diff --git a/GEMINI.md b/GEMINI.md
new file mode 100644
index 0000000..b000568
--- /dev/null
+++ b/GEMINI.md
@@ -0,0 +1,53 @@
+
+
+
+
+# Vibe Coding Guide - AI Assistant Context
+
+
+
+## 🚀 Project Overview
+
+**Vibe Coding** is a comprehensive guide and workflow for AI-assisted pair programming. The goal is to provide a structured and efficient way to turn ideas into reality by leveraging the power of AI. This project is not about a specific programming language or framework, but rather a methodology that can be applied to any software development project.
+
+The core of the project is a collection of documents, prompts, and tools that guide the developer and the AI assistant through the entire development process, from initial conception to final implementation.
+
+## 的核心理念 (Core Philosophy)
+
+The "Vibe Coding" methodology is based on the following key principles:
+
+* **Planning is everything:** A detailed implementation plan is created before any code is written. This plan is then executed step-by-step, with each step being tested and verified before moving on to the next.
+* **Modularization:** The project is broken down into small, manageable modules that can be developed and tested independently.
+* **Context is king:** The AI assistant is provided with a "memory-bank" of all the relevant project documents, such as the game design document, tech stack, and implementation plan. This ensures that the AI has a deep understanding of the project's context and can provide accurate and relevant assistance.
+* **AI as a partner:** The AI assistant is not just a code generator, but a true partner in the development process. The developer and the AI work together, with the developer providing the high-level guidance and the AI providing the low-level implementation details.
+
+## 📂 Folder Structure
+
+The most important files and directories in this project are:
+
+* `README.md`: The main entry point for the project, providing an overview and links to all the other resources.
+* `i18n/`: Contains the internationalization files for the project, with subdirectories for each supported language.
+* `i18n/en/`: The English version of the project documentation.
+* `i18n/zh/`: The Chinese version of the project documentation.
+* `libs/`: Contains common library code that can be used across different projects.
+* `prompts/`: A collection of prompts for different stages of the development process.
+* `skills/`: A collection of reusable skills that can be used to extend the functionality of the AI assistant.
+
+## 🤖 AI Assistant's Role
+
+As the AI assistant for this project, you are expected to:
+
+* **Be a true partner:** Work collaboratively with the developer to achieve the project's goals.
+* **Be proactive:** Ask clarifying questions and provide suggestions to improve the project.
+* **Be a good communicator:** Clearly explain your reasoning and provide detailed explanations of your work.
+* **Be a good learner:** Continuously learn from your interactions with the developer and the project's context.
+* **Follow the "Vibe Coding" methodology:** Adhere to the principles of plan-driven development, modularization, and context-awareness.
+
+By following these guidelines, you will be able to provide the best possible assistance to the developer and help them turn their ideas into reality.
diff --git a/README.md b/README.md
index ba17c97..712ad50 100644
--- a/README.md
+++ b/README.md
@@ -303,6 +303,7 @@
### 项目内部文档
* [**胶水编程 (Glue Coding)**](./i18n/zh/documents/胶水编程/): 软件工程的圣杯与银弹，Vibe Coding 的终极进化形态。
+* [**Chat Vault**](./libs/external/chat-vault/): AI 聊天记录保存工具，支持 Codex/Kiro/Gemini/Claude CLI。
* [**prompts-library 工具说明**](./libs/external/prompts-library/): 支持 Excel 与 Markdown 格式互转，包含数百个精选提示词。
* [**coding_prompts 集合**](./i18n/zh/prompts/coding_prompts/): 适用于 Vibe Coding 流程的专用提示词。
* [**系统提示词构建原则**](./i18n/zh/documents/方法论与原则/系统提示词构建原则.md): 构建高效 AI 系统提示词的综合指南。
diff --git a/i18n/zh/README.md b/i18n/zh/README.md
index 8e2a2af..950a47b 100644
--- a/i18n/zh/README.md
+++ b/i18n/zh/README.md
@@ -221,6 +221,7 @@
* [**在线提示词数据库**](https://docs.google.com/spreadsheets/d/1ngoQOhJqdguwNAilCl1joNwTje7FWWN9WiI2bo5VhpU/edit?gid=2093180351#gid=2093180351&range=A1): 包含数百个适用于各场景的用户及系统提示词的在线表格。
* [**第三方系统提示词仓库**](https://github.com/x1xhlol/system-prompts-and-models-of-ai-tools): 汇集了多种 AI 工具的系统提示词。
* **项目内部文档**:
+ * [**Chat Vault**](../../libs/external/chat-vault/): AI 聊天记录保存工具，支持 Codex/Kiro/Gemini/Claude CLI。
* [**prompts-library 工具说明**](./libs/external/prompts-library/): 该工具支持在 Excel 和 Markdown 格式之间转换提示词，并包含数百个精选提示词。
