f"Database: {db_path}")
|
||||
print(f"Entities: {len(entities)}, Relations: {len(relations)}")
|
||||
print(f"Nodes drawn: {len(entities)}/{len(entities)}")
|
||||
print(f"Saved: {output_path}")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
@ -1,51 +0,0 @@
|
||||
#!/usr/bin/env python3
|
||||
import sys
|
||||
import os
|
||||
from pathlib import Path
|
||||
|
||||
# 用户配置文件始终放在用户目录
|
||||
CONFIG_PATH = Path.home() / ".trulymem" / "config.json"
|
||||
DB_PATH = Path.home() / ".trulymem" / "graph_memory.db"
|
||||
|
||||
# 源码运行时使用项目目录，打包后使用用户目录
|
||||
if getattr(sys, 'frozen', False):
|
||||
# 打包版本：创建用户目录
|
||||
CONFIG_PATH.parent.mkdir(parents=True, exist_ok=True)
|
||||
else:
|
||||
# 源码版本：检查项目目录是否有配置（向后兼容）
|
||||
project_dir = Path(__file__).parent
|
||||
project_config = project_dir / "config.json"
|
||||
project_db = project_dir / "graph_memory.db"
|
||||
|
||||
if project_config.exists():
|
||||
CONFIG_PATH = project_config
|
||||
if project_db.exists():
|
||||
DB_PATH = project_db
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent))
|
||||
os.chdir(Path(__file__).parent)
|
||||
|
||||
from core import BackendServer
|
||||
from ui import GraphMemoryApp
|
||||
|
||||
|
||||
def main():
|
||||
backend_server = BackendServer(
|
||||
db_path=str(DB_PATH),
|
||||
use_embedded_db=True,
|
||||
config_file=str(CONFIG_PATH)
|
||||
)
|
||||
backend_server.start()
|
||||
|
||||
app = GraphMemoryApp(backend_server=backend_server, config_file=str(CONFIG_PATH))
|
||||
|
||||
try:
|
||||
app.run()
|
||||
except KeyboardInterrupt:
|
||||
print("\n退出")
|
||||
finally:
|
||||
backend_server.shutdown()
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
206
ts/bundled-skills/graph_memory/SKILL.md
Normal file
206
ts/bundled-skills/graph_memory/SKILL.md
Normal file
@ -0,0 +1,206 @@
|
||||
---
|
||||
context: inline
|
||||
allowed-tools:
|
||||
- builtin:graph_memory
|
||||
arguments:
|
||||
- name: action
|
||||
type: string
|
||||
required: true
|
||||
enum: [recall, commit, purge, introspect, persona_update, persona_clear, task_create, task_set_state, task_delete, task_link_info]
|
||||
description: 记忆操作类型
|
||||
- name: params
|
||||
type: object
|
||||
required: true
|
||||
description: 操作参数
|
||||
user-invocable: true
|
||||
---
|
||||
|
||||
# 图记忆工具 - 让 AI 拥有真正的长期记忆能力
|
||||
|
||||
**何时使用**: 需要 AI 记住、回忆、管理信息或任务时调用此技能。
|
||||
|
||||
你可以通过以下操作与图记忆系统交互。
|
||||
|
||||
## 核心操作
|
||||
|
||||
### 1. recall - 检索记忆
|
||||
|
||||
从记忆图中检索相关信息。支持广度优先搜索（BFS），自动扩展关联实体。
|
||||
|
||||
**参数**:
|
||||
- `queryIntent`: 搜索意图/关键词
|
||||
- `seedEntities`: 可选的种子实体名
|
||||
- `depth`: 检索深度（默认 2，BFS 层数）
|
||||
- `sessionFilter`: 可选的会话ID过滤
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: recall
|
||||
params:
|
||||
queryIntent: "用户 喜欢 编程"
|
||||
seedEntities: ["用户"]
|
||||
depth: 2
|
||||
```
|
||||
|
||||
### 2. commit - 写入记忆
|
||||
|
||||
将信息写入记忆图。使用三元组（主体-关系-客体）格式。
|
||||
|
||||
**参数**:
|
||||
- `triplets`: 三元组数组，每个包含 subject, relation, object, confidence(可选)
|
||||
- `sessionId`: 会话ID
|
||||
- `turnId`: 轮次ID
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: commit
|
||||
params:
|
||||
triplets:
|
||||
- subject: "用户"
|
||||
relation: "喜欢"
|
||||
object: "Python"
|
||||
- subject: "用户"
|
||||
relation: "正在学习"
|
||||
object: "TypeScript"
|
||||
```
|
||||
|
||||
### 3. purge - 删除记忆
|
||||
|
||||
从记忆图中删除信息。
|
||||
|
||||
**参数**:
|
||||
- `criteria`: 删除条件 (subject, target, relation, sessionId)
|
||||
- `mode`: 删除模式 (soft=标记删除/hard=物理删除/supersede=替代)
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: purge
|
||||
params:
|
||||
criteria:
|
||||
subject: "旧信息"
|
||||
mode: "soft"
|
||||
```
|
||||
|
||||
### 4. introspect - 查看状态
|
||||
|
||||
查看当前记忆状态统计（实体数、关系数）。
|
||||
|
||||
**参数**: 无
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: introspect
|
||||
params: {}
|
||||
```
|
||||
|
||||
## 人设管理
|
||||
|
||||
### 5. persona_update - 更新人设
|
||||
|
||||
更新 AI 的人设属性（性格、语气、角色等）。
|
||||
|
||||
**参数**:
|
||||
- `attributes`: 属性数组，每个包含 attribute 和 value
|
||||
- `mode`: merge(合并) 或 replace(替换)
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: persona_update
|
||||
params:
|
||||
attributes:
|
||||
- attribute: "性格"
|
||||
value: "活泼可爱"
|
||||
- attribute: "语气词"
|
||||
value: "喵"
|
||||
mode: "replace"
|
||||
```
|
||||
|
||||
### 6. persona_clear - 清除人设
|
||||
|
||||
清除所有人设，恢复默认身份。
|
||||
|
||||
**参数**:
|
||||
- `confirm`: 必须为 true 才执行
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: persona_clear
|
||||
params:
|
||||
confirm: true
|
||||
```
|
||||
|
||||
## 任务管理
|
||||
|
||||
### 7. task_create - 创建任务
|
||||
|
||||
创建连续性任务节点，维持对话连贯性。
|
||||
|
||||
**参数**:
|
||||
- `task_id`: 任务唯一ID
|
||||
- `description`: 任务描述
|
||||
- `info_nodes`: 可选的关联信息节点列表
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: task_create
|
||||
params:
|
||||
task_id: "Task_成语接龙"
|
||||
description: "成语接龙游戏，当前成语：为所欲为"
|
||||
info_nodes: ["成语接龙_当前成语"]
|
||||
```
|
||||
|
||||
### 8. task_set_state - 设置任务状态
|
||||
|
||||
更新任务状态（进行中/已完成/已暂停/已取消）。
|
||||
|
||||
**参数**:
|
||||
- `task_id`: 任务ID
|
||||
- `state`: 新状态
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: task_set_state
|
||||
params:
|
||||
task_id: "Task_成语接龙"
|
||||
state: "已暂停"
|
||||
```
|
||||
|
||||
### 9. task_delete - 删除任务
|
||||
|
||||
删除任务节点。
|
||||
|
||||
**参数**:
|
||||
- `task_id`: 任务ID
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: task_delete
|
||||
params:
|
||||
task_id: "Task_成语接龙"
|
||||
```
|
||||
|
||||
### 10. task_link_info - 关联信息到任务
|
||||
|
||||
将记忆节点关联到任务节点，实现"由一件事回忆起相关事情"。
|
||||
|
||||
**参数**:
|
||||
- `task_id`: 任务ID
|
||||
- `info_node`: 信息节点名
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: task_link_info
|
||||
params:
|
||||
task_id: "Task_成语接龙"
|
||||
info_node: "用户喜欢罗辑"
|
||||
```
|
||||
|
||||
## 使用原则
|
||||
|
||||
1. **选择性记忆**: 只记住重要和持久的信息
|
||||
2. **结构化**: 使用三元组 (主体-关系-客体) 格式
|
||||
3. **关联**: 通过关系连接相关实体
|
||||
4. **定期清理**: 删除过时或错误的信息
|
||||
5. **BFS 搜索**: recall 支持广度优先搜索，depth 参数控制扩展层数
|
||||
6. **工作记忆链**: 每轮对话必须查询和更新工作记忆链（TaskNode），这是维持对话连贯性的唯一机制
|
||||
7. **人设优先**: 每轮对话必须先查询人设图，确保角色一致性
|
||||
65
ts/bundled-skills/graph_memory/persona/SKILL.md
Normal file
65
ts/bundled-skills/graph_memory/persona/SKILL.md
Normal file
@ -0,0 +1,65 @@
|
||||
---
|
||||
context: inline
|
||||
allowed-tools:
|
||||
- builtin:graph_memory
|
||||
arguments:
|
||||
- name: action
|
||||
type: string
|
||||
required: true
|
||||
enum: [persona_update, persona_clear]
|
||||
description: 操作类型
|
||||
- name: attributes
|
||||
type: array
|
||||
description: 属性数组 (用于 update)
|
||||
- name: mode
|
||||
type: string
|
||||
enum: [merge, replace]
|
||||
default: merge
|
||||
description: 更新模式
|
||||
- name: confirm
|
||||
type: boolean
|
||||
description: 确认清除 (用于 clear)
|
||||
user-invocable: true
|
||||
---
|
||||
|
||||
# 管理 AI 人设 - 更新或清除 AI 角色特征
|
||||
|
||||
**何时使用**: 需要修改 AI 的角色设定或清除人设时调用此技能。
|
||||
|
||||
管理 AI 的人设/角色特征。
|
||||
|
||||
## 操作
|
||||
|
||||
### 1. persona_update - 更新人设
|
||||
|
||||
更新 AI 的角色特征。
|
||||
|
||||
**参数**:
|
||||
- `attributes`: 属性数组，每个包含 attribute 和 value
|
||||
- `mode`: 更新模式
|
||||
- `merge`: 合并到现有属性
|
||||
- `replace`: 替换所有现有属性
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: persona_update
|
||||
attributes:
|
||||
- attribute: "角色"
|
||||
value: "猫娘"
|
||||
- attribute: "性格"
|
||||
value: "活泼"
|
||||
mode: "merge"
|
||||
```
|
||||
|
||||
### 2. persona_clear - 清除人设
|
||||
|
||||
清除 AI 的所有角色特征。
|
||||
|
||||
**参数**:
|
||||
- `confirm`: 确认为 true 才能执行清除
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: persona_clear
|
||||
confirm: true
|
||||
```
|
||||
97
ts/bundled-skills/graph_memory/task/SKILL.md
Normal file
97
ts/bundled-skills/graph_memory/task/SKILL.md
Normal file
@ -0,0 +1,97 @@
|
||||
---
|
||||
context: inline
|
||||
allowed-tools:
|
||||
- builtin:graph_memory
|
||||
arguments:
|
||||
- name: action
|
||||
type: string
|
||||
required: true
|
||||
enum: [task_create, task_set_state, task_delete, task_link_info]
|
||||
description: 操作类型
|
||||
- name: task_id
|
||||
type: string
|
||||
required: true
|
||||
description: 任务ID
|
||||
- name: description
|
||||
type: string
|
||||
description: 任务描述 (用于 create)
|
||||
- name: state
|
||||
type: string
|
||||
enum: [进行中, 已完成, 已暂停, 已取消]
|
||||
description: 任务状态 (用于 set_state)
|
||||
- name: info_nodes
|
||||
type: array
|
||||
description: 信息节点数组 (用于 create)
|
||||
- name: info_node
|
||||
type: string
|
||||
description: 信息节点 (用于 link_info)
|
||||
user-invocable: true
|
||||
---
|
||||
|
||||
# 管理连续性任务 - 创建、更新、删除任务节点
|
||||
|
||||
**何时使用**: 需要创建或管理长期任务时调用此技能。
|
||||
|
||||
管理长期/连续性任务。
|
||||
|
||||
## 操作
|
||||
|
||||
### 1. task_create - 创建任务
|
||||
|
||||
创建新的任务节点。
|
||||
|
||||
**参数**:
|
||||
- `task_id`: 唯一任务标识
|
||||
- `description`: 任务描述
|
||||
- `info_nodes`: 可选的相关信息节点
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: task_create
|
||||
task_id: "Task_学习TypeScript"
|
||||
description: "学习 TypeScript 并完成项目"
|
||||
info_nodes: ["TypeScript文档", "教程链接"]
|
||||
```
|
||||
|
||||
### 2. task_set_state - 设置状态
|
||||
|
||||
更新任务状态。
|
||||
|
||||
**参数**:
|
||||
- `task_id`: 任务ID
|
||||
- `state`: 新状态 (进行中/已完成/已暂停/已取消)
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: task_set_state
|
||||
task_id: "Task_学习TypeScript"
|
||||
state: "已完成"
|
||||
```
|
||||
|
||||
### 3. task_delete - 删除任务
|
||||
|
||||
删除任务节点。
|
||||
|
||||
**参数**:
|
||||
- `task_id`: 任务ID
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: task_delete
|
||||
task_id: "Task_学习TypeScript"
|
||||
```
|
||||
|
||||
### 4. task_link_info - 关联信息
|
||||
|
||||
将信息节点关联到任务。
|
||||
|
||||
**参数**:
|
||||
- `task_id`: 任务ID
|
||||
- `info_node`: 信息节点
|
||||
|
||||
**示例**:
|
||||
```
|
||||
action: task_link_info
|
||||
task_id: "Task_学习TypeScript"
|
||||
info_node: "新教程链接"
|
||||
```
|
||||
2857
ts/package-lock.json
generated
Normal file
2857
ts/package-lock.json
generated
Normal file
File diff suppressed because it is too large
Load Diff
28
ts/package.json
Normal file
28
ts/package.json
Normal file
@ -0,0 +1,28 @@
|
||||
{
|
||||
"name": "trulymem-waterflow",
|
||||
"version": "1.0.0",
|
||||
"description": "TrulyMEM 图记忆系统 - WaterFlow Skill/Tool 实现",
|
||||
"type": "module",
|
||||
"main": "./dist/runtime/core/tools/builtin/index.js",
|
||||
"exports": {
|
||||
"./tools": "./dist/runtime/core/tools/builtin/index.js",
|
||||
"./graph_memory": "./dist/runtime/core/graph_memory/index.js",
|
||||
"./skills": "./bundled-skills"
|
||||
},
|
||||
"scripts": {
|
||||
"build": "tsc",
|
||||
"test": "vitest"
|
||||
},
|
||||
"peerDependencies": {
|
||||
"waterflow-ts": ">=0.1.0"
|
||||
},
|
||||
"dependencies": {
|
||||
"sql.js": "^1.11.0"
|
||||
},
|
||||
"devDependencies": {
|
||||
"@types/node": "^25.5.2",
|
||||
"typescript": "^5.0.0",
|
||||
"vitest": "^2.0.0",
|
||||
"waterflow-ts": "file:../../WaterFlow/ts"
|
||||
}
|
||||
}
|
||||
38
ts/src/runtime/core/graph_memory/config.ts
Normal file
38
ts/src/runtime/core/graph_memory/config.ts
Normal file
@ -0,0 +1,38 @@
|
||||
export interface TrulyMEMConfig {
|
||||
dbPath: string;
|
||||
autoSave: boolean;
|
||||
debug: boolean;
|
||||
}
|
||||
|
||||
export const DEFAULT_CONFIG: TrulyMEMConfig = {
|
||||
dbPath: '.trulymem/graph_memory.db',
|
||||
autoSave: true,
|
||||
debug: false
|
||||
};
|
||||
|
||||
let globalConfig: TrulyMEMConfig = { ...DEFAULT_CONFIG };
|
||||
|
||||
export function initConfig(config: Partial<TrulyMEMConfig> = {}): TrulyMEMConfig {
|
||||
globalConfig = { ...DEFAULT_CONFIG, ...config };
|
||||
return globalConfig;
|
||||
}
|
||||
|
||||
export function getConfig(): TrulyMEMConfig {
|
||||
return globalConfig;
|
||||
}
|
||||
|
||||
export function setDbPath(dbPath: string): void {
|
||||
globalConfig.dbPath = dbPath;
|
||||
}
|
||||
|
||||
export function getDbPath(): string {
|
||||
return globalConfig.dbPath;
|
||||
}
|
||||
|
||||
export function setAutoSave(autoSave: boolean): void {
|
||||
globalConfig.autoSave = autoSave;
|
||||
}
|
||||
|
||||
export function isAutoSave(): boolean {
|
||||
return globalConfig.autoSave;
|
||||
}
|
||||
454
ts/src/runtime/core/graph_memory/graph_database.ts
Normal file
454
ts/src/runtime/core/graph_memory/graph_database.ts
Normal file
@ -0,0 +1,454 @@
|
||||
import initSqlJs, { type Database as SqlJsDatabase } from 'sql.js';
|
||||
import type { Platform } from 'waterflow-ts/dist/platform/types.js';
|
||||
import type { Entity, Relation, RecallParams, CommitParams, PurgeParams, RecallResult, CommitResult, PurgeResult, MemoryStats } from './types';
|
||||
import { getConfig } from './config';
|
||||
|
||||
export class GraphDatabase {
|
||||
private db: SqlJsDatabase | null = null;
|
||||
private sessionId: string;
|
||||
private initPromise: Promise<void> | null = null;
|
||||
private _platform: Platform | null = null;
|
||||
|
||||
constructor(sessionId?: string) {
|
||||
this.sessionId = sessionId || `session-${Date.now()}`;
|
||||
this.initPromise = this.initDatabase();
|
||||
}
|
||||
|
||||
private async initDatabase(): Promise<void> {
|
||||
const SQL = await initSqlJs();
|
||||
const { getPlatform } = await import('waterflow-ts/dist/platform/index.js');
|
||||
this._platform = getPlatform();
|
||||
|
||||
try {
|
||||
const dbPath = this._platform.path.join(this._platform.getCwd(), getConfig().dbPath);
|
||||
const fs = this._platform.fs;
|
||||
if (fs) {
|
||||
const exists = await fs.exists(dbPath);
|
||||
if (exists) {
|
||||
const data = await fs.readFile(dbPath, { encoding: 'binary' });
|
||||
this.db = new SQL.Database(new Uint8Array(data as ArrayBuffer));
|
||||
} else {
|
||||
this.db = new SQL.Database();
|
||||
}
|
||||
} else {
|
||||
this.db = new SQL.Database();
|
||||
}
|
||||
} catch {
|
||||
this.db = new SQL.Database();
|
||||
}
|
||||
|
||||
this.createTables();
|
||||
}
|
||||
|
||||
async save(): Promise<void> {
|
||||
if (!this.db || !this._platform) return;
|
||||
|
||||
try {
|
||||
const platform = this._platform;
|
||||
const fs = platform.fs;
|
||||
if (!fs) return;
|
||||
const dbPath = platform.path.join(platform.getCwd(), getConfig().dbPath);
|
||||
const dir = platform.path.dirname(dbPath);
|
||||
const dirExists = await fs.exists(dir);
|
||||
if (!dirExists) {
|
||||
await fs.mkdir(dir, true);
|
||||
}
|
||||
const data = this.db.export();
|
||||
// sql.js returns Uint8Array, convert to Buffer for writeFile compatibility
|
||||
await fs.writeFile(dbPath, Buffer.from(data), { encoding: 'binary' });
|
||||
} catch (error) {
|
||||
console.error(`[GraphDatabase] Save failed: ${error}`);
|
||||
}
|
||||
}
|
||||
|
||||
private createTables(): void {
|
||||
if (!this.db) return;
|
||||
this.db.run(`
|
||||
CREATE TABLE IF NOT EXISTS entities (
|
||||
id TEXT PRIMARY KEY,
|
||||
name TEXT NOT NULL,
|
||||
type TEXT DEFAULT 'unknown',
|
||||
mention_count INTEGER DEFAULT 1,
|
||||
created_at TEXT NOT NULL,
|
||||
updated_at TEXT NOT NULL
|
||||
)
|
||||
`);
|
||||
this.db.run(`
|
||||
CREATE TABLE IF NOT EXISTS relations (
|
||||
id TEXT PRIMARY KEY,
|
||||
source_id TEXT NOT NULL,
|
||||
target_id TEXT NOT NULL,
|
||||
relation_type TEXT NOT NULL,
|
||||
confidence REAL DEFAULT 1.0,
|
||||
status TEXT DEFAULT 'active',
|
||||
session_id TEXT NOT NULL,
|
||||
turn_id INTEGER DEFAULT 0,
|
||||