* [**coding_prompts 集合**](./i18n/zh/prompts/coding_prompts/): 适用于 Vibe Coding 流程的专用提示词。
* [**系统提示词构建原则**](./documents/方法论与原则/系统提示词构建原则.md): 关于如何构建高效、可靠的 AI 系统提示词的综合指南。
diff --git a/libs/README.md b/libs/README.md
index 2bed630..a667be6 100644
--- a/libs/README.md
+++ b/libs/README.md
@@ -27,8 +27,11 @@ libs/
│ └── .gitkeep
└── external/
├── README.md
+ ├── chat-vault/
├── prompts-library/
+ ├── l10n-tool/
├── my-nvim/
+ ├── MCPlayerTransfer/
├── XHS-image-to-PDF-conversion/
└── .gitkeep
```
@@ -57,6 +60,7 @@ libs/
## 常用入口
+- AI 聊天记录保存：[`external/chat-vault/`](./external/chat-vault/)（支持 Codex/Kiro/Gemini/Claude CLI）
- 提示词批量管理：[`external/prompts-library/`](./external/prompts-library/)（配合 `../prompts/` 使用）
- 备份工具：优先使用仓库根目录的 `backups/`（当前与 `libs/common/utils/backups/` 内容一致）
diff --git a/libs/external/README.md b/libs/external/README.md
index e2542e1..40da5df 100644
--- a/libs/external/README.md
+++ b/libs/external/README.md
@@ -11,16 +11,22 @@
```
libs/external/
├── README.md
+├── chat-vault/ # AI 聊天记录保存工具
├── prompts-library/ # 提示词库管理工具（Excel ↔ Markdown）
+├── l10n-tool/ # 多语言翻译脚本
├── my-nvim/ # Neovim 配置（含 nvim-config/）
+├── MCPlayerTransfer/ # MC 玩家迁移工具
├── XHS-image-to-PDF-conversion/ # 图片合并 PDF 工具
└── .gitkeep
```
## 工具清单（入口与文档）
+- `chat-vault/`：AI 聊天记录保存工具，支持 Codex/Kiro/Gemini/Claude CLI（详见 [`chat-vault/README_CN.md`](./chat-vault/README_CN.md)）
- `prompts-library/`：提示词 Excel ↔ Markdown 批量互转与索引生成（详见 [`prompts-library/README.md`](./prompts-library/README.md)）
+- `l10n-tool/`：多语言批量翻译脚本
- `my-nvim/`：个人 Neovim 配置（详见 [`my-nvim/README.md`](./my-nvim/README.md)）
+- `MCPlayerTransfer/`：MC 玩家迁移工具
- `XHS-image-to-PDF-conversion/`：图片合并 PDF（详见 [`XHS-image-to-PDF-conversion/README.md`](./XHS-image-to-PDF-conversion/README.md)）
## 新增外部工具（最小清单）
diff --git a/libs/external/chat-vault/.env.example b/libs/external/chat-vault/.env.example
new file mode 100644
index 0000000..3b45d02
--- /dev/null
+++ b/libs/external/chat-vault/.env.example
@@ -0,0 +1,11 @@
+# AI Chat Converter Configuration (Optional)
+# Default: Auto-detect paths, no configuration needed
+
+# Custom paths (comma-separated for multiple)
+# CODEX_PATHS=~/.codex/sessions
+# KIRO_PATHS=~/.local/share/kiro-cli
+# GEMINI_PATHS=~/.gemini/tmp
+# CLAUDE_PATHS=~/.claude
+
+# WSL paths also supported
+# CODEX_PATHS=\\wsl.localhost\Ubuntu\home\user\.codex\sessions
diff --git a/libs/external/chat-vault/.gitignore b/libs/external/chat-vault/.gitignore
new file mode 100644
index 0000000..3660ee3
--- /dev/null
+++ b/libs/external/chat-vault/.gitignore
@@ -0,0 +1,28 @@
+# Python
+__pycache__/
+*.py[cod]
+*.so
+*.egg-info/
+dist/
+build/
+*.spec
+
+# Output
+output/
+*.db
+*.sqlite3
+*.log
+
+# Environment
+.env
+.venv/
+venv/
+
+# IDE
+.vscode/
+.idea/
+*.swp
+
+# OS
+.DS_Store
+Thumbs.db
diff --git a/libs/external/chat-vault/LICENSE b/libs/external/chat-vault/LICENSE
new file mode 100644
index 0000000..c13f991
--- /dev/null
+++ b/libs/external/chat-vault/LICENSE
@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/libs/external/chat-vault/README.md b/libs/external/chat-vault/README.md
new file mode 100644
index 0000000..c82ab18
--- /dev/null
+++ b/libs/external/chat-vault/README.md
@@ -0,0 +1,318 @@
+
+
+# 🔐 Chat Vault
+
+**One tool to save ALL your AI chat history**
+
+[](https://python.org)
+[](LICENSE)
+[]()
+[]()
+
+[English](README.md) | [中文](README_CN.md)
+
+[✨ Features](#-features) •
+[🚀 Quick Start](#-quick-start) •
+[📋 Commands](#-commands) •
+[📁 Project Structure](#-project-structure) •
+[❓ FAQ](#-faq)
+
+[📞 Contact](#-contact) •
+[✨ Support](#-support) •
+[🤝 Contributing](#-contributing)
+
+AI-powered docs: [zread.ai/tukuaiai/chat-vault](https://zread.ai/tukuaiai/chat-vault)
+
+> 📦 This tool is part of [vibe-coding-cn](https://github.com/tukuaiai/vibe-coding-cn) - A comprehensive Vibe Coding guide
+
+
+
+---
+
+## ✨ Features
+
+
+
+🔄 Multi-CLI
+Codex, Kiro, Gemini, Claude - all supported
+
+
+⚡ Real-time
+Watch mode with system-level file monitoring
+
+
+🔢 Token Stats
+Accurate counting using tiktoken (cl100k_base)
+
+
+🔍 Search
+Find any conversation instantly
+
+
+📤 Export
+JSON or CSV, your choice
+
+
+🚀 Zero Config
+Auto-detects paths, just run it
+
+
+
+---
+
+## 🏗️ Architecture
+
+```mermaid
+graph LR
+ subgraph Sources
+ A[~/.codex]
+ B[~/.kiro]
+ C[~/.gemini]
+ D[~/.claude]
+ end
+
+ subgraph Chat Vault
+ E[Watcher]
+ F[Parsers]
+ G[Storage]
+ end
+
+ subgraph Output
+ H[(SQLite DB)]
+ end
+
+ A --> E
+ B --> E
+ C --> E
+ D --> E
+ E --> F
+ F --> G
+ G --> H
+```
+
+---
+
+## 🔄 How It Works
+
+```mermaid
+sequenceDiagram
+ participant User
+ participant CLI as AI CLI (Codex/Kiro/...)
+ participant Watcher
+ participant Parser
+ participant DB as SQLite
+
+ User->>CLI: Chat with AI
+ CLI->>CLI: Save to local file
+ Watcher->>Watcher: Detect file change
+ Watcher->>Parser: Parse new content
+ Parser->>DB: Upsert session
+ DB-->>User: Query anytime
+```
+
+---
+
+## 🚀 Quick Start
+
+### 30 Seconds Setup
+
+```bash
+# Clone
+git clone https://github.com/tukuaiai/vibe-coding-cn.git
+cd vibe-coding-cn/libs/external/chat-vault
+
+# Run (auto-installs dependencies)
+./start.sh # Linux/macOS
+start.bat # Windows
+```
+
+**That's it!** 🎉
+
+---
+
+## 📊 Example Output
+
+```
+==================================================
+AI 聊天记录 → 集中存储
+==================================================
+数据库: ./output/chat_history.db
+
+[Codex] 新增:1241 更新:0 跳过:0 错误:0
+[Kiro] 新增:21 更新:0 跳过:0 错误:0
+[Gemini] 新增:332 更新:0 跳过:0 错误:0
+[Claude] 新增:168 更新:0 跳过:0 错误:0
+
+==================================================
+总计: 1762 会话, 40000+ 消息
+✓ 同步完成!
+
+=== Token 统计 (tiktoken) ===
+ codex: 11,659,952 tokens
+ kiro: 26,337 tokens
+ gemini: 3,195,821 tokens
+ claude: 29,725 tokens
+ 总计: 14,911,835 tokens
+```
+
+---
+
+## 📋 Commands
+
+| Command | Description |
+|---------|-------------|
+| `python src/main.py` | Sync once |
+| `python src/main.py -w` | Watch mode (real-time) |
+| `python src/main.py --stats` | Show statistics |
+| `python src/main.py --search "keyword"` | Search messages |
+| `python src/main.py --export json` | Export to JSON |
+| `python src/main.py --export csv --source codex` | Export specific source |
+| `python src/main.py --prune` | Clean orphaned records |
+
+---
+
+## 📁 Project Structure
+
+```
+chat-vault/
+├── 🚀 start.sh / start.bat # One-click start
+├── 📦 build.py # Build standalone exe
+├── 📂 src/
+│ ├── main.py # CLI entry
+│ ├── config.py # Auto-detection
+│ ├── storage.py # SQLite + tiktoken
+│ ├── watcher.py # File monitoring
+│ └── parsers/ # CLI parsers
+├── 📂 docs/
+│ ├── AI_PROMPT.md # AI assistant guide
+│ └── schema.md # Database schema
+└── 📂 output/
+ ├── chat_history.db # Your database
+ └── logs/ # Sync logs
+```
+
+---
+
+## 🗄️ Database Schema
+
+```mermaid
+erDiagram
+ sessions {
+ TEXT file_path PK
+ TEXT session_id
+ TEXT source
+ TEXT cwd
+ TEXT messages
+ INTEGER file_mtime
+ TEXT start_time
+ INTEGER token_count
+ }
+
+ meta {
+ TEXT key PK
+ TEXT value
+ }
+
+ meta_codex {
+ TEXT key PK
+ TEXT value
+ }
+```
+
+---
+
+## 🤖 For AI Assistants
+
+Send [docs/AI_PROMPT.md](docs/AI_PROMPT.md) to your AI assistant for:
+- SQL query examples
+- Python code snippets
+- Task guidance
+
+---
+
+## ❓ FAQ
+
+
+Do I need to configure anything?
+
+
+Does it work with WSL?
+
+
+How do I view the database?
+
+
+Is my data safe?
+
+---
+
+## 📞 Contact
+
+- **GitHub**: [tukuaiai](https://github.com/tukuaiai)
+- **Twitter / X**: [123olp](https://x.com/123olp)
+- **Telegram**: [@desci0](https://t.me/desci0)
+- **Telegram Group**: [glue_coding](https://t.me/glue_coding)
+- **Telegram Channel**: [tradecat_ai_channel](https://t.me/tradecat_ai_channel)
+- **Email**: tukuai.ai@gmail.com
+
+---
+
+## ✨ Support
+
+If this project helped you, consider supporting:
+
+- **Binance UID**: `572155580`
+- **Tron (TRC20)**: `TQtBXCSTwLFHjBqTS4rNUp7ufiGx51BRey`
+- **Solana**: `HjYhozVf9AQmfv7yv79xSNs6uaEU5oUk2USasYQfUYau`
+- **Ethereum (ERC20)**: `0xa396923a71ee7D9480b346a17dDeEb2c0C287BBC`
+- **BNB Smart Chain (BEP20)**: `0xa396923a71ee7D9480b346a17dDeEb2c0C287BBC`
+- **Bitcoin**: `bc1plslluj3zq3snpnnczplu7ywf37h89dyudqua04pz4txwh8z5z5vsre7nlm`
+- **Sui**: `0xb720c98a48c77f2d49d375932b2867e793029e6337f1562522640e4f84203d2e`
+
+---
+
+## 🤝 Contributing
+
+We welcome all contributions! Feel free to open an [Issue](https://github.com/tukuaiai/vibe-coding-cn/issues) or submit a [Pull Request](https://github.com/tukuaiai/vibe-coding-cn/pulls).