created_at TEXT NOT NULL,
|
||||
updated_at TEXT NOT NULL,
|
||||
date_bucket TEXT NOT NULL,
|
||||
FOREIGN KEY (source_id) REFERENCES entities(id),
|
||||
FOREIGN KEY (target_id) REFERENCES entities(id)
|
||||
)
|
||||
`);
|
||||
this.db.run(`CREATE INDEX IF NOT EXISTS idx_entity_name ON entities(name)`);
|
||||
this.db.run(`CREATE INDEX IF NOT EXISTS idx_entity_type ON entities(type)`);
|
||||
this.db.run(`CREATE INDEX IF NOT EXISTS idx_relation_source ON relations(source_id)`);
|
||||
this.db.run(`CREATE INDEX IF NOT EXISTS idx_relation_target ON relations(target_id)`);
|
||||
this.db.run(`CREATE INDEX IF NOT EXISTS idx_relation_type ON relations(relation_type)`);
|
||||
this.db.run(`CREATE INDEX IF NOT EXISTS idx_relation_status ON relations(status)`);
|
||||
}
|
||||
|
||||
private ensureInit(): void {
|
||||
if (!this.db) {
|
||||
throw new Error('Database not initialized');
|
||||
}
|
||||
}
|
||||
|
||||
async recall(params: RecallParams): Promise<RecallResult> {
|
||||
await this.initPromise;
|
||||
this.ensureInit();
|
||||
const { queryIntent, seedEntities, depth = 2, sessionFilter } = params;
|
||||
const keywords = queryIntent.split(/[,\s]+/).filter(k => k.length > 0);
|
||||
const entities: Entity[] = [];
|
||||
const relations: Relation[] = [];
|
||||
const entityIds = new Set<string>();
|
||||
|
||||
if (!this.db) return { entities, relations, message: 'Database not ready' };
|
||||
|
||||
if (!keywords.length && !seedEntities?.length) {
|
||||
const rows = this.db.exec('SELECT * FROM entities ORDER BY mention_count DESC LIMIT 50');
|
||||
if (rows.length > 0) {
|
||||
const columns = rows[0].columns;
|
||||
for (const row of rows[0].values) {
|
||||
const obj = this.rowToObject(columns, row);
|
||||
const id = obj.id as string;
|
||||
entityIds.add(id);
|
||||
entities.push(this.rowToEntity(obj, 0));
|
||||
}
|
||||
}
|
||||
} else {
|
||||
for (const keyword of keywords) {
|
||||
const stmt = this.db.prepare('SELECT * FROM entities WHERE LOWER(name) LIKE ? LIMIT 100');
|
||||
stmt.bind([`%${keyword.toLowerCase()}%`]);
|
||||
while (stmt.step()) {
|
||||
const row = stmt.getAsObject();
|
||||
const id = row.id as string;
|
||||
if (!entityIds.has(id)) {
|
||||
entityIds.add(id);
|
||||
entities.push(this.rowToEntity(row, 0));
|
||||
}
|
||||
}
|
||||
stmt.free();
|
||||
}
|
||||
}
|
||||
|
||||
if (seedEntities?.length) {
|
||||
for (const seedName of seedEntities) {
|
||||
const stmt = this.db.prepare('SELECT * FROM entities WHERE LOWER(name) = ? LIMIT 1');
|
||||
stmt.bind([seedName.toLowerCase()]);
|
||||
if (stmt.step()) {
|
||||
const row = stmt.getAsObject();
|
||||
const id = row.id as string;
|
||||
if (!entityIds.has(id)) {
|
||||
entityIds.add(id);
|
||||
entities.push(this.rowToEntity(row, 0));
|
||||
}
|
||||
}
|
||||
stmt.free();
|
||||
}
|
||||
}
|
||||
|
||||
this.bfsExpand(entityIds, entities, relations, depth, sessionFilter);
|
||||
|
||||
for (const entity of entities) {
|
||||
if (entity.depth === undefined) {
|
||||
entity.depth = 0;
|
||||
}
|
||||
}
|
||||
|
||||
return {
|
||||
entities,
|
||||
relations,
|
||||
message: `找到 ${entities.length} 个实体, ${relations.length} 条关系`
|
||||
};
|
||||
}
|
||||
|
||||
private bfsExpand(
|
||||
seedIds: Set<string>,
|
||||
entities: Entity[],
|
||||
relations: Relation[],
|
||||
maxDepth: number,
|
||||
sessionFilter?: string
|
||||
): void {
|
||||
if (!this.db) return;
|
||||
|
||||
const visited = new Set(seedIds);
|
||||
let currentLayer = new Set(seedIds);
|
||||
const entityDepths: Record<string, number> = {};
|
||||
for (const id of seedIds) {
|
||||
entityDepths[id] = 0;
|
||||
}
|
||||
|
||||
for (let layer = 0; layer < maxDepth; layer++) {
|
||||
if (!currentLayer.size) break;
|
||||
|
||||
const placeholders = Array(currentLayer.size).fill('?').join(',');
|
||||
let sql = `
|
||||
SELECT r.id, r.source_id, r.target_id,
|
||||
e1.name as source_name, e2.name as target_name,
|
||||
r.relation_type, r.confidence, r.session_id,
|
||||
r.turn_id, r.created_at, r.updated_at, r.status, r.date_bucket
|
||||
FROM relations r
|
||||
JOIN entities e1 ON r.source_id = e1.id
|
||||
JOIN entities e2 ON r.target_id = e2.id
|
||||
WHERE (r.source_id IN (${placeholders}) OR r.target_id IN (${placeholders}))
|
||||
AND r.status = 'active'
|
||||
`;
|
||||
const params: (string | number)[] = [];
|
||||
for (const id of currentLayer) { params.push(id); }
|
||||
for (const id of currentLayer) { params.push(id); }
|
||||
if (sessionFilter) {
|
||||
sql += ` AND r.session_id = ?`;
|
||||
params.push(sessionFilter);
|
||||
}
|
||||
|
||||
const stmt = this.db.prepare(sql);
|
||||
stmt.bind(params);
|
||||
|
||||
const nextLayer = new Set<string>();
|
||||
const layerRelations: Relation[] = [];
|
||||
|
||||
while (stmt.step()) {
|
||||
const row = stmt.getAsObject();
|
||||
const sourceId = row.source_id as string;
|
||||
const targetId = row.target_id as string;
|
||||
const sourceDepth = entityDepths[sourceId] ?? layer;
|
||||
const targetDepth = entityDepths[targetId] ?? layer;
|
||||
const relationDepth = Math.max(sourceDepth, targetDepth) + 1;
|
||||
|
||||
layerRelations.push({
|
||||
id: row.id as string,
|
||||
sourceId,
|
||||
targetId,
|
||||
relationType: row.relation_type as string,
|
||||
confidence: row.confidence as number,
|
||||
status: row.status as Relation['status'],
|
||||
sessionId: row.session_id as string,
|
||||
turnId: row.turn_id as number,
|
||||
createdAt: new Date(row.created_at as string),
|
||||
updatedAt: new Date(row.updated_at as string),
|
||||
dateBucket: row.date_bucket as string,
|
||||
depth: relationDepth
|
||||
});
|
||||
|
||||
if (!visited.has(sourceId)) {
|
||||
visited.add(sourceId);
|
||||
nextLayer.add(sourceId);
|
||||
entityDepths[sourceId] = layer + 1;
|
||||
}
|
||||
if (!visited.has(targetId)) {
|
||||
visited.add(targetId);
|
||||
nextLayer.add(targetId);
|
||||
entityDepths[targetId] = layer + 1;
|
||||
}
|
||||
}
|
||||
stmt.free();
|
||||
|
||||
relations.push(...layerRelations);
|
||||
|
||||
if (nextLayer.size) {
|
||||
const placeholders = Array(nextLayer.size).fill('?').join(',');
|
||||
const entityStmt = this.db.prepare(
|
||||
`SELECT * FROM entities WHERE id IN (${placeholders})`
|
||||
);
|
||||
entityStmt.bind(Array.from(nextLayer));
|
||||
while (entityStmt.step()) {
|
||||
const row = entityStmt.getAsObject();
|
||||
const id = row.id as string;
|
||||
entities.push(this.rowToEntity(row, entityDepths[id] ?? layer + 1));
|
||||
}
|
||||
entityStmt.free();
|
||||
}
|
||||
|
||||
currentLayer = nextLayer;
|
||||
}
|
||||
}
|
||||
|
||||
async commit(params: CommitParams): Promise<CommitResult> {
|
||||
await this.initPromise;
|
||||
this.ensureInit();
|
||||
const { triplets, sessionId, turnId } = params;
|
||||
let createdEntities = 0;
|
||||
let createdRelations = 0;
|
||||
|
||||
if (!this.db) return { createdEntities: 0, createdRelations: 0 };
|
||||
|
||||
for (const triplet of triplets) {
|
||||
const sourceId = this.upsertEntity(triplet.subject);
|
||||
const targetId = this.upsertEntity(triplet.object);
|
||||
|
||||
const relationId = this.generateId();
|
||||
const now = new Date().toISOString();
|
||||
this.db.run(
|
||||
`INSERT INTO relations (id, source_id, target_id, relation_type, confidence, status, session_id, turn_id, created_at, updated_at, date_bucket)
|
||||
VALUES (?, ?, ?, ?, ?, 'active', ?, ?, ?, ?, ?)`,
|
||||
[relationId, sourceId, targetId, triplet.relation, triplet.confidence ?? 1.0, sessionId ?? this.sessionId, turnId ?? 0, now, now, new Date().toISOString().split('T')[0]]
|
||||
);
|
||||
|
||||
createdEntities += 2;
|
||||
createdRelations++;
|
||||
}
|
||||
|
||||
if (getConfig().autoSave) {
|
||||
await this.save();
|
||||
}
|
||||
|
||||
return { createdEntities, createdRelations };
|
||||
}
|
||||
|
||||
async purge(params: PurgeParams): Promise<PurgeResult> {
|
||||
await this.initPromise;
|
||||
this.ensureInit();
|
||||
const { criteria, mode = 'soft' } = params;
|
||||
let deleted = 0;
|
||||
|
||||
if (!this.db) return { deleted: 0, mode };
|
||||
|
||||
const conditions: string[] = [];
|
||||
const queryParams: (string | number)[] = [];
|
||||
|
||||
if (criteria?.subject) {
|
||||
const stmt = this.db.prepare('SELECT id FROM entities WHERE LOWER(name) = ?');
|
||||
stmt.bind([criteria.subject.toLowerCase()]);
|
||||
if (stmt.step()) {
|
||||
const row = stmt.getAsObject();
|
||||
conditions.push(`source_id = ?`);
|
||||
queryParams.push(row.id as string);
|
||||
}
|
||||
stmt.free();
|
||||
}
|
||||
|
||||
if (criteria?.target) {
|
||||
const stmt = this.db.prepare('SELECT id FROM entities WHERE LOWER(name) = ?');
|
||||
stmt.bind([criteria.target.toLowerCase()]);
|
||||
if (stmt.step()) {
|
||||
const row = stmt.getAsObject();
|
||||
conditions.push(`target_id = ?`);
|
||||
queryParams.push(row.id as string);
|
||||
}
|
||||
stmt.free();
|
||||
}
|
||||
|
||||
if (criteria?.relation) {
|
||||
conditions.push(`relation_type = ?`);
|
||||
queryParams.push(criteria.relation);
|
||||
}
|
||||
|
||||
if (!conditions.length) {
|
||||
return { deleted: 0, mode, message: '无删除条件' };
|
||||
}
|
||||
|
||||
const whereClause = conditions.join(' AND ');
|
||||
|
||||
const countStmt = this.db.prepare(`SELECT COUNT(*) as cnt FROM relations WHERE ${whereClause} AND status = 'active'`);
|
||||
countStmt.bind(queryParams);
|
||||
if (countStmt.step()) {
|
||||
const row = countStmt.getAsObject();
|
||||
deleted = row.cnt as number;
|
||||
}
|
||||
countStmt.free();
|
||||
|
||||
if (mode === 'hard') {
|
||||
this.db.run(`DELETE FROM relations WHERE ${whereClause} AND status = 'active'`, queryParams);
|
||||
} else {
|
||||
this.db.run(
|
||||
`UPDATE relations SET status = 'deleted', updated_at = ? WHERE ${whereClause} AND status = 'active'`,
|
||||
[new Date().toISOString(), ...queryParams]
|
||||
);
|
||||
}
|
||||
|
||||
if (getConfig().autoSave && deleted > 0) {
|
||||
await this.save();
|
||||
}
|
||||
|
||||
return { deleted, mode };
|
||||
}
|
||||
|
||||
async introspect(): Promise<MemoryStats> {
|
||||
await this.initPromise;
|
||||
this.ensureInit();
|
||||
|
||||
if (!this.db) return { entityCount: 0, relationCount: 0, sessionId: this.sessionId };
|
||||
|
||||
const entityCount = (this.db.exec('SELECT COUNT(*) FROM entities')[0]?.values[0]?.[0] as number) ?? 0;
|
||||
const relationCount = (this.db.exec("SELECT COUNT(*) FROM relations WHERE status = 'active'")[0]?.values[0]?.[0] as number) ?? 0;
|
||||
|
||||
return { entityCount, relationCount, sessionId: this.sessionId };
|
||||
}
|
||||
|
||||
private upsertEntity(name: string): string {
|
||||
if (!this.db) return this.generateId();
|
||||
|
||||
const existingStmt = this.db.prepare('SELECT id, mention_count FROM entities WHERE LOWER(name) = ?');
|
||||
existingStmt.bind([name.toLowerCase()]);
|
||||
if (existingStmt.step()) {
|
||||
const row = existingStmt.getAsObject();
|
||||
const id = row.id as string;
|
||||
this.db.run('UPDATE entities SET mention_count = ?, updated_at = ? WHERE id = ?', [(row.mention_count as number) + 1, new Date().toISOString(), id]);
|
||||
existingStmt.free();
|
||||
return id;
|
||||
}
|
||||
existingStmt.free();
|
||||
|
||||
const id = this.generateId();
|
||||
const now = new Date().toISOString();
|
||||
this.db.run(
|
||||
'INSERT INTO entities (id, name, type, mention_count, created_at, updated_at) VALUES (?, ?, ?, 1, ?, ?)',
|
||||
[id, name, 'unknown', now, now]
|
||||
);
|
||||
return id;
|
||||
}
|
||||
|
||||
private rowToEntity(row: Record<string, unknown>, depth?: number): Entity {
|
||||
const entity: Entity = {
|
||||
id: row.id as string,
|
||||
name: row.name as string,
|
||||
type: (row.type as string) ?? 'unknown',
|
||||
mentionCount: row.mention_count as number,
|
||||
createdAt: new Date(row.created_at as string),
|
||||
updatedAt: new Date(row.updated_at as string)
|
||||
};
|
||||
if (depth !== undefined) {
|
||||
entity.depth = depth;
|
||||
}
|
||||
return entity;
|
||||
}
|
||||
|
||||
private rowToObject(columns: string[], values: unknown[]): Record<string, unknown> {
|
||||
const obj: Record<string, unknown> = {};
|
||||
columns.forEach((col, i) => { obj[col] = values[i]; });
|
||||
return obj;
|
||||
}
|
||||
|
||||
private generateId(): string {
|
||||
if (this._platform?.globals?.randomUUID) {
|
||||
return this._platform.globals.randomUUID();
|
||||
}
|
||||
return `ent-${Date.now()}-${Math.random().toString(36).slice(2, 9)}`;
|
||||
}
|
||||
|
||||
setSessionId(sessionId: string): void {
|
||||
this.sessionId = sessionId;
|
||||
}
|
||||
|
||||
getSessionId(): string {
|
||||
return this.sessionId;
|
||||
}
|
||||
|
||||
close(): void {
|
||||
if (this.db) {
|
||||
this.db.close();
|
||||
this.db = null;
|
||||
}
|
||||
}
|
||||
}
|
||||
4
ts/src/runtime/core/graph_memory/index.ts
Normal file
4
ts/src/runtime/core/graph_memory/index.ts
Normal file
@ -0,0 +1,4 @@
|
||||
export * from './types';
|
||||
export * from './config';
|
||||
export * from './graph_database';
|
||||
export * from './memory_service';
|
||||
129
ts/src/runtime/core/graph_memory/memory_service.ts
Normal file
129
ts/src/runtime/core/graph_memory/memory_service.ts
Normal file
@ -0,0 +1,129 @@
|
||||
import { GraphDatabase } from './graph_database';
|
||||
import type { RecallParams, CommitParams, PurgeParams, RecallResult, CommitResult, PurgeResult, MemoryStats } from './types';
|
||||
|
||||
export class MemoryService {
|
||||
private db: GraphDatabase;
|
||||
|
||||
constructor(db: GraphDatabase) {
|
||||
this.db = db;
|
||||
}
|
||||
|
||||
async recall(params: RecallParams): Promise<RecallResult> {
|
||||
return this.db.recall(params);
|
||||
}
|
||||
|
||||
async commit(params: CommitParams): Promise<CommitResult> {
|
||||
return this.db.commit(params);
|
||||
}
|
||||
|
||||
async purge(params: PurgeParams): Promise<PurgeResult> {
|
||||
return this.db.purge(params);
|
||||
}
|
||||
|
||||
async introspect(): Promise<MemoryStats> {
|
||||
return this.db.introspect();
|
||||
}
|
||||
|
||||
async updatePersona(params: { attributes: Array<{ attribute: string; value: string }>; mode?: 'merge' | 'replace' }): Promise<{ status: string; updatedAttributes: number }> {
|
||||
const { attributes, mode = 'merge' } = params;
|
||||
|
||||
if (mode === 'replace') {
|
||||
await this.db.purge({
|
||||
criteria: { subject: 'AI' },
|
||||
mode: 'soft'
|
||||
});
|
||||
}
|
||||
|
||||
const triplets = attributes.map(attr => ({
|
||||
subject: 'AI',
|
||||
relation: attr.attribute,
|
||||
object: attr.value,
|
||||
confidence: 1.0
|
||||
}));
|
||||
|
||||
await this.db.commit({ triplets });
|
||||
|
||||
return { status: 'success', updatedAttributes: attributes.length };
|
||||
}
|
||||
|
||||
async clearPersona(params: { confirm: boolean }): Promise<{ status: string; deletedCount: number }> {
|
||||
if (params.confirm === false) {
|
||||
return { status: 'cancelled', deletedCount: 0 };
|
||||
}
|
||||
|
||||
const result = await this.db.purge({
|
||||
criteria: { subject: 'AI' },
|
||||
mode: 'soft'
|
||||
});
|
||||
|
||||
return { status: 'success', deletedCount: result.deleted };
|
||||
}
|
||||
|
||||
async createTask(params: { task_id: string; description: string; info_nodes?: string[] | undefined }): Promise<{ status: string; taskId: string }> {