+
+---
+
+## 📄 License
+
+[MIT](LICENSE) - Do whatever you want with it.
+
+---
+
+
+
+**If this helped you, give it a ⭐!**
+
+## Star History
+
+
+
+
+
+
+---
+
+**Made with ❤️ by [tukuaiai](https://github.com/tukuaiai)**
+
+[⬆ Back to Top](#-chat-vault)
+
+
+
+# 🔐 Chat Vault
+
+**一个工具保存你所有的 AI 聊天记录**
+
+[](https://python.org)
+[](LICENSE)
+[]()
+[]()
+
+[English](README.md) | [中文](README_CN.md)
+
+[✨ 功能特性](#-功能特性) •
+[🚀 快速开始](#-30-秒快速开始) •
+[📋 命令一览](#-命令一览) •
+[📁 项目结构](#-项目结构) •
+[❓ 常见问题](#-常见问题)
+
+[📞 联系方式](#-联系方式) •
+[✨ 支持项目](#-支持项目) •
+[🤝 参与贡献](#-参与贡献)
+
+AI 解读文档: [zread.ai/tukuaiai/chat-vault](https://zread.ai/tukuaiai/chat-vault)
+
+> 📦 本工具是 [vibe-coding-cn](https://github.com/tukuaiai/vibe-coding-cn) 的一部分 - 一份全面的 Vibe Coding 指南
+
+
+
+---
+
+## ✨ 功能特性
+
+
+
+🔄 多 CLI 支持
+Codex、Kiro、Gemini、Claude 全都行
+
+
+⚡ 实时同步
+系统级文件监控，聊完自动保存
+
+
+🔢 Token 统计
+tiktoken 精确计算，知道你用了多少
+
+
+🔍 搜索
+秒找任何对话
+
+
+📤 导出
+JSON 或 CSV，随你选
+
+
+🚀 零配置
+自动检测路径，开箱即用
+
+
+
+---
+
+## 🏗️ 架构图
+
+```mermaid
+graph LR
+ subgraph 数据来源
+ A[~/.codex]
+ B[~/.kiro]
+ C[~/.gemini]
+ D[~/.claude]
+ end
+
+ subgraph Chat Vault
+ E[监控器]
+ F[解析器]
+ G[存储层]
+ end
+
+ subgraph 输出
+ H[(SQLite 数据库)]
+ end
+
+ A --> E
+ B --> E
+ C --> E
+ D --> E
+ E --> F
+ F --> G
+ G --> H
+```
+
+---
+
+## 🔄 工作流程
+
+```mermaid
+sequenceDiagram
+ participant 用户
+ participant CLI as AI CLI (Codex/Kiro/...)
+ participant 监控器
+ participant 解析器
+ participant DB as SQLite
+
+ 用户->>CLI: 和 AI 聊天
+ CLI->>CLI: 保存到本地文件
+ 监控器->>监控器: 检测文件变化
+ 监控器->>解析器: 解析新内容
+ 解析器->>DB: 写入数据库
+ DB-->>用户: 随时查询
+```
+
+---
+
+## 🚀 30 秒快速开始
+
+```bash
+# 下载
+git clone https://github.com/tukuaiai/vibe-coding-cn.git
+cd vibe-coding-cn/libs/external/chat-vault
+
+# 运行（自动安装依赖）
+./start.sh # Linux/macOS
+start.bat # Windows（双击）
+```
+
+**搞定！** 🎉
+
+---
+
+## 📊 运行效果
+
+```
+==================================================
+AI 聊天记录 → 集中存储
+==================================================
+数据库: ./output/chat_history.db
+
+[Codex] 新增:1241 更新:0 跳过:0 错误:0
+[Kiro] 新增:21 更新:0 跳过:0 错误:0
+[Gemini] 新增:332 更新:0 跳过:0 错误:0
+[Claude] 新增:168 更新:0 跳过:0 错误:0
+
+==================================================
+总计: 1762 会话, 40000+ 消息
+✓ 同步完成!