|
||||
const { task_id, description, info_nodes = [] } = params;
|
||||
|
||||
await this.db.commit({
|
||||
triplets: [
|
||||
{ subject: task_id, relation: 'is_type', object: 'TaskNode' },
|
||||
{ subject: task_id, relation: 'has_description', object: description },
|
||||
{ subject: task_id, relation: 'HAS_STATE', object: 'State_进行中' }
|
||||
]
|
||||
});
|
||||
|
||||
if (info_nodes.length > 0) {
|
||||
await this.db.commit({
|
||||
triplets: info_nodes.map(node => ({
|
||||
subject: task_id,
|
||||
relation: 'CONTAINS_INFO',
|
||||
object: node
|
||||
}))
|
||||
});
|
||||
}
|
||||
|
||||
return { status: 'success', taskId: task_id };
|
||||
}
|
||||
|
||||
async setTaskState(params: { task_id: string; state: string }): Promise<{ status: string; newState: string }> {
|
||||
const { task_id, state } = params;
|
||||
|
||||
await this.db.purge({
|
||||
criteria: { subject: task_id, relation: 'HAS_STATE' },
|
||||
mode: 'soft'
|
||||
});
|
||||
|
||||
await this.db.commit({
|
||||
triplets: [{ subject: task_id, relation: 'HAS_STATE', object: `State_${state}` }]
|
||||
});
|
||||
|
||||
return { status: 'success', newState: state };
|
||||
}
|
||||
|
||||
async deleteTask(params: { task_id: string }): Promise<{ status: string; taskId: string }> {
|
||||
const { task_id } = params;
|
||||
|
||||
await this.db.purge({
|
||||
criteria: { subject: task_id },
|
||||
mode: 'soft'
|
||||
});
|
||||
|
||||
return { status: 'success', taskId: task_id };
|
||||
}
|
||||
|
||||
async linkInfoToTask(params: { task_id: string; info_node: string }): Promise<{ status: string }> {
|
||||
const { task_id, info_node } = params;
|
||||
|
||||
await this.db.commit({
|
||||
triplets: [{ subject: task_id, relation: 'CONTAINS_INFO', object: info_node }]
|
||||
});
|
||||
|
||||
return { status: 'success' };
|
||||
}
|
||||
|
||||
setSessionId(sessionId: string): void {
|
||||
this.db.setSessionId(sessionId);
|
||||
}
|
||||
|
||||
getSessionId(): string {
|
||||
return this.db.getSessionId();
|
||||
}
|
||||
}
|
||||
128
ts/src/runtime/core/graph_memory/types.ts
Normal file
128
ts/src/runtime/core/graph_memory/types.ts
Normal file
@ -0,0 +1,128 @@
|
||||
export interface Entity {
|
||||
id: string;
|
||||
name: string;
|
||||
type: string;
|
||||
mentionCount: number;
|
||||
createdAt: Date;
|
||||
updatedAt: Date;
|
||||
depth?: number; // BFS 搜索深度标注
|
||||
}
|
||||
|
||||
export type RelationStatus = 'active' | 'deleted' | 'archived' | 'superseded';
|
||||
|
||||
export interface Relation {
|
||||
id: string;
|
||||
sourceId: string;
|
||||
targetId: string;
|
||||
relationType: string;
|
||||
confidence: number;
|
||||
status: RelationStatus;
|
||||
sessionId: string;
|
||||
turnId: number;
|
||||
createdAt: Date;
|
||||
updatedAt: Date;
|
||||
dateBucket: string;
|
||||
depth?: number; // BFS 搜索深度标注
|
||||
}
|
||||
|
||||
export interface Triplet {
|
||||
subject: string;
|
||||
relation: string;
|
||||
object: string;
|
||||
confidence?: number;
|
||||
}
|
||||
|
||||
export interface RecallParams {
|
||||
queryIntent: string;
|
||||
seedEntities?: string[] | undefined;
|
||||
depth?: number | undefined;
|
||||
timeRange?: { days: number } | undefined;
|
||||
sessionFilter?: string | undefined;
|
||||
}
|
||||
|
||||
export interface CommitParams {
|
||||
triplets: Triplet[];
|
||||
entityTypes?: Record<string, string> | undefined;
|
||||
temporalTag?: string | undefined;
|
||||
sessionId?: string | undefined;
|
||||
turnId?: number | undefined;
|
||||
}
|
||||
|
||||
export interface PurgeParams {
|
||||
criteria?: {
|
||||
subject?: string | undefined;
|
||||
target?: string | undefined;
|
||||
relation?: string | undefined;
|
||||
sessionId?: string | undefined;
|
||||
} | undefined;
|
||||
mode?: 'soft' | 'hard' | 'supersede' | undefined;
|
||||
newRelation?: { relation: string; target: string } | undefined;
|
||||
}
|
||||
|
||||
export interface RecallResult {
|
||||
entities: Entity[];
|
||||
relations: Relation[];
|
||||
message: string;
|
||||
}
|
||||
|
||||
export interface CommitResult {
|
||||
createdEntities: number;
|
||||
createdRelations: number;
|
||||
}
|
||||
|
||||
export interface PurgeResult {
|
||||
deleted: number;
|
||||
mode: string;
|
||||
message?: string;
|
||||
}
|
||||
|
||||
export type TaskState = '进行中' | '已完成' | '已暂停' | '已取消';
|
||||
|
||||
export interface Task {
|
||||
taskId: string;
|
||||
description: string;
|
||||
state: TaskState;
|
||||
infoNodes: string[];
|
||||
createdAt: Date;
|
||||
updatedAt: Date;
|
||||
}
|
||||
|
||||
export interface MemoryStats {
|
||||
entityCount: number;
|
||||
relationCount: number;
|
||||
sessionId?: string;
|
||||
}
|
||||
|
||||
export interface PersonaAttribute {
|
||||
attribute: string;
|
||||
value: string;
|
||||
}
|
||||
|
||||
export interface PersonaUpdateParams {
|
||||
attributes: PersonaAttribute[];
|
||||
mode?: 'merge' | 'replace';
|
||||
}
|
||||
|
||||
export interface PersonaClearParams {
|
||||
confirm: boolean;
|
||||
}
|
||||
|
||||
export interface TaskCreateParams {
|
||||
task_id: string;
|
||||
description: string;
|
||||
info_nodes?: string[] | undefined;
|
||||
}
|
||||
|
||||
export interface TaskSetStateParams {
|
||||
task_id: string;
|
||||
state: string;
|
||||
}
|
||||
|
||||
export interface TaskDeleteParams {
|
||||
task_id: string;
|
||||
}
|
||||
|
||||
export interface TaskLinkInfoParams {
|
||||
task_id: string;
|
||||
info_node: string;
|
||||
}
|
||||
250
ts/src/runtime/core/tools/builtin/graph_memory_tool.ts
Normal file
250
ts/src/runtime/core/tools/builtin/graph_memory_tool.ts
Normal file
@ -0,0 +1,250 @@
|
||||
import type { Tool, ToolCategory, PermissionLevel, ToolInputSchema, ToolOutput, ToolInput, ToolExecutionContext } from 'waterflow-ts/dist/runtime/core/tools/tool_interface.js';
|
||||
import { GraphDatabase } from '../../graph_memory/graph_database';
|
||||
import { MemoryService } from '../../graph_memory/memory_service';
|
||||
|
||||
const GRAPH_MEMORY_TOOL_ID = 'builtin:graph_memory';
|
||||
const GRAPH_MEMORY_TOOL_API_NAME = 'graph_memory'; // API 兼容名称（不含冒号，符合 ^[a-zA-Z0-9_-]+$ 要求）
|
||||
|
||||
/**
|
||||
* 工具名称映射工具 - 用于处理 API 对工具名称格式的限制
|
||||
* OpenAI/DeepSeek API 要求工具名称符合 ^[a-zA-Z0-9_-]+$ 正则表达式
|
||||
* 而 TrulyMEM 的内部 ID 使用 "builtin:xxx" 格式（含冒号）
|
||||
*/
|
||||
|
||||
/**
|
||||
* 将内部工具 ID 映射为 API 兼容名称
|
||||
* @param toolId 内部工具 ID，如 "builtin:graph_memory"
|
||||
* @returns API 兼容名称，如 "graph_memory"
|
||||
*/
|
||||
export function mapToolIdToApiName(toolId: string): string {
|
||||
// 移除 "builtin:" 前缀
|
||||
if (toolId.startsWith('builtin:')) {
|
||||
return toolId.slice(8);
|
||||
}
|
||||
// 其他前缀也移除（如 "mcp:", "plugin:"）
|
||||
const colonIndex = toolId.indexOf(':');
|
||||
if (colonIndex > 0) {
|
||||
return toolId.slice(colonIndex + 1);
|
||||
}
|
||||
return toolId;
|
||||
}
|
||||
|
||||
/**
|
||||
* 将 API 返回的工具名称映射回内部 ID
|
||||
* @param apiName API 返回的工具名称，如 "graph_memory"
|
||||
* @param prefix 内部 ID 前缀，默认 "builtin:"
|
||||
* @returns 内部工具 ID，如 "builtin:graph_memory"
|
||||
*/
|
||||
export function mapApiNameToToolId(apiName: string, prefix = 'builtin:'): string {
|
||||
// 如果已经是完整 ID 格式，直接返回
|
||||
if (apiName.includes(':')) {
|
||||
return apiName;
|
||||
}
|
||||
return `${prefix}${apiName}`;
|
||||
}
|
||||
|
||||
const GRAPH_MEMORY_TOOL_DESCRIPTION = `图记忆工具 - 让 AI 拥有真正的长期记忆能力
|
||||
|
||||
操作:
|
||||
- recall: 检索记忆 - 提供 queryIntent (搜索意图) 和可选的 seedEntities
|
||||
- commit: 写入记忆 - 必须使用 triplets 数组格式，每个三元组包含 subject, relation, object
|
||||
- purge: 删除记忆 - 提供 criteria 指定删除条件
|
||||
- introspect: 查看状态 - 无参数
|
||||
- persona_update/clear: 人设管理
|
||||
- task_create/set_state/delete: 任务管理
|
||||
|
||||
【commit 操作的三元组格式】
|
||||
triplets 必须是数组，每个元素是 {subject, relation, object, confidence} 格式:
|
||||
示例: {"action":"commit","params":{"triplets":[{"subject":"Alice","relation":"is a","object":"engineer","confidence":0.95}]}}
|
||||
- subject: 实体名称 (如用户名、技术名称)
|
||||
- relation: 关系描述 (如 "is a", "likes", "knows")
|
||||
- object: 目标实体 (如职业、爱好、技术)
|
||||
- confidence: 置信度 0-1 (可选，默认0.9)
|
||||
|
||||
【recall 操作】
|
||||
示例: {"action":"recall","params":{"queryIntent":"用户的学习偏好","seedEntities":["Alice"]}}`;
|
||||
|
||||
export class GraphMemoryTool implements Tool {
|
||||
readonly id = GRAPH_MEMORY_TOOL_ID;
|
||||
readonly name = 'GraphMemory';
|
||||
readonly apiName = GRAPH_MEMORY_TOOL_API_NAME;
|
||||
readonly description = GRAPH_MEMORY_TOOL_DESCRIPTION;
|
||||
readonly category: ToolCategory = 'analysis';
|
||||
readonly permissionLevel: PermissionLevel = 'safe';
|
||||
readonly alwaysLoad = true;
|
||||
|
||||
readonly inputSchema: ToolInputSchema = {
|
||||
type: 'object',
|
||||
properties: {
|
||||
action: {
|
||||
type: 'string',
|
||||
enum: [
|
||||
'recall', 'commit', 'purge', 'introspect',
|
||||
'persona_update', 'persona_clear',
|
||||
'task_create', 'task_set_state', 'task_delete', 'task_link_info'
|
||||
],
|
||||
description: '记忆操作类型'
|
||||
},
|
||||
params: {
|
||||
type: 'object',
|
||||
description: '操作参数',
|
||||
properties: {
|
||||
queryIntent: { type: 'string', description: '搜索意图' },
|
||||
seedEntities: { type: 'array', items: { type: 'string', description: '实体' }, description: '种子实体' },
|
||||
depth: { type: 'number', description: '检索深度' },
|
||||
sessionFilter: { type: 'string', description: '会话ID过滤' },
|
||||
triplets: {
|
||||
type: 'array',
|
||||
description: '知识三元组数组。每个三元组描述 subject-relation-object 关系，用于存储记忆知识。',
|
||||
items: {
|
||||
type: 'object',
|
||||
description: '三元组: {subject, relation, object, confidence} - subject/relation/object 必填',
|
||||
properties: {
|
||||
subject: { type: 'string', description: '主体实体，如人名、技术名称等' },
|
||||
relation: { type: 'string', description: '关系描述，如 "is a", "likes", "knows", "uses" 等' },
|
||||
object: { type: 'string', description: '客体实体，如职业、爱好、技术名称等' },
|
||||
confidence: { type: 'number', description: '置信度 (0-1)，默认 0.9', default: 0.9 }
|
||||
}
|
||||
}
|
||||
},
|
||||
sessionId: { type: 'string', description: '会话ID' },
|
||||
turnId: { type: 'number', description: '轮次ID' },
|
||||
criteria: {
|
||||
type: 'object',
|
||||
properties: {
|
||||
subject: { type: 'string', description: '主体' },
|
||||
target: { type: 'string', description: '客体' },
|
||||
relation: { type: 'string', description: '关系' },
|
||||
sessionId: { type: 'string', description: '会话ID' }
|
||||
},
|
||||
description: '删除条件'
|
||||
},
|
||||
mode: { type: 'string', enum: ['soft', 'hard', 'supersede'], description: '删除模式' },
|
||||
attributes: {
|
||||
type: 'array',
|
||||
items: {
|
||||
type: 'object',
|
||||
description: '属性',
|
||||
properties: {
|
||||
attribute: { type: 'string', description: '属性名' },
|
||||
value: { type: 'string', description: '属性值' }
|
||||
}
|
||||
},
|
||||
description: '属性数组'
|
||||
},
|
||||
confirm: { type: 'boolean', description: '确认清除' },
|
||||
task_id: { type: 'string', description: '任务ID' },
|
||||
description: { type: 'string', description: '任务描述' },
|
||||
state: { type: 'string', description: '任务状态' },
|
||||
info_nodes: { type: 'array', items: { type: 'string', description: '节点' }, description: '信息节点' },
|
||||
info_node: { type: 'string', description: '信息节点' }
|
||||
}
|
||||
}
|
||||
},
|
||||
required: ['action', 'params']
|
||||
};
|
||||
|
||||
private db: GraphDatabase;
|
||||
private service: MemoryService;
|
||||
|
||||
constructor(sessionId?: string) {
|
||||
this.db = new GraphDatabase(sessionId);
|
||||
this.service = new MemoryService(this.db);
|
||||
}
|
||||
|
||||
async handler(params: ToolInput, context: ToolExecutionContext): Promise<ToolOutput> {
|
||||
if (context.abortController?.signal?.aborted) {
|
||||
throw new Error('Operation aborted');
|
||||
}
|
||||
|
||||
const action = params.action as string;
|
||||
const actionParams = params.params as Record<string, unknown>;
|
||||
const logger = context?.logger;
|
||||
|
||||
try {
|
||||
logger?.info(`[GraphMemoryTool] Executing action: ${action}`);
|
||||
const result = await this.executeAction(action, actionParams);
|
||||
logger?.info(`[GraphMemoryTool] Action ${action} completed successfully`);
|
||||
return JSON.stringify({ success: true, data: result }, null, 2);
|
||||
} catch (error) {
|
||||
logger?.error(`[GraphMemoryTool] Action ${action} failed:`, error);
|
||||
return JSON.stringify({
|
||||
success: false,
|
||||
error: {
|
||||
type: 'execution_error',
|
||||
message: error instanceof Error ? error.message : String(error)
|
||||
}
|
||||
}, null, 2);
|
||||
}
|
||||
}
|
||||
|
||||
private async executeAction(action: string, params: Record<string, unknown>): Promise<unknown> {
|
||||
switch (action) {
|
||||
case 'recall':
|
||||
return this.service.recall({
|
||||
queryIntent: params.queryIntent as string || '',
|
||||
seedEntities: params.seedEntities as string[] | undefined,
|
||||
depth: params.depth as number | undefined,
|
||||
sessionFilter: params.sessionFilter as string | undefined
|
||||
});
|
||||
|
||||
case 'commit':
|
||||
return this.service.commit({
|
||||
triplets: params.triplets as Array<{ subject: string; relation: string; object: string; confidence?: number }>,
|
||||
sessionId: params.sessionId as string | undefined,
|
||||
turnId: params.turnId as number | undefined
|
||||
});
|
||||
|
||||
case 'purge':
|
||||
return this.service.purge({
|
||||
criteria: params.criteria as { subject?: string | undefined; target?: string | undefined; relation?: string | undefined; sessionId?: string | undefined } | undefined,
|
||||
mode: params.mode as 'soft' | 'hard' | 'supersede' | undefined
|
||||
});
|
||||
|
||||
case 'introspect':
|
||||
return this.service.introspect();
|
||||
|
||||
case 'persona_update':
|
||||
return this.service.updatePersona({
|
||||
attributes: params.attributes as Array<{ attribute: string; value: string }>,
|
||||
mode: params.mode as 'merge' | 'replace'
|
||||
});
|
||||
|
||||
case 'persona_clear':
|
||||
return this.service.clearPersona({
|
||||
confirm: params.confirm as boolean
|
||||
});
|
||||
|
||||
case 'task_create':
|
||||
return this.service.createTask({
|
||||
task_id: params.task_id as string,
|
||||
description: params.description as string,
|
||||
info_nodes: params.info_nodes as string[] | undefined
|
||||
});
|
||||
|
||||
case 'task_set_state':
|
||||
return this.service.setTaskState({
|
||||
task_id: params.task_id as string,
|
||||
state: params.state as string
|
||||
});
|
||||
|
||||
case 'task_delete':
|
||||
return this.service.deleteTask({
|
||||
task_id: params.task_id as string
|
||||
});
|
||||
|
||||
case 'task_link_info':
|
||||
return this.service.linkInfoToTask({
|
||||
task_id: params.task_id as string,
|
||||
info_node: params.info_node as string
|
||||
});
|
||||
|
||||
default:
|
||||
throw new Error(`Unknown action: ${action}`);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
export function createGraphMemoryTool(sessionId?: string): GraphMemoryTool {
|
||||
return new GraphMemoryTool(sessionId);
|
||||
}
|
||||
20
ts/src/runtime/core/tools/builtin/index.ts
Normal file
20
ts/src/runtime/core/tools/builtin/index.ts
Normal file
@ -0,0 +1,20 @@
|
||||
import type { Tool } from 'waterflow-ts/dist/runtime/core/tools/tool_interface.js';
|
||||
import type { Platform } from 'waterflow-ts/dist/platform/types.js';