+
+=== Token 统计 (tiktoken) ===
+ codex: 11,659,952 tokens
+ kiro: 26,337 tokens
+ gemini: 3,195,821 tokens
+ claude: 29,725 tokens
+ 总计: 14,911,835 tokens
+```
+
+---
+
+## 📋 命令一览
+
+| 命令 | 说明 |
+|------|------|
+| `python src/main.py` | 同步一次 |
+| `python src/main.py -w` | 实时监控（推荐） |
+| `python src/main.py --stats` | 查看统计 |
+| `python src/main.py --search "关键词"` | 搜索消息 |
+| `python src/main.py --export json` | 导出 JSON |
+| `python src/main.py --export csv --source codex` | 导出指定来源 |
+| `python src/main.py --prune` | 清理孤立记录 |
+
+---
+
+## 📁 项目结构
+
+```
+chat-vault/
+├── 🚀 start.sh / start.bat # 一键启动
+├── 📦 build.py # 打包脚本
+├── 📂 src/
+│ ├── main.py # 主程序
+│ ├── config.py # 配置检测
+│ ├── storage.py # SQLite + tiktoken
+│ ├── watcher.py # 文件监控
+│ └── parsers/ # 各 CLI 解析器
+├── 📂 docs/
+│ ├── AI_PROMPT.md # AI 助手指南
+│ └── schema.md # 数据库结构
+└── 📂 output/
+ ├── chat_history.db # 你的数据库
+ └── logs/ # 日志
+```
+
+---
+
+## 🗄️ 数据库结构
+
+```mermaid
+erDiagram
+ sessions {
+ TEXT file_path PK "文件路径"
+ TEXT session_id "会话ID"
+ TEXT source "来源"
+ TEXT cwd "工作目录"
+ TEXT messages "消息JSON"
+ INTEGER file_mtime "修改时间"
+ TEXT start_time "开始时间"
+ INTEGER token_count "Token数"
+ }
+
+ meta {
+ TEXT key PK
+ TEXT value
+ }
+```
+
+---
+
+## 🤖 让 AI 帮你查数据库
+
+把 [docs/AI_PROMPT.md](docs/AI_PROMPT.md) 发给 AI 助手，它就知道：
+- 怎么写 SQL 查询
+- 怎么用 Python 分析
+- 怎么帮你找对话
+
+---
+
+## ❓ 常见问题
+
+
+需要配置什么吗？
+
+
+WSL 能用吗？
+
+
+怎么看数据库？
+
+
+会不会搞坏我的数据？
+
+---
+
+## 📞 联系方式
+
+- **GitHub**: [tukuaiai](https://github.com/tukuaiai)
+- **Twitter / X**: [123olp](https://x.com/123olp)
+- **Telegram**: [@desci0](https://t.me/desci0)
+- **Telegram 交流群**: [glue_coding](https://t.me/glue_coding)
+- **Telegram 频道**: [tradecat_ai_channel](https://t.me/tradecat_ai_channel)
+- **邮箱**: tukuai.ai@gmail.com
+
+---
+
+## ✨ 支持项目
+
+如果这个项目帮到你了，考虑支持一下：
+
+- **币安 UID**: `572155580`
+- **Tron (TRC20)**: `TQtBXCSTwLFHjBqTS4rNUp7ufiGx51BRey`
+- **Solana**: `HjYhozVf9AQmfv7yv79xSNs6uaEU5oUk2USasYQfUYau`
+- **Ethereum (ERC20)**: `0xa396923a71ee7D9480b346a17dDeEb2c0C287BBC`
+- **BNB Smart Chain (BEP20)**: `0xa396923a71ee7D9480b346a17dDeEb2c0C287BBC`
+- **Bitcoin**: `bc1plslluj3zq3snpnnczplu7ywf37h89dyudqua04pz4txwh8z5z5vsre7nlm`
+- **Sui**: `0xb720c98a48c77f2d49d375932b2867e793029e6337f1562522640e4f84203d2e`
+
+---
+
+## 🤝 参与贡献
+
+欢迎各种形式的贡献！随时开启一个 [Issue](https://github.com/tukuaiai/vibe-coding-cn/issues) 或提交 [Pull Request](https://github.com/tukuaiai/vibe-coding-cn/pulls)。
+
+---
+
+## 📄 开源协议
+
+[MIT](LICENSE) - 随便用，不用管我
+
+---
+
+
+
+**如果帮到你了，点个 ⭐ 呗！**
+
+## Star History
+
+
+
+
+
+
+---
+
+**Made with ❤️ by [tukuaiai](https://github.com/tukuaiai)**
+
+[⬆ 返回顶部](#-chat-vault)
+
+
# 搜索
+ python main.py --export json|csv [--source codex|kiro|gemini|claude]
+"""
+import os
+import sys
+import subprocess
+
+# 项目根目录
+PROJECT_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+VENV_DIR = os.path.join(PROJECT_DIR, '.venv')
+REQUIREMENTS = os.path.join(PROJECT_DIR, 'requirements.txt')
+
+def ensure_venv():
+ """检测并创建虚拟环境，安装依赖"""
+ # 打包版本跳过
+ if getattr(sys, 'frozen', False):
+ return
+
+ # 已在虚拟环境中运行则跳过
+ if sys.prefix != sys.base_prefix:
+ return
+
+ # 检查 .venv 是否存在
+ venv_python = os.path.join(VENV_DIR, 'bin', 'python') if os.name != 'nt' else os.path.join(VENV_DIR, 'Scripts', 'python.exe')
+
+ if not os.path.exists(venv_python):
+ print("首次运行，创建虚拟环境...")
+ subprocess.run([sys.executable, '-m', 'venv', VENV_DIR], check=True)
+ print("安装依赖...")
+ pip = os.path.join(VENV_DIR, 'bin', 'pip') if os.name != 'nt' else os.path.join(VENV_DIR, 'Scripts', 'pip.exe')
+ subprocess.run([pip, 'install', '-r', REQUIREMENTS, '-q'], check=True)
+ print("环境准备完成，重新启动...\n")
+
+ # 使用虚拟环境重新执行
+ os.execv(venv_python, [venv_python] + sys.argv)
+
+# 启动前检测虚拟环境
+ensure_venv()
+
+# 支持 PyInstaller 打包
+if getattr(sys, 'frozen', False):
+ BASE_DIR = sys._MEIPASS
+ sys.path.insert(0, os.path.join(BASE_DIR, 'src'))
+else:
+ BASE_DIR = os.path.dirname(os.path.abspath(__file__))
+
+import argparse
+from config import CONFIG
+from parsers import CodexParser, GeminiParser, ClaudeParser, KiroParser
+from storage import ChatStorage
+from logger import setup_logger, get_logger
+
+storage: ChatStorage = None
+
+def main():
+ global storage
+
+ parser = argparse.ArgumentParser(description='AI Chat Converter')
+ parser.add_argument('-w', '--watch', action='store_true', help='持续监控模式')
+ parser.add_argument('--prune', action='store_true', help='清理孤立记录')
+ parser.add_argument('--stats', action='store_true', help='显示统计信息')
+ parser.add_argument('--search', type=str, help='搜索关键词')
+ parser.add_argument('--export', choices=['json', 'csv'], help='导出格式')
+ parser.add_argument('--source', choices=['codex', 'kiro', 'gemini', 'claude'], help='指定来源')
+ parser.add_argument('--output', type=str, help='导出文件路径')
+ args = parser.parse_args()
+
+ # 初始化
+ setup_logger(CONFIG["log_dir"])
+ log = get_logger()
+
+ storage = ChatStorage(CONFIG["db_path"])
+
+ # 命令分发
+ if args.prune:
+ cmd_prune()
+ elif args.stats:
+ cmd_stats()
+ elif args.search:
+ cmd_search(args.search, args.source)
+ elif args.export:
+ cmd_export(args.export, args.source, args.output)
+ elif args.watch:
+ cmd_sync()
+ cmd_watch()
+ else:
+ cmd_sync()
+
+def cmd_sync():
+ log = get_logger()
+ log.info("=" * 50)
+ log.info("AI 聊天记录 → 集中存储")
+ log.info("=" * 50)
+ log.info(f"数据库: {CONFIG['db_path']}")
+
+ total_added, total_updated, total_skipped, total_errors = 0, 0, 0, 0
+
+ for cli, key, parser_cls in [
+ ('codex', 'codex_paths', lambda: CodexParser('codex')),
+ ('kiro', 'kiro_paths', KiroParser),
+ ('gemini', 'gemini_paths', GeminiParser),
+ ('claude', 'claude_paths', ClaudeParser),
+ ]:
+ paths = CONFIG.get(key, [])
+ if not paths:
+ continue
+
+ parser = parser_cls()
+ if cli in ('claude', 'kiro'):
+ a, u, s, e = process_multi(parser, paths, cli)
+ else:
+ a, u, s, e = process(parser, paths)
+
+ log.info(f"[{cli.capitalize()}] 新增:{a} 更新:{u} 跳过:{s} 错误:{e}")
+ update_cli_meta(cli)
+ total_added += a
+ total_updated += u
+ total_skipped += s
+ total_errors += e
+
+ total = storage.get_total_stats()
+ storage.update_total_meta(total['sessions'], total['messages'], total['tokens'])
+
+ log.info("=" * 50)
+ log.info(f"总计: {total['sessions']} 会话, {total['messages']} 消息")
+ if total_errors > 0:
+ log.warning(f"错误: {total_errors} 个文件解析失败")
+ log.info("✓ 同步完成!")
+
+ print_token_stats()
+
+def cmd_watch():
+ from watcher import ChatWatcher
+ from datetime import datetime
+
+ log = get_logger()
+ log.info("")
+ log.info("=" * 50)
+ log.info("实时监听模式 (watchdog)")
+ log.info("=" * 50)
+
+ watch_paths = []
+ path_source_map = {}
+
+ for cli, key in [('codex', 'codex_paths'), ('kiro', 'kiro_paths'),
+ ('gemini', 'gemini_paths'), ('claude', 'claude_paths')]:
+ for p in CONFIG.get(key, []):
+ if os.path.isdir(p) or os.path.isfile(p):
+ watch_paths.append(p)
+ path_source_map[p] = cli
+
+ def on_change(file_path, event_type):
+ now = datetime.now().strftime('%H:%M:%S')
+ source = None
+ for p, s in path_source_map.items():
+ if file_path.startswith(p) or file_path == p:
+ source = s
+ break
+ if not source:
+ return
+
+ try:
+ if source == 'kiro':
+ parser = KiroParser()
+ for sess in parser.parse_file(file_path):
+ storage.upsert_session(sess.session_id, sess.source, sess.file_path, sess.cwd, sess.messages, int(sess.file_mtime))
+ log.info(f"[{now}] kiro 更新")
+ elif source == 'claude':
+ parser = ClaudeParser()
+ for sess in parser.parse_file(file_path):
+ fp = f"claude:{sess.session_id}"
+ storage.upsert_session(sess.session_id, sess.source, fp, sess.cwd, sess.messages, int(sess.file_mtime))
+ log.info(f"[{now}] claude 更新")
+ else:
+ parser = CodexParser(source) if source == 'codex' else GeminiParser()
+ sess = parser.parse_file(file_path)
+ fp = os.path.abspath(sess.file_path)
+ storage.upsert_session(sess.session_id, sess.source, fp, sess.cwd, sess.messages, int(sess.file_mtime))
+ log.info(f"[{now}] {source} {event_type}: {os.path.basename(file_path)}")
+
+ update_cli_meta(source)
+ total = storage.get_total_stats()
+ storage.update_total_meta(total['sessions'], total['messages'], total['tokens'])
+ except Exception as e:
+ log.error(f"[{now}] 处理失败 {file_path}: {e}")
+
+ log.info(f"监听目录: {len(watch_paths)} 个")
+ watcher = ChatWatcher(watch_paths, on_change)
+ watcher.start()
+
+def cmd_prune():
+ log = get_logger()
+ log.info("清理孤立记录...")