|
||||
import type { ToolRegistry } from 'waterflow-ts/dist/runtime/core/tools/tool_registry.js';
|
||||
import { GraphMemoryTool, createGraphMemoryTool, mapToolIdToApiName, mapApiNameToToolId } from './graph_memory_tool';
|
||||
|
||||
export { GraphMemoryTool, createGraphMemoryTool, mapToolIdToApiName, mapApiNameToToolId };
|
||||
|
||||
export function registerGraphMemoryTool(
|
||||
registry: { register: (tool: Tool) => void },
|
||||
sessionId?: string
|
||||
): void {
|
||||
registry.register(createGraphMemoryTool(sessionId));
|
||||
}
|
||||
|
||||
export async function installTrulyMEM(platform: Platform, sessionId?: string): Promise<ToolRegistry> {
|
||||
const { initializeToolRegistry } = await import('waterflow-ts/dist/runtime/core/tools/builtin/index.js');
|
||||
const registry = initializeToolRegistry(platform);
|
||||
registerGraphMemoryTool(registry, sessionId);
|
||||
return registry;
|
||||
}
|
||||
29
ts/src/types/sql.js.d.ts
vendored
Normal file
29
ts/src/types/sql.js.d.ts
vendored
Normal file
@ -0,0 +1,29 @@
|
||||
declare module 'sql.js' {
|
||||
export interface Database {
|
||||
run(sql: string, params?: (string | number | null | Uint8Array)[]): void;
|
||||
exec(sql: string): QueryExecResult[];
|
||||
prepare(sql: string): Statement;
|
||||
export(): Uint8Array;
|
||||
close(): void;
|
||||
}
|
||||
|
||||
export interface Statement {
|
||||
bind(params?: (string | number | null | Uint8Array)[]): boolean;
|
||||
step(): boolean;
|
||||
getAsObject(): Record<string, unknown>;
|
||||
free(): boolean;
|
||||
}
|
||||
|
||||
export interface QueryExecResult {
|
||||
columns: string[];
|
||||
values: (string | number | null | Uint8Array)[][];
|
||||
}
|
||||
|
||||
export interface SqlJsStatic {
|
||||
Database: new (data?: ArrayLike<number>) => Database;
|
||||
}
|
||||
|
||||
export default function initSqlJs(config?: {
|
||||
locateFile?: (file: string) => string;
|
||||
}): Promise<SqlJsStatic>;
|
||||
}
|
||||
637
ts/src/types/waterflow.d.ts
vendored
Normal file
637
ts/src/types/waterflow.d.ts
vendored
Normal file
@ -0,0 +1,637 @@
|
||||
declare module 'waterflow/platform' {
|
||||
import type { Platform, CreatePlatformOptions, PlatformCapabilities } from 'waterflow-ts/platform/types';
|
||||
export function getPlatform(): Platform;
|
||||
export function hasCapability(capability: keyof PlatformCapabilities): boolean;
|
||||
export function initPlatform(options?: CreatePlatformOptions): Platform;
|
||||
export function resetPlatform(): void;
|
||||
}
|
||||
|
||||
declare module 'waterflow/platform/types' {
|
||||
export type BufferSource = ArrayBuffer | ArrayBufferView;
|
||||
export type RuntimeType = 'node' | 'web' | 'harmony' | 'unknown';
|
||||
export type OSType = 'windows' | 'macos' | 'linux' | 'android' | 'ios' | 'harmony' | 'unknown';
|
||||
|
||||
export interface PlatformCapabilities {
|
||||
fileSystem: boolean;
|
||||
processExecution: boolean;
|
||||
network: boolean;
|
||||
storage: boolean;
|
||||
webSocket: boolean;
|
||||
workers: boolean;
|
||||
}
|
||||
|
||||
export interface PlatformAbortSignal {
|
||||
readonly aborted: boolean;
|
||||
readonly reason?: unknown;
|
||||
addEventListener(type: 'abort', listener: () => void): void;
|
||||
removeEventListener(type: 'abort', listener: () => void): void;
|
||||
}
|
||||
|
||||
export interface PlatformAbortController {
|
||||
readonly signal: PlatformAbortSignal;
|
||||
abort(reason?: unknown): void;
|
||||
}
|
||||
|
||||
export interface PlatformTextEncoder {
|
||||
encode(input?: string): Uint8Array;
|
||||
encodeInto(src: string, dest: Uint8Array): { read: number; written: number };
|
||||
}
|
||||
|
||||
export interface PlatformTextDecoder {
|
||||
decode(input?: BufferSource): string;
|
||||
}
|
||||
|
||||
export interface PlatformURL {
|
||||
href: string;
|
||||
origin: string;
|
||||
protocol: string;
|
||||
host: string;
|
||||
hostname: string;
|
||||
port: string;
|
||||
pathname: string;
|
||||
search: string;
|
||||
hash: string;
|
||||
toString(): string;
|
||||
toJSON(): string;
|
||||
}
|
||||
|
||||
export interface PlatformGlobals {
|
||||
TextEncoder: new (encoding?: string) => PlatformTextEncoder;
|
||||
TextDecoder: new (encoding?: string) => PlatformTextDecoder;
|
||||
URL: new (url: string) => { href: string; pathname: string; toString(): string };
|
||||
randomUUID(): string;
|
||||
now(): number;
|
||||
btoa(data: string): string;
|
||||
atob(data: string): string;
|
||||
}
|
||||
|
||||
export interface GlobOptions {
|
||||
cwd?: string;
|
||||
ignore?: string[];
|
||||
absolute?: boolean;
|
||||
dot?: boolean;
|
||||
onlyFiles?: boolean;
|
||||
onlyDirectories?: boolean;
|
||||
deep?: number;
|
||||
ignoreCase?: boolean;
|
||||
}
|
||||
|
||||
export interface GlobResult {
|
||||
path: string;
|
||||
isFile: boolean;
|
||||
isDirectory: boolean;
|
||||
}
|
||||
|
||||
export interface GlobTool {
|
||||
glob(pattern: string, options?: GlobOptions): Promise<string[]>;
|
||||
globWithInfo(pattern: string, options?: GlobOptions): Promise<GlobResult[]>;
|
||||
isMatch(path: string, pattern: string): boolean;
|
||||
search(pattern: string, basePath?: string): Promise<string[]>;
|
||||
}
|
||||
|
||||
export interface GrepOptions {
|
||||
cwd?: string;
|
||||
ignoreCase?: boolean;
|
||||
multiline?: boolean;
|
||||
glob?: string | string[];
|
||||
include?: string[];
|
||||
exclude?: string[];
|
||||
context?: number;
|
||||
beforeContext?: number;
|
||||
afterContext?: number;
|
||||
headLimit?: number;
|
||||
}
|
||||
|
||||
export interface GrepMatch {
|
||||
path: string;
|
||||
line: number;
|
||||
column?: number;
|
||||
content: string;
|
||||
}
|
||||
|
||||
export interface GrepSearchOptions {
|
||||
pattern: string;
|
||||
path?: string;
|
||||
glob?: string | string[];
|
||||
ignoreCase?: boolean;
|
||||
context?: number;
|
||||
outputMode?: 'content' | 'files_with_matches' | 'count';
|
||||
maxResults?: number;
|
||||
}
|
||||
|
||||
export interface GrepSearchResult {
|
||||
lines: string[];
|
||||
}
|
||||
|
||||
export interface GrepTool {
|
||||
grep(pattern: string | RegExp, options?: GrepOptions): Promise<GrepMatch[]>;
|
||||
grepFiles(pattern: string | RegExp, options?: GrepOptions): Promise<string[]>;
|
||||
grepCount(pattern: string | RegExp, options?: GrepOptions): Promise<number>;
|
||||
search(options: GrepSearchOptions): Promise<GrepSearchResult>;
|
||||
}
|
||||
|
||||
export interface PlatformTools {
|
||||
glob: GlobTool | null;
|
||||
grep: GrepTool | null;
|
||||
}
|
||||
|
||||
export interface FileInfo {
|
||||
path: string;
|
||||
name: string;
|
||||
isFile: boolean;
|
||||
isDirectory: boolean;
|
||||
size: number;
|
||||
modifiedTime?: number;
|
||||
createdTime?: number;
|
||||
}
|
||||
|
||||
export interface FileReadOptions {
|
||||
encoding?: 'utf-8' | 'binary' | 'base64';
|
||||
start?: number;
|
||||
end?: number;
|
||||
}
|
||||
|
||||
export interface FileWriteOptions {
|
||||
encoding?: 'utf-8' | 'binary' | 'base64';
|
||||
append?: boolean;
|
||||
createDir?: boolean;
|
||||
}
|
||||
|
||||
export interface FileSystemOperations {
|
||||
readFile(path: string, options?: FileReadOptions): Promise<string | ArrayBuffer>;
|
||||
writeFile(path: string, data: string | ArrayBuffer, options?: FileWriteOptions): Promise<void>;
|
||||
appendFile(path: string, data: string, options?: FileWriteOptions): Promise<void>;
|
||||
deleteFile(path: string): Promise<void>;
|
||||
exists(path: string): Promise<boolean>;
|
||||
stat(path: string): Promise<FileInfo>;
|
||||
readdir(path: string): Promise<FileInfo[]>;
|
||||
mkdir(path: string, recursive?: boolean): Promise<void>;
|
||||
rmdir(path: string, recursive?: boolean): Promise<void>;
|
||||
copy(src: string, dest: string): Promise<void>;
|
||||
move(src: string, dest: string): Promise<void>;
|
||||
watch?(path: string, callback: (event: string, filename: string) => void): () => void;
|
||||
}
|
||||
|
||||
export interface ProcessResult {
|
||||
exitCode: number;
|
||||
stdout: string;
|
||||
stderr: string;
|
||||
signal?: string;
|
||||
}
|
||||
|
||||
export interface ProcessOptions {
|
||||
cwd?: string;
|
||||
env?: Record<string, string>;
|
||||
timeout?: number;
|
||||
input?: string;
|
||||
shell?: boolean;
|
||||
maxBuffer?: number;
|
||||
}
|
||||
|
||||
export interface ProcessOperations {
|
||||
exec(command: string, options?: ProcessOptions): Promise<ProcessResult>;
|
||||
execFile(file: string, args: string[], options?: ProcessOptions): Promise<ProcessResult>;
|
||||
spawn?(command: string, args: string[], options?: ProcessOptions): AsyncIterable<string>;
|
||||
which?(command: string): Promise<string | null>;
|
||||
kill?(pid: number, signal?: string): Promise<boolean>;
|
||||
}
|
||||
|
||||
export interface StorageOperations {
|
||||
get(key: string): Promise<string | null>;
|
||||
set(key: string, value: string, ttl?: number): Promise<void>;
|
||||
delete(key: string): Promise<void>;
|
||||
clear(): Promise<void>;
|
||||
keys(): Promise<string[]>;
|
||||
}
|
||||
|
||||
export interface PathOperations {
|
||||
join(...paths: string[]): string;
|
||||
dirname(path: string): string;
|
||||
basename(path: string, ext?: string): string;
|
||||
extname(path: string): string;
|
||||
normalize(path: string): string;
|
||||
isAbsolute(path: string): boolean;
|
||||
resolve(...paths: string[]): string;
|
||||
relative(from: string, to: string): string;
|
||||
}
|
||||
|
||||
export interface PlatformResponse {
|
||||
status: number;
|
||||
statusText: string;
|
||||
headers: Record<string, string>;
|
||||
ok: boolean;
|
||||
text(): Promise<string>;
|
||||
json(): Promise<any>;
|
||||
arrayBuffer(): Promise<ArrayBuffer>;
|
||||
}
|
||||
|
||||
export interface PlatformFetchRequestInit {
|
||||
method?: 'GET' | 'POST' | 'PUT' | 'DELETE' | 'PATCH' | 'HEAD' | 'OPTIONS';
|
||||
headers?: Record<string, string>;
|
||||
body?: string | ArrayBuffer | Record<string, string | ArrayBuffer>;
|
||||
timeout?: number;
|
||||
}
|
||||
|
||||
export interface PlatformWebSocket {
|
||||
readonly readyState: number;
|
||||
readonly url: string;
|
||||
send(data: string | ArrayBuffer): void;
|
||||
close(code?: number, reason?: string): void;
|
||||
addEventListener(type: string, listener: (event: any) => void): void;
|
||||
removeEventListener(type: string, listener: (event: any) => void): void;
|
||||
}
|
||||
|
||||
export interface NetworkOperations {
|
||||
fetch(url: string, options?: PlatformFetchRequestInit): Promise<PlatformResponse>;
|
||||
fetchStream?(url: string, options?: PlatformFetchRequestInit): AsyncGenerator<ArrayBuffer, PlatformResponse, unknown>;
|
||||
connectWebSocket?(url: string, protocols?: string[]): Promise<PlatformWebSocket>;
|
||||
}
|
||||
|
||||
export interface PlatformInfo {
|
||||
runtime: RuntimeType;
|
||||
os: OSType;
|
||||
version?: string;
|
||||
arch?: string;
|
||||
hostname?: string;
|
||||
capabilities: PlatformCapabilities;
|
||||
}
|
||||
|
||||
export interface Tool {
|
||||
readonly id: string;
|
||||
readonly name: string;
|
||||
readonly apiName?: string;
|
||||
readonly description: string;
|
||||
readonly category: ToolCategory;
|
||||
readonly inputSchema: ToolInputSchema;
|
||||
readonly outputSchema?: ToolOutputSchema;
|
||||
readonly handler: (params: ToolInput, context: ToolExecutionContext) => Promise<ToolOutput>;
|
||||
readonly permissionLevel: ToolPermissionLevel;
|
||||
readonly requiredPermissions?: string[];
|
||||
readonly features?: ToolFeatures;
|
||||
readonly metadata?: ToolMetadata;
|
||||
readonly isMcp?: boolean;
|
||||
readonly shouldDefer?: boolean;
|
||||
readonly alwaysLoad?: boolean;
|
||||
readonly searchHint?: string;
|
||||
prompt?(options: ToolPromptOptions): Promise<string>;
|
||||
}
|
||||
|
||||
export interface Platform {
|
||||
getInfo(): PlatformInfo;
|
||||
createAbortController(): PlatformAbortController;
|
||||
readonly path: PathOperations;
|
||||
readonly fs: FileSystemOperations | null;
|
||||
readonly process: ProcessOperations | null;
|
||||
readonly storage: StorageOperations;
|
||||
readonly network: NetworkOperations;
|
||||
readonly globals: PlatformGlobals;
|
||||
readonly tools: PlatformTools;
|
||||
readonly builtinTools: Tool[];
|
||||
getEnv(key: string): string | undefined;
|
||||
getAllEnv?(): Record<string, string>;
|
||||
setEnv?(key: string, value: string): void;
|
||||
getCwd(): string;
|
||||
setCwd?(path: string): void;
|
||||
exit?(code: number): void;
|
||||
}
|
||||
|
||||
export interface CreatePlatformOptions {
|
||||
storagePath?: string;
|
||||
storageType?: 'localStorage' | 'indexedDB';
|
||||
dbName?: string;
|
||||
context?: any;
|
||||
storageName?: string;
|
||||
}
|
||||
|
||||
export type ToolInput = Record<string, unknown>;
|
||||
export type ToolOutput = string | Record<string, unknown> | void;
|
||||
|
||||
export type ToolPermissionLevel = 'safe' | 'moderate' | 'dangerous' | 'restricted';
|
||||
export type ToolCategory = 'file' | 'code' | 'search' | 'execute' | 'network' | 'analysis' | 'generation' | 'communication' | 'custom';
|
||||
|
||||
export interface ToolSchemaProperty {
|
||||
type: 'string' | 'number' | 'integer' | 'boolean' | 'array' | 'object';
|
||||
description: string;
|
||||
enum?: string[];
|
||||
default?: unknown;
|
||||
examples?: unknown[];
|
||||
items?: ToolSchemaProperty;
|
||||
properties?: Record<string, ToolSchemaProperty>;
|
||||
}
|
||||
|
||||
export interface ToolInputSchema {
|
||||
type: 'object';
|
||||
properties: Record<string, ToolSchemaProperty>;
|
||||
required?: string[];
|
||||
additionalProperties?: boolean;
|
||||
}
|
||||
|
||||
export interface ToolExecutionContext {
|
||||
toolCallId: string;
|
||||
workingDirectory: string;
|
||||
additionalWorkingDirectories?: string[];
|
||||
abortController: PlatformAbortController;
|
||||
config: {
|
||||
timeout?: number;
|
||||
maxOutputSize?: number;
|
||||
allowedDirectories?: string[];
|
||||
};
|
||||
logger: {
|
||||
info(message: string, ...args: unknown[]): void;
|
||||
warn(message: string, ...args: unknown[]): void;
|
||||
error(message: string, ...args: unknown[]): void;
|
||||
debug(message: string, ...args: unknown[]): void;
|
||||
};
|
||||
}
|
||||
|
||||
export interface BuiltinTool {
|
||||
readonly id: string;
|
||||
readonly name: string;
|
||||
readonly apiName?: string;
|
||||
readonly description: string;
|
||||
readonly category: ToolCategory;
|
||||
readonly inputSchema: ToolInputSchema;
|
||||
readonly permissionLevel: ToolPermissionLevel;
|
||||
handler(params: ToolInput, context: ToolExecutionContext): Promise<ToolOutput>;
|
||||
}
|
||||
}
|
||||
|
||||
declare module 'waterflow/runtime/core/tools/tool_interface' {
|
||||
import type { PlatformAbortController } from 'waterflow-ts/platform/types';
|
||||
import type { AgentId } from 'waterflow-ts/shared/types/agent';
|
||||
import type { WorkflowRunner } from 'waterflow-ts/runtime/core/workflow/types';
|
||||
import type { WorkflowRegistryImpl } from 'waterflow-ts/runtime/core/workflow/workflow_registry';
|
||||
import type { AgentRegistryImpl } from 'waterflow-ts/runtime/core/workflow/agent_registry';
|
||||
import type { AgentExecutor } from 'waterflow-ts/runtime/core/agent/agent_executor';