+ removed = storage.prune()
+ total = sum(removed.values())
+ if total > 0:
+ for cli, count in removed.items():
+ if count > 0:
+ log.info(f" {cli}: 删除 {count} 条")
+ log.info(f"✓ 共清理 {total} 条孤立记录")
+ else:
+ log.info("✓ 无孤立记录")
+
+def cmd_stats():
+ log = get_logger()
+ meta = storage.get_total_meta()
+ tokens = storage.get_token_stats()
+
+ log.info("=" * 50)
+ log.info("统计信息")
+ log.info("=" * 50)
+ log.info(f"数据库: {CONFIG['db_path']}")
+ log.info(f"总会话: {meta['total_sessions']}")
+ log.info(f"总消息: {meta['total_messages']}")
+ log.info(f"最后同步: {meta['last_sync']}")
+ log.info("")
+ log.info("Token 统计 (tiktoken):")
+ total_tokens = 0
+ for source in ['codex', 'kiro', 'gemini', 'claude']:
+ t = tokens.get(source, 0)
+ if t > 0:
+ log.info(f" {source}: {t:,}")
+ total_tokens += t
+ log.info(f" 总计: {total_tokens:,}")
+
+def cmd_search(keyword: str, source: str = None):
+ log = get_logger()
+ results = storage.search(keyword, source)
+ log.info(f"搜索 '{keyword}' 找到 {len(results)} 个会话:")
+ for r in results[:20]:
+ log.info(f" [{r['source']}] {r['session_id']} - {r['cwd'] or 'N/A'}")
+
+def cmd_export(fmt: str, source: str = None, output: str = None):
+ log = get_logger()
+ if not output:
+ output = os.path.join(CONFIG["output_dir"], f"export.{fmt}")
+
+ if fmt == 'json':
+ count = storage.export_json(output, source)
+ else:
+ count = storage.export_csv(output, source)
+
+ log.info(f"✓ 导出 {count} 条到 {output}")
+
+def print_token_stats():
+ log = get_logger()
+ tokens = storage.get_token_stats()
+ log.info("")
+ log.info("=== Token 统计 (tiktoken) ===")
+ total = 0
+ for source in ['codex', 'kiro', 'gemini', 'claude']:
+ t = tokens.get(source, 0)
+ if t > 0:
+ log.info(f" {source}: {t:,} tokens")
+ total += t
+ log.info(f" 总计: {total:,} tokens")
+
+def update_cli_meta(cli: str):
+ stats = storage.get_cli_stats(cli)
+ path = CONFIG.get(f"{cli}_paths", [""])[0] if CONFIG.get(f"{cli}_paths") else ""
+ storage.update_cli_meta(cli, path, stats['sessions'], stats['messages'], stats['tokens'])
+
+def process(parser, paths) -> tuple:
+ log = get_logger()
+ added, updated, skipped, errors = 0, 0, 0, 0
+ for f in parser.find_files(paths):
+ try:
+ s = parser.parse_file(f)
+ file_path = os.path.abspath(s.file_path)
+ db_mtime = storage.get_file_mtime(file_path)
+ file_mtime = int(s.file_mtime)
+ if db_mtime == 0:
+ storage.upsert_session(s.session_id, s.source, file_path, s.cwd, s.messages, file_mtime)
+ added += 1
+ elif file_mtime > db_mtime:
+ storage.upsert_session(s.session_id, s.source, file_path, s.cwd, s.messages, file_mtime)
+ updated += 1
+ else:
+ skipped += 1
+ except Exception as e:
+ log.debug(f"解析失败 {f}: {e}")
+ errors += 1
+ return added, updated, skipped, errors
+
+def process_multi(parser, paths, source: str) -> tuple:
+ """处理返回多个会话的解析器（Claude/Kiro）"""
+ log = get_logger()
+ added, updated, skipped, errors = 0, 0, 0, 0
+ for f in parser.find_files(paths):
+ try:
+ for s in parser.parse_file(f):
+ file_path = s.file_path # kiro:xxx 或 claude:xxx
+ db_mtime = storage.get_file_mtime(file_path)
+ file_mtime = int(s.file_mtime)
+ if db_mtime == 0:
+ storage.upsert_session(s.session_id, s.source, file_path, s.cwd, s.messages, file_mtime)
+ added += 1
+ elif file_mtime > db_mtime:
+ storage.upsert_session(s.session_id, s.source, file_path, s.cwd, s.messages, file_mtime)
+ updated += 1
+ else:
+ skipped += 1
+ except Exception as e:
+ log.debug(f"解析失败 {f}: {e}")
+ errors += 1
+ return added, updated, skipped, errors
+
+if __name__ == '__main__':
+ main()
diff --git a/libs/external/chat-vault/src/parsers/__init__.py b/libs/external/chat-vault/src/parsers/__init__.py
new file mode 100644
index 0000000..7f6e56c
--- /dev/null
+++ b/libs/external/chat-vault/src/parsers/__init__.py
@@ -0,0 +1,7 @@
+from .codex import CodexParser
+from .gemini import GeminiParser
+from .claude import ClaudeParser
+from .kiro import KiroParser
+from .base import SessionData
+
+__all__ = ["CodexParser", "GeminiParser", "ClaudeParser", "KiroParser", "SessionData"]
diff --git a/libs/external/chat-vault/src/parsers/base.py b/libs/external/chat-vault/src/parsers/base.py
new file mode 100644
index 0000000..feb2d63
--- /dev/null
+++ b/libs/external/chat-vault/src/parsers/base.py
@@ -0,0 +1,24 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+from abc import ABC, abstractmethod
+from dataclasses import dataclass, field
+from typing import List, Dict
+
+@dataclass
+class SessionData:
+ """会话数据"""
+ session_id: str
+ source: str
+ file_path: str
+ file_mtime: float = 0
+ cwd: str = None
+ messages: List[Dict] = field(default_factory=list) # [{"time", "role", "content"}]
+
+class BaseParser(ABC):
+ @abstractmethod
+ def find_files(self, paths: list) -> list:
+ pass
+
+ @abstractmethod
+ def parse_file(self, filepath: str) -> SessionData:
+ pass
diff --git a/libs/external/chat-vault/src/parsers/claude.py b/libs/external/chat-vault/src/parsers/claude.py
new file mode 100644
index 0000000..9df8d27