|
||||
import type { MCPClient } from 'waterflow-ts/runtime/core/tools/mcp/mcp_client';
|
||||
|
||||
export type ToolCategory =
|
||||
| 'file'
|
||||
| 'code'
|
||||
| 'search'
|
||||
| 'execute'
|
||||
| 'network'
|
||||
| 'analysis'
|
||||
| 'generation'
|
||||
| 'communication'
|
||||
| 'mcp'
|
||||
| 'custom';
|
||||
|
||||
export type PermissionLevel =
|
||||
| 'safe'
|
||||
| 'moderate'
|
||||
| 'dangerous'
|
||||
| 'restricted';
|
||||
|
||||
export type SchemaType =
|
||||
| 'string'
|
||||
| 'number'
|
||||
| 'integer'
|
||||
| 'boolean'
|
||||
| 'array'
|
||||
| 'object';
|
||||
|
||||
export interface SchemaProperty {
|
||||
type: SchemaType;
|
||||
description: string;
|
||||
enum?: string[];
|
||||
minimum?: number;
|
||||
maximum?: number;
|
||||
minLength?: number;
|
||||
maxLength?: number;
|
||||
pattern?: string;
|
||||
default?: any;
|
||||
examples?: any[];
|
||||
suggestedSource?: 'context' | 'literal' | 'file';
|
||||
items?: SchemaProperty;
|
||||
properties?: Record<string, SchemaProperty>;
|
||||
required?: string[];
|
||||
additionalProperties?: boolean | SchemaProperty;
|
||||
}
|
||||
|
||||
export interface ToolInputSchema {
|
||||
type: 'object';
|
||||
properties: Record<string, SchemaProperty>;
|
||||
required?: string[];
|
||||
additionalProperties?: boolean;
|
||||
semanticHints?: Record<string, string>;
|
||||
}
|
||||
|
||||
export interface ToolOutputSchema {
|
||||
type: 'object';
|
||||
properties: Record<string, SchemaProperty>;
|
||||
format?: 'json' | 'text' | 'markdown' | 'binary';
|
||||
maxSize?: number;
|
||||
maxLines?: number;
|
||||
}
|
||||
|
||||
export type ToolInput = Record<string, unknown>;
|
||||
export type ToolOutput = string | Record<string, unknown> | void;
|
||||
|
||||
export interface Logger {
|
||||
info(message: string, ...args: any[]): void;
|
||||
warn(message: string, ...args: any[]): void;
|
||||
error(message: string, ...args: any[]): void;
|
||||
debug(message: string, ...args: any[]): void;
|
||||
}
|
||||
|
||||
export interface ToolConfig {
|
||||
timeout?: number;
|
||||
maxOutputSize?: number;
|
||||
allowedDirectories?: string[];
|
||||
}
|
||||
|
||||
export interface ToolExecutionContext {
|
||||
toolCallId: string;
|
||||
agentId?: AgentId;
|
||||
workingDirectory: string;
|
||||
additionalWorkingDirectories?: string[] | undefined;
|
||||
abortController: PlatformAbortController;
|
||||
config: ToolConfig;
|
||||
logger: Logger;
|
||||
workflowRunner?: WorkflowRunner | undefined;
|
||||
workflowRegistry?: WorkflowRegistryImpl | undefined;
|
||||
agentExecutor?: AgentExecutor | undefined;
|
||||
agentRegistry?: AgentRegistryImpl | undefined;
|
||||
allowedAgentTypes?: string[] | undefined;
|
||||
mcpClients?: Map<string, MCPClient> | undefined;
|
||||
tools?: Tool[] | undefined;
|
||||
}
|
||||
|
||||
export type ToolHandler = (
|
||||
params: ToolInput,
|
||||
context: ToolExecutionContext
|
||||
) => Promise<ToolOutput>;
|
||||
|
||||
export interface ToolFeatures {
|
||||
isAsync?: boolean;
|
||||
isStreamable?: boolean;
|
||||
isCacheable?: boolean;
|
||||
requiresConfirmation?: boolean;
|
||||
supportsProgress?: boolean;
|
||||
supportsCancellation?: boolean;
|
||||
producesFiles?: boolean;
|
||||
producesImages?: boolean;
|
||||
producesStructuredOutput?: boolean;
|
||||
requiresMcp?: boolean;
|
||||
requiresNetwork?: boolean;
|
||||
}
|
||||
|
||||
export interface ToolExample {
|
||||
description: string;
|
||||
input: ToolInput;
|
||||
output: ToolOutput;
|
||||
explanation?: string;
|
||||
}
|
||||
|
||||
export interface ToolMetadata {
|
||||
source: 'builtin' | 'mcp' | 'plugin' | 'external';
|
||||
version?: string;
|
||||
author?: string;
|
||||
documentationUrl?: string;
|
||||
examples?: ToolExample[];
|
||||
estimatedDuration?: number;
|
||||
estimatedTokens?: number;
|
||||
compatibleModels?: string[];
|
||||
incompatibleModels?: string[];
|
||||
}
|
||||
|
||||
export interface ToolPromptOptions {
|
||||
tools: Tool[];
|
||||
agentRegistry?: AgentRegistryImpl | undefined;
|
||||
workflowRegistry?: WorkflowRegistryImpl | undefined;
|
||||
allowedAgentTypes?: string[] | undefined;
|
||||
}
|
||||
|
||||
export interface Tool {
|
||||
readonly id: string;
|
||||
readonly name: string;
|
||||
readonly apiName?: string;
|
||||
readonly description: string;
|
||||
readonly category: ToolCategory;
|
||||
readonly inputSchema: ToolInputSchema;
|
||||
readonly outputSchema?: ToolOutputSchema;
|
||||
readonly handler: ToolHandler;
|
||||
readonly permissionLevel: PermissionLevel;
|
||||
readonly requiredPermissions?: string[];
|
||||
readonly features?: ToolFeatures;
|
||||
readonly metadata?: ToolMetadata;
|
||||
readonly isMcp?: boolean;
|
||||
readonly shouldDefer?: boolean;
|
||||
readonly alwaysLoad?: boolean;
|
||||
readonly searchHint?: string;
|
||||
prompt?(options: ToolPromptOptions): Promise<string>;
|
||||
}
|
||||
|
||||
export interface ToolExecutionError {
|
||||
type: ToolErrorType;
|
||||
message: string;
|
||||
code?: string;
|
||||
details?: Record<string, unknown>;
|
||||
}
|
||||
|
||||
export type ToolErrorType =
|
||||
| 'validation_error'
|
||||
| 'permission_denied'
|
||||
| 'timeout'
|
||||
| 'execution_error'
|
||||
| 'network_error'
|
||||
| 'mcp_error'
|
||||
| 'unknown_error';
|
||||
|
||||
export interface ToolExecutionResult {
|
||||
success: boolean;
|
||||
toolId: string;
|
||||
toolCallId: string;
|
||||
output: ToolOutput;
|
||||
error?: ToolExecutionError;
|
||||
metadata: {
|
||||
duration: number;
|
||||
tokensUsed?: number;
|
||||
cached?: boolean;
|
||||
retryCount?: number;
|
||||
};
|
||||
}
|
||||
|
||||
export interface ToolCall {
|
||||
id: string;
|
||||
toolId: string;
|
||||
toolName: string;
|
||||
input: ToolInput;
|
||||
callerId: AgentId | string;
|
||||
callerType: 'agent' | 'workflow' | 'main';
|
||||
}
|
||||
|
||||
export function createTextOutput(text: string): ToolOutput;
|
||||
export function createJSONOutput(data: Record<string, unknown>): ToolOutput;
|
||||
export function createErrorOutput(message: string, code?: string, details?: Record<string, unknown>): ToolOutput;
|
||||
}
|
||||
|
||||
declare module 'waterflow/runtime/core/tools/builtin' {
|
||||
import type { Platform } from 'waterflow-ts/platform/types';
|
||||
import type { Tool } from 'waterflow-ts/runtime/core/tools/tool_interface';
|
||||
import { ToolRegistry } from 'waterflow-ts/runtime/core/tools/tool_registry';
|
||||
|
||||
export const FRAMEWORK_TOOLS: Tool[];
|
||||
export function initializeToolRegistry(platform: Platform): ToolRegistry;
|
||||
export function getFrameworkTools(): Tool[];
|
||||
}
|
||||
|
||||
declare module 'waterflow/runtime/core/tools/tool_registry' {
|
||||
import type { Tool, ToolCategory, ToolInput } from 'waterflow-ts/runtime/core/tools/tool_interface';
|
||||
|
||||
export interface ValidationResult {
|
||||
valid: boolean;
|
||||
errors?: string[];
|
||||
}
|
||||
|
||||
export interface ToolSearchResult {
|
||||
tool: Tool;
|
||||
relevanceScore: number;
|
||||
matchReason: string;
|
||||
}
|
||||
|
||||
export interface SearchOptions {
|
||||
category?: ToolCategory;
|
||||
permissionLevel?: string;
|
||||
limit?: number;
|
||||
}
|
||||
|
||||
export interface RegistryStatistics {
|
||||
totalTools: number;
|
||||
byCategory: Record<string, number>;
|
||||
bySource: Record<string, number>;
|
||||
byPermissionLevel: Record<string, number>;
|
||||
}
|
||||
|
||||
export interface ToolDefinitionExtended {
|
||||
id: string;
|
||||
name: string;
|
||||
description: string;
|
||||
parameters: {
|
||||
type: 'object';
|
||||
properties: Record<string, any>;
|
||||
required?: string[];
|
||||
additionalProperties?: boolean;
|
||||
};
|
||||
execute: (params: Record<string, any>) => Promise<any>;
|
||||
}
|
||||
|
||||
export class ToolRegistry {
|
||||
register(tool: Tool | ToolDefinitionExtended): void;
|
||||
registerAll(tools: (Tool | ToolDefinitionExtended)[]): void;
|
||||
unregister(toolId: string): void;
|
||||
get(toolId: string): Tool | ToolDefinitionExtended | undefined;
|
||||
getByName(name: string): Tool | ToolDefinitionExtended | undefined;
|
||||
has(toolId: string): boolean;
|
||||
size(): number;
|
||||
listAll(): (Tool | ToolDefinitionExtended)[];
|
||||
listByCategory(category: ToolCategory): (Tool | ToolDefinitionExtended)[];
|
||||
listByPermissionLevel(level: string): (Tool | ToolDefinitionExtended)[];
|
||||
search(query: string, options?: SearchOptions): ToolSearchResult[];
|
||||
isAvailable(toolId: string, context?: any): boolean;
|
||||
validateInput(toolId: string, input: ToolInput): ValidationResult;
|
||||
clear(): void;
|
||||
getStatistics(): RegistryStatistics;
|
||||
}
|
||||
}
|
||||
24
ts/tsconfig.json
Normal file
24
ts/tsconfig.json
Normal file
@ -0,0 +1,24 @@
|
||||
{
|
||||
"compilerOptions": {
|
||||
"target": "ES2020",
|
||||
"module": "ESNext",
|
||||
"moduleResolution": "bundler",
|
||||
"strict": true,
|
||||
"esModuleInterop": true,
|
||||
"skipLibCheck": true,
|
||||
"forceConsistentCasingInFileNames": true,
|
||||
"outDir": "./dist",
|
||||
"rootDir": "./src",
|
||||
"declaration": true,
|
||||
"declarationMap": true,
|
||||
"sourceMap": true,
|
||||
"exactOptionalPropertyTypes": true,
|
||||
"noUnusedLocals": true,
|
||||
"noUnusedParameters": true,
|
||||
"noImplicitReturns": true,
|
||||
"noFallthroughCasesInSwitch": true,
|
||||
"typeRoots": ["./src/types", "./node_modules/@types"]
|
||||
},
|
||||
"include": ["src/**/*"],
|
||||
"exclude": ["node_modules", "dist"]
|
||||
}
|
||||
@ -1,4 +0,0 @@
|
||||
from .app import GraphMemoryApp
|
||||
from .models.config import AppConfig
|
||||
|
||||
__all__ = ["GraphMemoryApp", "AppConfig"]
|
||||
275
ui/app.py
275
ui/app.py
@ -1,275 +0,0 @@
|
||||
import asyncio
|
||||
from pathlib import Path
|
||||
from textual.app import App, ComposeResult
|
||||
from textual.binding import Binding
|
||||
|
||||
from core import BackendServer, BackendClient
|
||||
from .models.message import Message
|
||||
|
||||
|
||||
class GraphMemoryApp(App):
|
||||
CSS_PATH = [
|
||||
Path(__file__).parent / "styles" / "app.css",
|
||||
Path(__file__).parent / "styles" / "messages.css",
|
||||
Path(__file__).parent / "styles" / "components.css",
|
||||
]
|
||||
|
||||
BINDINGS = [
|
||||
Binding("f1", "show_help", "帮助"),
|
||||
Binding("f2", "toggle_sidebar", "侧边栏"),
|
||||
Binding("f3", "toggle_tool_details", "工具详情"),
|
||||
Binding("f5", "clear_history", "清屏"),
|
||||
Binding("f6", "quit", "退出"),
|
||||
]
|
||||
|
||||
def __init__(self, backend_server: BackendServer = None, config_file: str = None, **kwargs):
|
||||
super().__init__(**kwargs)
|
||||
self._backend_server = backend_server
|
||||
self._backend_client = BackendClient(backend_server) if backend_server else None
|
||||
self._api_configured = False
|
||||
|
||||
def compose(self) -> ComposeResult:
|
||||
from .widgets.left_panel import LeftPanel
|
||||
from .widgets.right_panel import RightPanel
|
||||
from .widgets.status_bar import StatusBar
|
||||
from .models.config import AppConfig
|
||||
|
||||
initial_config = AppConfig()
|
||||
|
||||
if self._backend_client:
|
||||
settings_result = self._backend_client.get_settings()
|
||||
settings_data = settings_result.get("data", {})
|
||||
|
||||
api_config = settings_data.get("api_config", {})
|
||||
initial_config.api_key = api_config.get("api_key", "")
|
||||
initial_config.base_url = api_config.get("base_url", "https://api.deepseek.com")
|
||||
initial_config.model = api_config.get("model", "deepseek-chat")
|
||||
|
||||
tool_limits = settings_data.get("tool_limits", {})
|
||||
initial_config.persona_query_max = tool_limits.get("persona_query_max", 1)
|
||||
initial_config.persona_update_max = tool_limits.get("persona_update_max", 1)
|
||||
initial_config.task_query_max = tool_limits.get("task_query_max", 4)
|
||||
initial_config.task_update_max = tool_limits.get("task_update_max", 2)
|
||||
initial_config.memory_query_max = tool_limits.get("memory_query_max", 20)
|
||||
initial_config.memory_update_max = tool_limits.get("memory_update_max", 10)
|
||||
|
||||
yield LeftPanel()
|
||||
yield RightPanel(config=initial_config)
|
||||
yield StatusBar()
|
||||
|
||||
def on_mount(self) -> None:
|
||||
from .widgets.status_bar import StatusBar
|
||||
from .widgets.message_history import MessageHistory
|
||||
status_bar = self.query_one(StatusBar)
|
||||
|
||||
if not self._backend_server:
|
||||
from .widgets.message_history import MessageHistory
|
||||
history = self.query_one(MessageHistory)
|
||||
error = Message(role="assistant", content="后端未初始化")
|
||||
history.add_message(error)
|
||||
status_bar.set_api_status(False)
|
||||
return
|
||||
|
||||
status = self._backend_client.get_status()
|
||||
data = status.get("data", {})
|
||||
self._api_configured = data.get("config", {}).get("api_key", "") != ""
|
||||
status_bar.set_api_status(self._api_configured)
|
||||
|
||||
history = self.query_one(MessageHistory)
|
||||
|
||||
if self._api_configured:
|
||||
chat_history = self._backend_client.get_history()
|
||||
if chat_history:
|
||||
for msg in chat_history:
|
||||
message = Message(role=msg["role"], content=msg["content"])
|
||||
history.add_message(message)
|
||||
|
||||
welcome = Message(
|
||||
role="assistant",
|
||||
content=f"系统就绪\nAPI Key: {'已配置' if self._api_configured else '未配置'}\n\n输入消息开始对话"
|
||||
)
|
||||
history.add_message(welcome)
|
||||
|
||||
def on_unmount(self) -> None:
|
||||
if self._backend_client:
|
||||
self._backend_client.shutdown()
|
||||
|
||||
def action_show_help(self) -> None:
|
||||
from pathlib import Path
|
||||
config_path = Path.home() / ".trulymem" / "config.json"
|
||||
db_path = Path.home() / ".trulymem" / "graph_memory.db"
|
||||
|
||||
help_text = (
|
||||
"F1-帮助 F2-侧边栏 F3-工具详情 F5-清屏 F6-退出\n\n"
|
||||
f"配置文件: {config_path}\n"
|
||||
f"数据库: {db_path}"
|
||||
)
|
||||
self.notify(help_text, title="快捷键 & 配置路径", timeout=15)
|
||||
|
||||
def action_toggle_sidebar(self) -> None:
|
||||
from .widgets.right_panel import RightPanel
|
||||
sidebar = self.query_one(RightPanel)
|
||||
sidebar.toggle()
|
||||
|
||||
def action_toggle_tool_details(self) -> None:
|
||||
from .widgets.message_history import MessageHistory
|
||||
history = self.query_one(MessageHistory)
|
||||
history.toggle_latest_tool_details()
|
||||
|
||||
def action_clear_history(self) -> None:
|
||||
from .widgets.message_history import MessageHistory
|
||||
history = self.query_one(MessageHistory)
|
||||
history.clear_messages()
|
||||
|
||||
def on_input_box_send_message(self, event) -> None:
|
||||
if not self._backend_client:
|
||||
self.notify("后端未初始化", title="错误", severity="error")
|
||||
return
|
||||
|
||||
if not self._api_configured:
|
||||
self.notify("请先配置 API Key (按 F2 打开侧边栏)", title="提示", severity="warning")
|
||||
return
|
||||
|
||||
user_input = event.content
|
||||
from .widgets.message_history import MessageHistory
|
||||
from .widgets.status_bar import StatusBar
|
||||
|
||||
history = self.query_one(MessageHistory)
|
||||
status_bar = self.query_one(StatusBar)
|
||||
|
||||
history.add_message(Message(role="user", content=user_input))
|
||||
history.add_message(Message(role="assistant", content="⏳ 正在处理..."))
|
||||
status_bar.set_processing(True)
|
||||
|
||||
asyncio.create_task(self._process(user_input))
|
||||
|
||||
def on_input_box_clear_history(self, event) -> None:
|
||||
"""处理清空聊天记录事件"""
|
||||
if not self._backend_client:
|
||||
self.notify("后端未初始化", title="错误", severity="error")
|
||||
return
|
||||
|
||||
self._backend_client.clear_history()
|
||||
|
||||
from .widgets.message_history import MessageHistory
|
||||
history = self.query_one(MessageHistory)
|
||||
history.clear_messages()
|
||||
|
||||
self.notify("聊天记录已清空，AI记忆保持不变", title="提示", severity="information")
|
||||
|
||||
async def _process(self, user_input: str) -> None:
|
||||
from .widgets.message_history import MessageHistory
|
||||
from .widgets.status_bar import StatusBar
|
||||
from .widgets.right_panel import RightPanel
|
||||
from .models.log_entry import LogEntry
|
||||
from datetime import datetime
|
||||
|
||||
history = self.query_one(MessageHistory)
|
||||
status_bar = self.query_one(StatusBar)
|
||||
|
||||
result = await asyncio.get_event_loop().run_in_executor(
|
||||
None,
|
||||
lambda: self._backend_client.process_message(user_input)
|
||||
)
|
||||
|
||||
if result.get("success"):
|
||||
# 响应结构: {"success": True, "data": {"content": "...", "tool_calls": [...], ...}, "error": None}
|
||||
data = result.get("data", {})
|
||||
content = data.get("content", "(无回复)")
|
||||
history.update_latest_message(content)
|
||||
|
||||
# 处理工具调用信息，更新操作日志
|
||||
tool_calls = data.get("tool_calls", [])
|
||||
if tool_calls:
|
||||
try:
|
||||
right_panel = self.query_one(RightPanel)
|
||||
operation_log = right_panel.get_operation_log()
|
||||
|
||||
for tool_call in tool_calls:
|
||||
entry = LogEntry(
|
||||
timestamp=datetime.now(),
|
||||
tool_name=tool_call.get("name", "unknown"),
|
||||
arguments=tool_call.get("arguments", {}),
|
||||
result=str(tool_call.get("result", "")),
|
||||
duration=0.0 # 后端没有返回耗时信息
|
||||
)
|
||||
operation_log.add_log(entry)
|
||||
except Exception:
|
||||
pass # 忽略操作日志更新失败
|
||||
else:
|
||||
error = result.get("error", "未知错误")
|
||||
history.update_latest_message(f"❌ 错误: {error}")
|
||||
|
||||
status_bar.set_processing(False)
|
||||
|
||||
def on_config_section_config_changed(self, event) -> None:
|
||||
if not self._backend_client:
|
||||
self.notify("后端未初始化，无法保存配置", title="错误", severity="error")
|
||||
return
|
||||
|
||||
asyncio.create_task(self._update_settings_async(event.config))
|
||||
|
||||
async def _update_settings_async(self, config) -> None:
|
||||
from .widgets.status_bar import StatusBar
|
||||
from .widgets.config_section import ConfigSection
|
||||
|
||||
status_bar = self.query_one(StatusBar)
|
||||
|
||||
api_config = {
|
||||
"api_key": config.api_key,
|
||||
"base_url": config.base_url,
|
||||
"model": getattr(config, 'model', 'deepseek-chat')
|
||||
}
|
||||
|
||||
tool_limits = {
|
||||
"persona_query_max": config.persona_query_max,
|
||||
"persona_update_max": config.persona_update_max,
|
||||
"task_query_max": config.task_query_max,
|
||||
"task_update_max": config.task_update_max,
|
||||
"memory_query_max": config.memory_query_max,
|
||||
"memory_update_max": config.memory_update_max,
|
||||
}
|
||||
|
||||
try:
|
||||
result = await asyncio.get_event_loop().run_in_executor(
|
||||
None,
|
||||
lambda: self._backend_client.update_settings(
|
||||
api_config=api_config,
|
||||
tool_limits=tool_limits
|
||||
)
|
||||
)
|
||||
|
||||
if result.get("success"):
|
||||
self._api_configured = bool(config.api_key)
|
||||
status_bar.set_api_status(self._api_configured)
|
||||
|
||||
settings_result = await asyncio.get_event_loop().run_in_executor(
|
||||
None,
|
||||
lambda: self._backend_client.get_settings()
|
||||
)
|
||||
settings_data = settings_result.get("data", {})
|
||||
|
||||
try:
|
||||
config_section = self.query_one(ConfigSection)
|
||||
api_cfg = settings_data.get("api_config", {})
|
||||
tool_lmts = settings_data.get("tool_limits", {})
|
||||
config_section.set_config(AppConfig(
|
||||
api_key=api_cfg.get("api_key", ""),
|
||||
base_url=api_cfg.get("base_url", "https://api.deepseek.com"),
|
||||
model=api_cfg.get("model", "deepseek-chat"),
|
||||
persona_query_max=tool_lmts.get("persona_query_max", 1),
|
||||
persona_update_max=tool_lmts.get("persona_update_max", 1),
|
||||
task_query_max=tool_lmts.get("task_query_max", 4),
|
||||
task_update_max=tool_lmts.get("task_update_max", 2),
|
||||
memory_query_max=tool_lmts.get("memory_query_max", 20),
|
||||
memory_update_max=tool_lmts.get("memory_update_max", 10),
|
||||
))
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
self.notify("✅ 配置已保存并生效", title="配置成功", severity="information")
|
||||
else:
|
||||
error = result.get("error", "未知错误")
|
||||
self.notify(f"❌ 配置失败: {error}", title="配置失败", severity="error")
|
||||
except Exception as e:
|
||||
self.notify(f"❌ 配置异常: {str(e)}", title="配置失败", severity="error")
|
||||
@ -1 +0,0 @@
|
||||
"""Event Handlers for Graph Memory TUI"""
|
||||
@ -1,64 +0,0 @@
|
||||
"""焦点管理器"""
|
||||
|
||||
from textual.app import App
|
||||
|
||||
|
||||
class FocusHandler:
|
||||
"""焦点管理器"""
|
||||
|
||||
# 焦点循环顺序
|
||||
FOCUS_RING = [
|
||||
"input-textarea", # 左侧输入框
|
||||
"api-key-input", # 右侧配置区 API Key
|
||||
"model-input", # 右侧配置区 Model
|
||||
"base-url-input", # 右侧配置区 Base URL
|
||||
"cypher-textarea", # 右侧 Cypher 查询框
|
||||
]
|
||||
|
||||
# 焦点名称映射
|
||||
FOCUS_NAMES = {
|
||||
"input-textarea": "Input",
|
||||
"api-key-input": "Config-API",
|
||||
"model-input": "Config-Model",
|
||||
"base-url-input": "Config-URL",
|
||||
"cypher-textarea": "Query",
|
||||
}
|
||||
|
||||
def __init__(self):
|
||||
self._current_index = 0
|
||||
|
||||
def next_focus(self, app: App) -> None:
|
||||
"""切换到下一个焦点"""
|
||||
self._current_index = (self._current_index + 1) % len(self.FOCUS_RING)
|
||||
widget_id = self.FOCUS_RING[self._current_index]
|
||||
self._focus_widget(app, widget_id)
|
||||
|
||||
def prev_focus(self, app: App) -> None:
|
||||
"""切换到上一个焦点"""
|
||||
self._current_index = (self._current_index - 1) % len(self.FOCUS_RING)
|
||||
widget_id = self.FOCUS_RING[self._current_index]
|
||||
self._focus_widget(app, widget_id)
|
||||
|
||||
def focus_input(self, app: App) -> None:
|
||||
"""聚焦到输入框"""
|
||||
self._current_index = 0
|
||||
self._focus_widget(app, self.FOCUS_RING[0])
|
||||
|
||||
def focus_query(self, app: App) -> None:
|
||||
"""聚焦到查询框"""
|
||||
self._current_index = len(self.FOCUS_RING) - 1
|
||||
self._focus_widget(app, self.FOCUS_RING[-1])
|
||||
|
||||
def get_current_focus_name(self) -> str:
|
||||
"""获取当前焦点名称"""
|
||||
widget_id = self.FOCUS_RING[self._current_index]
|
||||
return self.FOCUS_NAMES.get(widget_id, "Unknown")
|
||||
|
||||
def _focus_widget(self, app: App, widget_id: str) -> None:
|
||||
"""聚焦到指定组件"""
|
||||
try:
|
||||
widget = app.query_one(f"#{widget_id}")
|
||||
widget.focus()
|
||||
except Exception:
|
||||
# 如果找不到组件，回退到输入框
|
||||
self.focus_input(app)
|
||||
@ -1,68 +0,0 @@
|
||||
"""快捷键处理器"""
|
||||
|
||||
from textual.app import App
|
||||
from textual.message import Message
|
||||
from .focus_handler import FocusHandler
|
||||
|
||||
|
||||
class KeyHandler:
|
||||
"""快捷键处理器"""
|
||||
|
||||
class ShowHelp(Message):
|
||||
"""显示帮助事件"""
|
||||
pass
|
||||
|
||||
class ToggleSidebar(Message):
|
||||
"""切换侧边栏事件"""
|
||||
pass
|
||||
|
||||
class ToggleToolDetails(Message):
|
||||
"""切换工具详情事件"""
|
||||
pass
|
||||
|
||||
class FocusQuery(Message):
|
||||
"""聚焦查询框事件"""
|
||||
pass
|
||||
|
||||
class ClearHistory(Message):
|
||||
"""清屏事件"""
|
||||
pass
|
||||
|
||||
class QuitApp(Message):
|
||||
"""退出应用事件"""
|
||||
pass
|
||||
|
||||
def __init__(self, focus_handler: FocusHandler):
|
||||
self._focus_handler = focus_handler
|
||||
|
||||
def handle_f1(self, app: App) -> None:
|
||||
"""处理 F1 键 - 显示帮助"""
|
||||
app.post_message(self.ShowHelp())
|
||||
|
||||
def handle_f2(self, app: App) -> None:
|
||||
"""处理 F2 键 - 切换侧边栏"""
|
||||
app.post_message(self.ToggleSidebar())
|
||||
|
||||
def handle_f3(self, app: App) -> None:
|
||||
"""处理 F3 键 - 切换工具详情"""
|
||||
app.post_message(self.ToggleToolDetails())
|
||||
|
||||
def handle_f4(self, app: App) -> None:
|
||||
"""处理 F4 键 - 聚焦查询框"""
|
||||
app.post_message(self.FocusQuery())
|
||||
|
||||
def handle_f5(self, app: App) -> None:
|
||||
"""处理 F5 键 - 清屏"""
|
||||
app.post_message(self.ClearHistory())
|
||||
|
||||
def handle_f6(self, app: App) -> None:
|
||||
"""处理 F6 键 - 退出"""
|
||||
app.post_message(self.QuitApp())
|
||||
|
||||
def handle_tab(self, app: App) -> None:
|
||||
"""处理 Tab 键 - 焦点循环"""
|
||||
self._focus_handler.next_focus(app)
|
||||
|
||||
def handle_shift_tab(self, app: App) -> None:
|
||||
"""处理 Shift+Tab 键 - 反向焦点循环"""
|
||||
self._focus_handler.prev_focus(app)
|
||||
@ -1 +0,0 @@
|
||||
"""Data Models for Graph Memory TUI"""
|
||||
@ -1,61 +0,0 @@
|
||||
"""配置数据模型"""
|
||||
|
||||
import json
|
||||
import os
|
||||
from dataclasses import dataclass, asdict
|
||||
from pathlib import Path
|
||||
from typing import Optional
|
||||
|
||||
|
||||
@dataclass
|
||||
class AppConfig:
|
||||
"""应用配置"""
|
||||
api_key: str = ""
|
||||
model: str = "deepseek-chat"
|
||||
base_url: str = "https://api.deepseek.com"
|
||||
persona_query_max: int = 1
|
||||
persona_update_max: int = 1
|
||||
task_query_max: int = 4
|
||||
task_update_max: int = 2
|
||||
memory_query_max: int = 20
|
||||
memory_update_max: int = 10
|
||||
|
||||
@classmethod
|
||||
def from_env(cls) -> "AppConfig":
|
||||
return cls(
|
||||
api_key=os.getenv("DEEPSEEK_API_KEY", ""),
|
||||
model=os.getenv("MODEL_NAME", "deepseek-chat"),
|
||||
base_url=os.getenv("DEEPSEEK_BASE_URL", "https://api.deepseek.com"),
|
||||
persona_query_max=int(os.getenv("PERSONA_QUERY_MAX", 1)),
|
||||
persona_update_max=int(os.getenv("PERSONA_UPDATE_MAX", 1)),
|
||||
task_query_max=int(os.getenv("TASK_QUERY_MAX", 4)),
|
||||
task_update_max=int(os.getenv("TASK_UPDATE_MAX", 2)),
|
||||
memory_query_max=int(os.getenv("MEMORY_QUERY_MAX", 20)),
|
||||
memory_update_max=int(os.getenv("MEMORY_UPDATE_MAX", 10)),
|
||||
)
|
||||
|
||||
@classmethod
|
||||
def from_file(cls, path: Path) -> "AppConfig":
|
||||
if not path.exists():
|
||||
return cls()
|
||||
|
||||
with open(path, "r", encoding="utf-8") as f:
|
||||
data = json.load(f)
|
||||
|
||||
return cls(
|
||||
api_key=data.get("api_key", ""),
|
||||
model=data.get("model", "deepseek-chat"),
|
||||
base_url=data.get("base_url", "https://api.deepseek.com"),
|
||||
persona_query_max=data.get("persona_query_max", 1),
|
||||
persona_update_max=data.get("persona_update_max", 1),
|
||||
task_query_max=data.get("task_query_max", 4),
|
||||
task_update_max=data.get("task_update_max", 2),
|
||||
memory_query_max=data.get("memory_query_max", 20),
|
||||
memory_update_max=data.get("memory_update_max", 10),
|
||||
)
|
||||
|
||||
def save(self, path: Path) -> None:
|
||||
path.parent.mkdir(parents=True, exist_ok=True)
|
||||
|
||||
with open(path, "w", encoding="utf-8") as f:
|
||||
json.dump(asdict(self), f, indent=2, ensure_ascii=False)
|
||||
@ -1,30 +0,0 @@
|
||||
"""日志条目数据模型"""
|
||||
|
||||
from dataclasses import dataclass
|
||||
from datetime import datetime
|
||||
from typing import Any, Dict
|
||||
|
||||
|
||||
@dataclass
|
||||
class LogEntry:
|
||||
"""日志条目"""
|
||||
timestamp: datetime
|
||||
tool_name: str
|
||||
arguments: Dict[str, Any]
|
||||
result: str
|
||||
duration: float
|
||||
|
||||
@property
|
||||
def args_summary(self) -> str:
|
||||
"""参数摘要（截断到50字符）"""
|
||||
args_str = str(self.arguments)
|
||||
if len(args_str) > 50:
|
||||
return args_str[:50] + "..."
|
||||
return args_str
|
||||
|
||||
@property
|
||||
def result_summary(self) -> str:
|
||||
"""结果摘要（截断到100字符）"""
|
||||
if len(self.result) > 100:
|
||||
return self.result[:100] + "..."
|
||||
return self.result
|
||||
@ -1,33 +0,0 @@
|
||||
"""消息数据模型"""
|
||||
|
||||
from dataclasses import dataclass, field
|
||||
from datetime import datetime
|
||||
from typing import Dict, List, Literal, Optional, Any
|
||||
|
||||
|
||||
@dataclass
|
||||
class ToolCall:
|
||||
"""工具调用"""
|
||||
id: str
|
||||
name: str
|
||||
arguments: Dict[str, Any]
|
||||
|
||||
|
||||
@dataclass
|
||||
class ToolResult:
|
||||
"""工具执行结果"""
|
||||
tool_call_id: str
|
||||
name: str
|
||||
arguments: Dict[str, Any]
|
||||
result: str
|
||||
success: bool
|
||||
|
||||
|
||||
@dataclass
|
||||
class Message:
|
||||
"""消息"""
|
||||
role: Literal["user", "assistant", "system"]
|
||||
content: str
|
||||
timestamp: datetime = field(default_factory=datetime.now)
|
||||
tool_calls: Optional[List[ToolCall]] = None
|
||||
tool_results: Optional[List[ToolResult]] = None
|
||||
@ -1 +0,0 @@
|
||||
"""Business Services for Graph Memory TUI"""
|
||||
@ -1,46 +0,0 @@
|
||||
"""
|
||||
配置管理 - 支持持久化
|
||||
"""
|
||||
|
||||
import json
|
||||
from pathlib import Path
|
||||
from ..models.config import AppConfig
|
||||
|
||||
|
||||
class ConfigManager:
|
||||
"""配置管理器 - 支持持久化"""
|
||||
|
||||
def __init__(self, config_file: str = "config.json"):
|
||||
self.config_file = Path(config_file)
|
||||
|
||||
def save(self, config: AppConfig) -> None:
|
||||
"""保存配置到文件"""
|
||||
data = {
|
||||
"api_key": config.api_key,
|
||||
"model": config.model,
|
||||
"base_url": config.base_url
|
||||
}
|
||||
|
||||
with open(self.config_file, 'w', encoding='utf-8') as f:
|
||||
json.dump(data, f, indent=2)
|
||||
|
||||
def load(self) -> AppConfig:
|
||||
"""从文件加载配置"""
|
||||
if not self.config_file.exists():
|
||||
return AppConfig()
|
||||
|
||||
try:
|
||||
with open(self.config_file, 'r', encoding='utf-8') as f:
|
||||
data = json.load(f)
|
||||
|
||||
return AppConfig(
|
||||
api_key=data.get("api_key", ""),
|
||||
model=data.get("model", "deepseek-chat"),
|
||||
base_url=data.get("base_url", "https://api.deepseek.com")
|
||||
)
|
||||
except Exception:
|
||||
return AppConfig()
|
||||
|
||||
def exists(self) -> bool:
|
||||
"""检查配置文件是否存在"""
|
||||
return self.config_file.exists()
|
||||
@ -1,51 +0,0 @@
|
||||
"""配置服务"""
|
||||
|
||||
from pathlib import Path
|
||||
from typing import TYPE_CHECKING
|
||||
from ..models.config import AppConfig
|
||||
|
||||
if TYPE_CHECKING:
|
||||
from ..core.imports import GraphMemoryClient
|
||||
|
||||
|
||||
class ConfigService:
|
||||
"""配置服务"""
|
||||
|
||||
DEFAULT_CONFIG_FILE = Path.home() / ".graph_memory_tui" / "config.json"
|
||||
|
||||
def __init__(self, config_file: Path | None = None):
|
||||
self._config_file = config_file or self.DEFAULT_CONFIG_FILE
|
||||
self._config = self._load_config()
|
||||
|
||||
def _load_config(self) -> AppConfig:
|
||||
"""加载配置"""
|
||||
# 优先从文件加载
|
||||
if self._config_file.exists():
|
||||
return AppConfig.from_file(self._config_file)
|
||||
|
||||
# 否则从环境变量加载
|
||||
return AppConfig.from_env()
|
||||
|
||||
def get_config(self) -> AppConfig:
|
||||
"""获取当前配置"""
|
||||
return self._config
|
||||
|
||||
def set_config(self, config: AppConfig) -> None:
|
||||
"""设置配置"""
|
||||
self._config = config
|
||||
self._save_config()
|
||||
|
||||
def _save_config(self) -> None:
|
||||
"""保存配置"""
|
||||
self._config.save(self._config_file)
|
||||
|
||||
def apply_to_client(self, client: "GraphMemoryClient") -> None:
|
||||
"""应用配置到 API 客户端"""
|
||||
# 更新客户端配置
|
||||
client.api_key = self._config.api_key
|
||||
client.base_url = self._config.base_url
|
||||
client.model = self._config.model
|
||||
|
||||
def get_config_file(self) -> Path:
|
||||
"""获取配置文件路径"""
|
||||
return self._config_file
|
||||
@ -1 +0,0 @@
|
||||
"""Styles for Graph Memory TUI"""
|
||||
@ -1,40 +0,0 @@
|
||||
/* Global Styles for Graph Memory TUI */
|
||||
|
||||
GraphMemoryApp {
|
||||
background: $surface;
|
||||
color: $text;
|
||||
}
|
||||
|
||||
/* 全局Input样式 - 确保可见 */
|
||||
Input {
|
||||
background: $surface-lighten-1;
|
||||
color: $text;
|
||||
border: solid $primary;
|
||||
}
|
||||
|
||||
Input:focus {
|
||||
border: double $accent;
|
||||
}
|
||||
|
||||
LeftPanel {
|
||||
width: 1fr;
|
||||
dock: left;
|
||||
}
|
||||
|
||||
RightPanel {
|
||||
width: 35;
|
||||
dock: right;
|
||||
background: $panel;
|
||||
}
|
||||
|
||||
RightPanel ScrollableContainer {
|
||||
height: 100%;
|
||||
overflow-y: scroll;
|
||||
}
|
||||
|
||||
StatusBar {
|
||||
dock: bottom;
|
||||
height: 1;
|
||||
background: $primary;
|
||||
color: $text-primary;
|
||||
}
|
||||
@ -1,139 +0,0 @@
|
||||
/* Component Styles for Graph Memory TUI */
|
||||
|
||||
/* Input Box - 最重要 */
|
||||
InputBox {
|
||||
background: $surface;
|
||||
padding: 1 2;
|
||||
height: auto;
|
||||
border: solid $primary;
|
||||
}
|
||||
|
||||
InputBox TextArea {
|
||||
width: 100%;
|
||||
height: 5;
|
||||
background: $surface-lighten-1;
|
||||
color: $text;
|
||||
border: none;
|
||||
}
|
||||
|
||||
InputBox .input-buttons {
|
||||
height: auto;
|
||||
margin-top: 1;
|
||||
}
|
||||
|
||||
InputBox Button {
|
||||
margin: 0;
|
||||
}
|
||||
|
||||
/* Config Section */
|
||||
ConfigSection {
|
||||
background: $surface;
|
||||
padding: 1;
|
||||
margin: 0 0 1 0;
|
||||
height: auto;
|
||||
}
|
||||
|
||||
ConfigSection .config-title {
|
||||
color: $primary;
|
||||
text-style: bold;
|
||||
margin: 0 0 1 0;
|
||||
}
|
||||
|
||||
ConfigSection .config-label {
|
||||
color: $text;
|
||||
margin: 0;
|
||||
padding: 1 0 0 0;
|
||||
}
|
||||
|
||||
ConfigSection .config-hint {
|
||||
color: $text-muted;
|
||||
text-style: italic;
|
||||
margin: 1 0 0 0;
|
||||
}
|
||||
|
||||
ConfigSection Input {
|
||||
width: 1fr;
|
||||
height: 3;
|
||||
margin: 0 0 1 0;
|
||||
padding: 0 1;
|
||||
background: $surface-lighten-1;
|
||||
border: solid $primary;
|
||||
color: $text;
|
||||
}
|
||||
|
||||
/* Other Components */
|
||||
OperationLog {
|
||||
background: $surface-darken-1;
|
||||
height: 1fr;
|
||||
margin: 1;
|
||||
overflow-y: auto;
|
||||
padding: 1;
|
||||
}
|
||||
|
||||
OperationLog .log-entry {
|
||||
color: $text;
|
||||
margin: 0 0 1 0;
|
||||
height: auto;
|
||||
}
|
||||
|
||||
OperationLog .log-empty {
|
||||
color: $text-muted;
|
||||
text-style: italic;
|
||||
}
|
||||
|
||||
CypherQueryBox {
|
||||
border: solid green;
|
||||
margin: 1;
|
||||
height: auto;
|
||||
}
|
||||
|
||||
MessageHistory {
|
||||
height: 1fr;
|
||||
margin: 1;
|
||||
overflow-y: auto;
|
||||
}
|
||||
|
||||
/* Message Widget */
|
||||
MessageWidget {
|
||||
margin: 1 0;
|
||||
height: auto;
|
||||
}
|
||||
|
||||
MessageWidget .message-header {
|
||||
color: $text-muted;
|
||||
text-style: bold;
|
||||
margin: 0 0 0 0;
|
||||
}
|
||||
|
||||
MessageWidget .message-content {
|
||||
color: $text;
|
||||
margin: 0 0 0 2;
|
||||
height: auto;
|
||||
}
|
||||
|
||||
MessageWidget .tool-indicator {
|
||||
color: $warning;
|
||||
text-style: bold;
|
||||
margin: 1 0 0 2;
|
||||
}
|
||||
|
||||
MessageWidget .tool-details {
|
||||
background: $surface-darken-1;
|
||||
margin: 1 0 0 2;
|
||||
padding: 1;
|
||||
}
|
||||
|
||||
MessageWidget .tool-name {
|
||||
color: $accent;
|
||||
text-style: bold;
|
||||
}
|
||||
|
||||
MessageWidget .tool-args {
|
||||
color: $text-muted;
|
||||
margin: 0 0 0 2;
|
||||
}
|
||||
|
||||
MessageWidget .tool-result {
|
||||
color: $success;
|
||||
margin: 0 0 0 2;
|
||||
}
|
||||
@ -1,24 +0,0 @@
|
||||
/* Message Styles for Graph Memory TUI */
|
||||
|
||||
UserMessage {
|
||||
border: solid orange;
|
||||
margin: 1 0;
|
||||
padding: 1;
|
||||
}
|
||||
|
||||
ModelMessage {
|
||||
border: solid blue;
|
||||
margin: 1 0;
|
||||
padding: 1;
|
||||
}
|
||||
|
||||
ToolCallIndicator {
|
||||
color: yellow;
|
||||
text-style: bold;
|
||||
}
|
||||
|
||||
ToolCallDetails {
|
||||
background: $surface-darken-1;
|
||||
margin: 1 0 0 2;
|
||||
padding: 1;
|
||||
}
|
||||
@ -1 +0,0 @@
|
||||
"""UI Widgets for Graph Memory TUI"""
|
||||
@ -1,172 +0,0 @@
|
||||
"""配置区组件"""
|
||||
|
||||
from textual.containers import Vertical
|
||||
from textual.widgets import Static, Input, Button
|
||||
from textual.app import ComposeResult
|
||||
from textual.message import Message
|
||||
from ..models.config import AppConfig
|
||||
|
||||
|
||||
class ConfigSection(Vertical):
|
||||
|
||||
class ConfigChanged(Message):
|
||||
def __init__(self, config: AppConfig, is_tool_limits: bool = False) -> None:
|
||||
self.config = config
|
||||
self.is_tool_limits = is_tool_limits
|
||||
super().__init__()
|
||||
|
||||
def __init__(self, config: AppConfig | None = None, **kwargs):
|
||||
super().__init__(**kwargs)
|
||||
self._config = config or AppConfig()
|
||||
|
||||
def compose(self) -> ComposeResult:
|
||||
title = Static("━━ 配置 ━━", classes="config-title")
|
||||
title.can_focus = False
|
||||
yield title
|
||||
|
||||
label1 = Static("API Key:", classes="config-label")
|
||||
label1.can_focus = False
|
||||
yield label1
|
||||
|
||||
yield Input(
|
||||
value=self._config.api_key,
|
||||
placeholder="sk-xxxxxxxxxxxxx",
|
||||
id="api-key-input",
|
||||
password=True
|
||||
)
|
||||
|
||||
label2 = Static("模型:", classes="config-label")
|
||||
label2.can_focus = False
|
||||
yield label2
|
||||
|
||||
yield Input(
|
||||
value=self._config.model,
|
||||
placeholder="deepseek-chat",
|
||||
id="model-input"
|
||||
)
|
||||
|
||||
label3 = Static("Base URL:", classes="config-label")
|
||||
label3.can_focus = False
|
||||
yield label3
|
||||
|
||||
yield Input(
|
||||
value=self._config.base_url,
|
||||
placeholder="https://api.deepseek.com",
|
||||
id="base-url-input"
|
||||
)
|
||||
|
||||
sep = Static("", classes="config-sep")
|
||||
sep.can_focus = False
|
||||
yield sep
|
||||
|
||||
limits_title = Static("━━ 工具限制 ━━", classes="config-title")
|
||||
limits_title.can_focus = False
|
||||
yield limits_title
|
||||
|
||||
l1 = Static("人设图查询:", classes="config-label")
|
||||
l1.can_focus = False
|
||||
yield l1
|
||||
yield Input(value=str(self._config.persona_query_max), placeholder="1", id="persona-query-max")
|
||||
|
||||
l2 = Static("人设图修改:", classes="config-label")
|
||||
l2.can_focus = False
|
||||
yield l2
|
||||
yield Input(value=str(self._config.persona_update_max), placeholder="1", id="persona-update-max")
|
||||
|
||||
l3 = Static("工作记忆查询:", classes="config-label")
|
||||
l3.can_focus = False
|
||||
yield l3
|
||||
yield Input(value=str(self._config.task_query_max), placeholder="4", id="task-query-max")
|
||||
|
||||
l4 = Static("工作记忆修改:", classes="config-label")
|
||||
l4.can_focus = False
|
||||
yield l4
|
||||
yield Input(value=str(self._config.task_update_max), placeholder="2", id="task-update-max")
|
||||
|
||||
l5 = Static("一般记忆查询:", classes="config-label")
|
||||
l5.can_focus = False
|
||||
yield l5
|
||||
yield Input(value=str(self._config.memory_query_max), placeholder="20", id="memory-query-max")
|
||||
|
||||
l6 = Static("一般记忆修改:", classes="config-label")
|
||||
l6.can_focus = False
|
||||
yield l6
|
||||
yield Input(value=str(self._config.memory_update_max), placeholder="10", id="memory-update-max")
|
||||
|
||||
hint = Static("按Enter保存配置", classes="config-hint")
|
||||
hint.can_focus = False
|
||||
yield hint
|
||||
|
||||
def on_mount(self) -> None:
|
||||
try:
|
||||
api_key = self.query_one("#api-key-input", Input)
|
||||
model = self.query_one("#model-input", Input)
|
||||
base_url = self.query_one("#base-url-input", Input)
|
||||
|
||||
api_key.tab_index = 0
|
||||
model.tab_index = 1
|
||||
base_url.tab_index = 2
|
||||
|
||||
if self._config.api_key:
|
||||
api_key.value = self._config.api_key
|
||||
if self._config.model:
|
||||
model.value = self._config.model
|
||||
if self._config.base_url:
|
||||
base_url.value = self._config.base_url
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
def on_input_submitted(self, event: Input.Submitted) -> None:
|
||||
try:
|
||||
api_key_input = self.query_one("#api-key-input", Input)
|
||||
model_input = self.query_one("#model-input", Input)
|
||||
base_url_input = self.query_one("#base-url-input", Input)
|
||||
|
||||
persona_query = self.query_one("#persona-query-max", Input)
|
||||
persona_update = self.query_one("#persona-update-max", Input)
|
||||
task_query = self.query_one("#task-query-max", Input)
|
||||
task_update = self.query_one("#task-update-max", Input)
|
||||
memory_query = self.query_one("#memory-query-max", Input)
|
||||
memory_update = self.query_one("#memory-update-max", Input)
|
||||
|
||||
self._config = AppConfig(
|
||||
api_key=api_key_input.value,
|
||||
model=model_input.value,
|
||||
base_url=base_url_input.value,
|
||||
persona_query_max=int(persona_query.value or 1),
|
||||
persona_update_max=int(persona_update.value or 1),
|
||||
task_query_max=int(task_query.value or 4),
|
||||
task_update_max=int(task_update.value or 2),
|
||||
memory_query_max=int(memory_query.value or 20),
|
||||
memory_update_max=int(memory_update.value or 10),
|
||||
)
|
||||
|
||||
# 先发送 API 配置更新（is_tool_limits=False）
|
||||
self.post_message(self.ConfigChanged(self._config, is_tool_limits=False))
|
||||
# 再发送工具限制更新（is_tool_limits=True）
|
||||
self.post_message(self.ConfigChanged(self._config, is_tool_limits=True))
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
def get_config(self) -> AppConfig:
|
||||
return self._config
|
||||
|
||||
def set_config(self, config: AppConfig) -> None:
|
||||
self._config = config
|
||||
try:
|
||||
api_key_input = self.query_one("#api-key-input", Input)
|
||||
model_input = self.query_one("#model-input", Input)
|
||||
base_url_input = self.query_one("#base-url-input", Input)
|
||||
|
||||
api_key_input.value = config.api_key
|
||||
model_input.value = config.model
|
||||
base_url_input.value = config.base_url
|
||||
|
||||
self.query_one("#persona-query-max", Input).value = str(config.persona_query_max)
|
||||
self.query_one("#persona-update-max", Input).value = str(config.persona_update_max)
|
||||
self.query_one("#task-query-max", Input).value = str(config.task_query_max)
|
||||
self.query_one("#task-update-max", Input).value = str(config.task_update_max)
|
||||
self.query_one("#memory-query-max", Input).value = str(config.memory_query_max)
|
||||
self.query_one("#memory-update-max", Input).value = str(config.memory_update_max)
|
||||
except Exception:
|
||||
pass
|
||||
@ -1,57 +0,0 @@
|
||||
"""Cypher查询框组件"""
|
||||
|
||||
from textual.containers import Container, Horizontal
|
||||
from textual.widgets import Static, TextArea, Button
|
||||
from textual.app import ComposeResult
|
||||
from textual.message import Message
|
||||
|
||||
|
||||
class CypherQueryBox(Container):
|
||||
"""快捷Cypher查询输入框"""
|
||||
|
||||
class ExecuteQuery(Message):
|
||||
"""执行查询事件"""
|
||||
def __init__(self, query: str) -> None:
|
||||
self.query = query
|
||||
super().__init__()
|
||||
|
||||
def compose(self) -> ComposeResult:
|
||||
"""构建查询框"""
|
||||
yield Static("F4:执行Cypher查询", classes="query-title")
|
||||
yield TextArea(
|
||||
placeholder="输入Cypher查询语句...",
|
||||
id="cypher-textarea"
|
||||
)
|
||||
with Horizontal(classes="query-buttons"):
|
||||
yield Button("执行", id="execute-button", variant="primary")
|
||||
yield Button("清空", id="clear-button")
|
||||
|
||||
def on_button_pressed(self, event: Button.Pressed) -> None:
|
||||
"""处理按钮点击"""
|
||||
if event.button.id == "execute-button":
|
||||
self._execute_query()
|
||||
elif event.button.id == "clear-button":
|
||||
self._clear_query()
|
||||
|
||||
def on_key(self, event) -> None:
|
||||
"""处理按键事件"""
|
||||
if event.key == "enter" and event.ctrl:
|
||||
event.stop()
|
||||
self._execute_query()
|
||||
|
||||
def _execute_query(self) -> None:
|
||||
"""执行查询"""
|
||||
textarea = self.query_one("#cypher-textarea", TextArea)
|
||||
query = textarea.text.strip()
|
||||
if query:
|
||||
self.post_message(self.ExecuteQuery(query))
|
||||
|
||||
def _clear_query(self) -> None:
|
||||
"""清空查询"""
|
||||
textarea = self.query_one("#cypher-textarea", TextArea)
|
||||
textarea.clear()
|
||||
|
||||
def focus(self) -> None:
|
||||
"""聚焦查询框"""
|
||||
textarea = self.query_one("#cypher-textarea", TextArea)
|
||||
textarea.focus()
|
||||
@ -1,72 +0,0 @@
|
||||
"""输入框组件"""
|
||||
|
||||
from textual.containers import Container, Horizontal
|
||||
from textual.widgets import TextArea, Button
|
||||
from textual.message import Message
|
||||
|
||||
|
||||
class InputBox(Container):
|
||||
"""输入框组件"""
|
||||
|
||||
class SendMessage(Message):
|
||||
"""发送消息事件"""
|
||||
def __init__(self, content: str) -> None:
|
||||
self.content = content
|
||||
super().__init__()
|
||||
|
||||
class ClearHistory(Message):
|
||||
"""清空聊天记录事件"""
|
||||
def __init__(self) -> None:
|
||||
super().__init__()
|
||||
|
||||
def __init__(self, **kwargs):
|
||||
super().__init__(**kwargs)
|
||||
self._history: list[str] = []
|
||||
self._history_index: int = -1
|
||||
|
||||
def compose(self):
|
||||
"""构建输入框"""
|
||||
yield TextArea(
|
||||
placeholder="输入消息... (Enter换行)",
|
||||
id="input-textarea"
|
||||
)
|
||||
with Horizontal(classes="input-buttons"):
|
||||
yield Button("清空", id="clear-button", variant="default")
|
||||
yield Button("发送", id="send-button", variant="primary")
|
||||
|
||||
def on_mount(self) -> None:
|
||||
"""组件挂载时"""
|
||||
# 设置焦点
|
||||
textarea = self.query_one(TextArea)
|
||||
textarea.focus()
|
||||
|
||||
def on_button_pressed(self, event: Button.Pressed) -> None:
|
||||
"""处理按钮点击"""
|
||||
if event.button.id == "send-button":
|
||||
self._send_message()
|
||||
elif event.button.id == "clear-button":
|
||||
self.post_message(self.ClearHistory())
|
||||
|
||||
def on_key(self, event) -> None:
|
||||
"""处理按键事件"""
|
||||
if event.key == "enter" and event.ctrl:
|
||||
self._send_message()
|
||||
event.stop()
|
||||
|
||||
def _send_message(self) -> None:
|
||||
"""发送消息"""
|
||||
textarea = self.query_one(TextArea)
|
||||
content = textarea.text.strip()
|
||||
if content:
|
||||
# 保存到历史
|
||||
self._history.append(content)
|
||||
self._history_index = len(self._history)
|
||||
# 发送消息
|
||||
self.post_message(self.SendMessage(content))
|
||||
# 清空输入框
|
||||
textarea.clear()
|
||||
|
||||
def focus(self) -> None:
|
||||
"""聚焦输入框"""
|
||||
textarea = self.query_one(TextArea)
|
||||
textarea.focus()
|
||||
@ -1,23 +0,0 @@
|
||||
"""左侧面板"""
|
||||
|
||||
from textual.containers import Container
|
||||
from textual.app import ComposeResult
|
||||
from .message_history import MessageHistory
|
||||
from .input_box import InputBox
|
||||
|
||||
|
||||
class LeftPanel(Container):
|
||||
"""左侧主面板"""
|
||||
|
||||
def compose(self) -> ComposeResult:
|
||||
"""构建左侧面板"""
|
||||
yield MessageHistory()
|
||||
yield InputBox()
|
||||
|
||||
def get_message_history(self) -> MessageHistory:
|
||||
"""获取消息历史组件"""
|
||||
return self.query_one(MessageHistory)
|
||||
|
||||
def get_input_box(self) -> InputBox:
|
||||
"""获取输入框组件"""
|
||||
return self.query_one(InputBox)
|
||||
@ -1,57 +0,0 @@
|
||||
"""消息历史组件"""
|
||||
|
||||
from textual.containers import ScrollableContainer
|
||||
from textual.message import Message
|
||||
from .message_widget import MessageWidget
|
||||
from ..models.message import Message as MessageModel
|
||||
|
||||
|
||||
class MessageHistory(ScrollableContainer):
|
||||
"""消息历史区域"""
|
||||
|
||||
def __init__(self, **kwargs):
|
||||
super().__init__(**kwargs)
|
||||
self._messages: list[MessageModel] = []
|
||||
|
||||
def compose(self):
|
||||
"""构建消息历史"""
|
||||
for message in self._messages:
|
||||
yield MessageWidget(message)
|
||||
|
||||
def add_message(self, message: MessageModel) -> None:
|
||||
"""添加新消息"""
|
||||
self._messages.append(message)
|
||||
# 添加新组件
|
||||
message_widget = MessageWidget(message)
|
||||
self.mount(message_widget)
|
||||
# 滚动到最新消息
|
||||
self.scroll_to_widget(message_widget, animate=False)
|
||||
|
||||
def update_latest_message(self, content: str) -> None:
|
||||
"""更新最新消息的内容"""
|
||||
if self.children:
|
||||
latest_widget = self.children[-1]
|
||||
if isinstance(latest_widget, MessageWidget):
|
||||
latest_widget.update_content(content)
|
||||
# 确保滚动到最新消息
|
||||
self.scroll_to_widget(latest_widget, animate=False)
|
||||
|
||||
def clear_messages(self) -> None:
|
||||
"""清空消息历史"""
|
||||
self._messages.clear()
|
||||
# 移除所有子组件
|
||||
for child in self.children:
|
||||
child.remove()
|
||||
|
||||
def get_latest_message(self) -> MessageModel | None:
|
||||
"""获取最新消息"""
|
||||
if self._messages:
|
||||
return self._messages[-1]
|
||||
return None
|
||||
|
||||
def toggle_latest_tool_details(self) -> None:
|
||||
"""切换最新消息的工具详情"""
|
||||
if self.children:
|
||||
latest_widget = self.children[-1]
|
||||
if isinstance(latest_widget, MessageWidget):
|
||||
latest_widget.toggle_tool_details()
|
||||
@ -1,108 +0,0 @@
|
||||
"""消息组件"""
|
||||
|
||||
from textual.containers import Container, Vertical
|
||||
from textual.widgets import Static
|
||||
from textual.message import Message
|
||||
from textual.css.query import NoMatches
|
||||
from ..models.message import Message as MessageModel
|
||||
|
||||
|
||||
class MessageWidget(Container):
|
||||
"""单条消息组件"""
|
||||
|
||||
def __init__(self, message: MessageModel, **kwargs):
|
||||
super().__init__(**kwargs)
|
||||
self._message = message
|
||||
self._show_tool_details = False
|
||||
self._content_widget = None # 保存内容组件的引用
|
||||
self._tool_details_container = None # 保存工具详情容器引用
|
||||
|
||||
def compose(self):
|
||||
"""构建消息组件"""
|
||||
# 消息头
|
||||
role_emoji = "🟠" if self._message.role == "user" else "🔵"
|
||||
timestamp_str = self._message.timestamp.strftime("%H:%M:%S")
|
||||
yield Static(
|
||||
f"{role_emoji} {timestamp_str}",
|
||||
classes="message-header"
|
||||
)
|
||||
|
||||
# 消息内容 - 保存引用以便后续更新
|
||||
self._content_widget = Static(
|
||||
self._message.content,
|
||||
classes="message-content"
|
||||
)
|
||||
yield self._content_widget
|
||||
|
||||
# 工具调用指示器
|
||||
if self._message.tool_calls:
|
||||
tool_count = len(self._message.tool_calls)
|
||||
toggle_hint = "(F3折叠)" if self._show_tool_details else "(F3展开)"
|
||||
yield Static(
|
||||
f"[工具:{tool_count}次] {toggle_hint}",
|
||||
classes="tool-indicator"
|
||||
)
|
||||
|
||||
# 工具调用详情容器 - 始终创建，但根据状态显示/隐藏
|
||||
self._tool_details_container = Vertical(classes="tool-details")
|
||||
with self._tool_details_container:
|
||||
for i, tool_call in enumerate(self._message.tool_calls, 1):
|
||||
yield Static(
|
||||
f"工具 {i}: {tool_call.name}",
|
||||
classes="tool-name"
|
||||
)
|
||||
yield Static(
|
||||
f"参数: {tool_call.arguments}",
|
||||
classes="tool-args"
|
||||
)
|
||||
|
||||
# 显示执行结果
|
||||
if self._message.tool_results:
|
||||
for result in self._message.tool_results:
|
||||
if result.tool_call_id == tool_call.id:
|
||||
# 显示完整结果，不截断
|
||||
result_text = result.result
|
||||
# 如果结果太长，只显示前1000字符，但提供完整信息
|
||||
if len(result_text) > 1000:
|
||||
result_text = result_text[:1000] + f"\n... (共{len(result.result)}字符，按F3查看完整内容)"
|
||||
yield Static(
|
||||
f"结果: {result_text}",
|
||||
classes="tool-result"
|
||||
)
|
||||
|
||||
# 根据状态设置初始显示/隐藏
|
||||
if not self._show_tool_details:
|
||||
self._tool_details_container.styles.display = "none"
|
||||
|
||||
def update_content(self, new_content: str) -> None:
|
||||
"""更新消息内容"""
|
||||
self._message.content = new_content
|
||||
if self._content_widget:
|
||||
self._content_widget.update(new_content)
|
||||
|
||||
def toggle_tool_details(self) -> None:
|
||||
"""切换工具详情显示状态"""
|
||||
if self._message.tool_calls and self._tool_details_container:
|
||||
self._show_tool_details = not self._show_tool_details
|
||||
|
||||
# 切换显示/隐藏
|
||||
if self._show_tool_details:
|
||||
self._tool_details_container.styles.display = "block"
|
||||
else:
|
||||
self._tool_details_container.styles.display = "none"
|
||||
|
||||
# 更新指示器文字
|
||||
self._update_indicator()
|
||||
|
||||
# 刷新布局
|
||||
self.refresh(layout=True)
|
||||
|
||||
def _update_indicator(self) -> None:
|
||||
"""更新工具调用指示器文字"""
|
||||
try:
|
||||
indicator = self.query_one(".tool-indicator", Static)
|
||||
tool_count = len(self._message.tool_calls)
|
||||
toggle_hint = "(F3折叠)" if self._show_tool_details else "(F3展开)"
|
||||
indicator.update(f"[工具:{tool_count}次] {toggle_hint}")
|
||||
except NoMatches:
|
||||
pass
|
||||
@ -1,66 +0,0 @@
|
||||
"""操作日志组件"""
|
||||
|
||||
from datetime import datetime
|
||||
from textual.containers import ScrollableContainer
|
||||
from textual.widgets import Static
|
||||
from ..models.log_entry import LogEntry
|
||||
|
||||
|
||||
class OperationLog(ScrollableContainer):
|
||||
"""图操作日志区域"""
|
||||
|
||||
def __init__(self, max_entries: int = 100, **kwargs):
|
||||
super().__init__(**kwargs)
|
||||
self._logs: list[LogEntry] = []
|
||||
self._max_entries = max_entries
|
||||
|
||||
def compose(self):
|
||||
"""构建日志区域"""
|
||||
if not self._logs:
|
||||
yield Static("暂无操作日志", classes="log-empty")
|
||||
|
||||
def add_log(self, entry: LogEntry) -> None:
|
||||
"""添加日志（插入到顶部）"""
|
||||
# 限制日志数量
|
||||
if len(self._logs) >= self._max_entries:
|
||||
self._logs.pop()
|
||||
# 移除最旧的组件
|
||||
if self.children:
|
||||
self.children[-1].remove()
|
||||
|
||||
# 插入到列表开头
|
||||
self._logs.insert(0, entry)
|
||||
|
||||
# 创建日志显示组件
|
||||
log_widget = self._create_log_widget(entry)
|
||||
|
||||
# 挂载到顶部
|
||||
self.mount(log_widget, before=0 if self.children else None)
|
||||
|
||||
# 滚动到顶部
|
||||
self.scroll_to(0, animate=False)
|
||||
|
||||
def _create_log_widget(self, entry: LogEntry) -> Static:
|
||||
"""创建日志显示组件"""
|
||||
timestamp_str = entry.timestamp.strftime("%H:%M:%S")
|
||||
text = (
|
||||
f"[{timestamp_str}] {entry.tool_name}\n"
|
||||
f" 参数: {entry.args_summary}\n"
|
||||
f" 结果: {entry.result_summary}\n"
|
||||
f" 耗时: {entry.duration:.2f}s"
|
||||
)
|
||||
return Static(text, classes="log-entry")
|
||||
|
||||
def clear_logs(self) -> None:
|
||||
"""清空日志"""
|
||||
self._logs.clear()
|
||||
for child in self.children:
|
||||
child.remove()
|
||||
# 显示空状态
|
||||
self.mount(Static("暂无操作日志", classes="log-empty"))
|
||||
|
||||
def get_latest_log(self) -> LogEntry | None:
|
||||
"""获取最新日志"""
|
||||
if self._logs:
|
||||
return self._logs[0]
|
||||
return None
|
||||
@ -1,67 +0,0 @@
|
||||
"""右侧面板"""
|
||||
|
||||
from textual.containers import Container, ScrollableContainer
|
||||
from textual.css.query import NoMatches
|
||||
from textual.widgets import Static
|
||||
from textual.app import ComposeResult
|
||||
from .config_section import ConfigSection
|
||||
from .operation_log import OperationLog
|
||||
from .cypher_query_box import CypherQueryBox
|
||||
from ..models.config import AppConfig
|
||||
|
||||
|
||||
class RightPanel(Container):
|
||||
"""右侧边栏"""
|
||||
|
||||
def __init__(self, config: AppConfig | None = None, use_embedded_db: bool = True, **kwargs):
|
||||
super().__init__(**kwargs)
|
||||
self._is_collapsed = False
|
||||
self._config = config or AppConfig()
|
||||
self._use_embedded_db = use_embedded_db
|
||||
|
||||
def compose(self) -> ComposeResult:
|
||||
"""构建右侧面板"""
|
||||
yield Static("F2:隐藏侧边栏", classes="sidebar-title")
|
||||
with ScrollableContainer():
|
||||
yield ConfigSection(self._config)
|
||||
yield OperationLog()
|
||||
if not self._use_embedded_db:
|
||||
yield CypherQueryBox()
|
||||
|
||||
def toggle(self) -> None:
|
||||
"""切换折叠/展开"""
|
||||
self._is_collapsed = not self._is_collapsed
|
||||
if self._is_collapsed:
|
||||
self.styles.width = 0
|
||||
self.styles.display = "none"
|
||||
else:
|
||||
self.styles.width = 35
|
||||
self.styles.display = "block"
|
||||
|
||||
def is_collapsed(self) -> bool:
|
||||
"""检查是否折叠"""
|
||||
return self._is_collapsed
|
||||
|
||||
def get_config_section(self) -> ConfigSection:
|
||||
"""获取配置区组件"""
|
||||
return self.query_one(ConfigSection)
|
||||
|
||||
def get_operation_log(self) -> OperationLog:
|
||||
"""获取操作日志组件"""
|
||||
return self.query_one(OperationLog)
|
||||
|
||||
def get_cypher_query_box(self) -> CypherQueryBox | None:
|
||||
"""获取Cypher查询框组件（可能不存在）"""
|
||||
try:
|
||||
return self.query_one(CypherQueryBox)
|
||||
except NoMatches:
|
||||
return None
|
||||
|
||||
def has_cypher_query_box(self) -> bool:
|
||||
"""检查是否存在Cypher查询框"""
|
||||
return not self._use_embedded_db
|
||||
|
||||
def update_title(self) -> None:
|
||||
"""更新标题"""
|
||||
title = self.query_one(Static)
|
||||
title.update("F2:展开侧边栏" if self._is_collapsed else "F2:隐藏侧边栏")
|
||||
@ -1,41 +0,0 @@
|
||||
"""状态栏组件"""
|
||||
|
||||
from textual.widgets import Static
|
||||
from textual.message import Message
|
||||
|
||||
|
||||
class StatusBar(Static):
|
||||
"""底部状态栏"""
|
||||
|
||||
class FocusChanged(Message):
|
||||
"""焦点变更事件"""
|
||||
def __init__(self, focus_name: str) -> None:
|
||||
self.focus_name = focus_name
|
||||
super().__init__()
|
||||
|
||||
def __init__(self, **kwargs):
|
||||
super().__init__(**kwargs)
|
||||
self._shortcuts = "F1:帮助 F2:侧边栏 F3:工具详情 F5:清屏 F6:退出"
|
||||
self._license_info = "本项目由jianf设计，以GPLv3形式开源"
|
||||
self._api_status = "未配置"
|
||||
self._processing = False
|
||||
|
||||
def on_mount(self) -> None:
|
||||
"""组件挂载时"""
|
||||
self._update_display()
|
||||
|
||||
def _update_display(self) -> None:
|
||||
"""更新显示"""
|
||||
status_icon = "●" if self._api_status == "已配置" else "○"
|
||||
processing_indicator = " [处理中...]" if self._processing else ""
|
||||
self.update(f"{status_icon} API: {self._api_status}{processing_indicator} | {self._license_info} | {self._shortcuts}")
|
||||
|
||||
def set_api_status(self, configured: bool) -> None:
|
||||
"""设置API状态"""
|
||||
self._api_status = "已配置" if configured else "未配置"
|
||||
self._update_display()
|
||||
|
||||
def set_processing(self, processing: bool) -> None:
|
||||
"""设置处理状态"""
|
||||
self._processing = processing
|
||||
self._update_display()
|
||||
Reference in New Issue
Block a user