--- /dev/null
+++ b/libs/external/chat-vault/src/parsers/claude.py
@@ -0,0 +1,54 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+import os
+import json
+import hashlib
+from datetime import datetime
+from collections import defaultdict
+from .base import BaseParser, SessionData
+
+class ClaudeParser(BaseParser):
+ def find_files(self, paths: list) -> list:
+ files = []
+ for base in paths:
+ history = os.path.join(base, "history.jsonl")
+ if os.path.exists(history):
+ files.append(history)
+ return files
+
+ def parse_file(self, filepath: str) -> list:
+ """返回多个 SessionData（按 project 分组）"""
+ projects = defaultdict(list)
+ file_mtime = os.path.getmtime(filepath)
+
+ with open(filepath, 'r', encoding='utf-8') as f:
+ for line in f:
+ line = line.strip()
+ if not line:
+ continue
+ data = json.loads(line)
+ content = data.get('display', '')
+ if not content:
+ continue
+
+ project = data.get('project', 'unknown')
+ ts_ms = data.get('timestamp', 0)
+ ts = datetime.fromtimestamp(ts_ms / 1000).isoformat() if ts_ms else ''
+
+ projects[project].append({
+ 'time': ts,
+ 'role': 'user',
+ 'content': content
+ })
+
+ return [
+ SessionData(
+ session_id='claude-' + hashlib.md5(proj.encode()).hexdigest()[:12],
+ source='claude',
+ file_path=filepath,
+ file_mtime=file_mtime,
+ cwd=proj,
+ messages=msgs
+ )
+ for proj, msgs in projects.items()
+ ]
diff --git a/libs/external/chat-vault/src/parsers/codex.py b/libs/external/chat-vault/src/parsers/codex.py
new file mode 100644
index 0000000..66c47d4
--- /dev/null
+++ b/libs/external/chat-vault/src/parsers/codex.py
@@ -0,0 +1,70 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+import os
+import json
+import re
+from .base import BaseParser, SessionData
+
+class CodexParser(BaseParser):
+ def __init__(self, source: str = 'codex'):
+ self.source = source
+
+ def find_files(self, paths: list) -> list:
+ files = []
+ for base in paths:
+ if not os.path.exists(base):
+ continue
+ for root, _, names in os.walk(base):
+ for f in names:
+ if f.endswith('.jsonl') and f != 'history.jsonl':
+ files.append(os.path.join(root, f))
+ return files
+
+ def _extract_id(self, filepath: str) -> str:
+ match = re.search(r'([0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12})',
+ os.path.basename(filepath))
+ return match.group(1) if match else os.path.basename(filepath).replace('.jsonl', '')
+
+ def parse_file(self, filepath: str) -> SessionData:
+ s = SessionData(
+ session_id=self._extract_id(filepath),
+ source=self.source,
+ file_path=filepath,
+ file_mtime=os.path.getmtime(filepath)
+ )
+
+ with open(filepath, 'r', encoding='utf-8', errors='ignore') as f:
+ for line in f:
+ line = line.strip()
+ if not line or line[0] != '{':
+ continue
+ try:
+ data = json.loads(line)
+ except json.JSONDecodeError:
+ continue
+
+ if data.get('type') == 'session_meta':
+ p = data.get('payload', {})
+ s.cwd = p.get('cwd')
+ s.session_id = p.get('id', s.session_id)
+ continue
+
+ if data.get('type') != 'response_item':
+ continue
+ payload = data.get('payload', {})
+ if payload.get('type') != 'message':
+ continue
+ role = payload.get('role')
+ if role not in ('user', 'assistant'):
+ continue
+
+ parts = [item.get('text', '') for item in payload.get('content', [])
+ if isinstance(item, dict) and item.get('type') in ('input_text', 'output_text', 'text')]
+ if parts:
+ s.messages.append({
+ 'time': data.get('timestamp', ''),
+ 'role': 'user' if role == 'user' else 'ai',
+ 'content': ' '.join(parts)
+ })
+
+ return s
diff --git a/libs/external/chat-vault/src/parsers/gemini.py b/libs/external/chat-vault/src/parsers/gemini.py
new file mode 100644
index 0000000..957be7f
--- /dev/null
+++ b/libs/external/chat-vault/src/parsers/gemini.py
@@ -0,0 +1,40 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+import os
+import glob
+import json
+from .base import BaseParser, SessionData
+
+class GeminiParser(BaseParser):
+ def find_files(self, paths: list) -> list:
+ files = []
+ for base in paths:
+ if os.path.exists(base):
+ files.extend(glob.glob(os.path.join(base, "*", "chats", "*.json")))
+ return files
+
+ def parse_file(self, filepath: str) -> SessionData:
+ s = SessionData(
+ session_id=os.path.basename(filepath).replace('.json', ''),
+ source='gemini',
+ file_path=filepath,
+ file_mtime=os.path.getmtime(filepath)
+ )
+
+ with open(filepath, 'r', encoding='utf-8') as f:
+ data = json.load(f)
+
+ s.session_id = data.get('sessionId', s.session_id)
+
+ for msg in data.get('messages', []):
+ if msg.get('type') not in ('user', 'gemini'):
+ continue
+ content = msg.get('content', '')
+ if content:
+ s.messages.append({
+ 'time': msg.get('timestamp', ''),
+ 'role': 'user' if msg.get('type') == 'user' else 'ai',
+ 'content': content
+ })
+
+ return s
diff --git a/libs/external/chat-vault/src/parsers/kiro.py b/libs/external/chat-vault/src/parsers/kiro.py
new file mode 100644
index 0000000..0d58bea
--- /dev/null
+++ b/libs/external/chat-vault/src/parsers/kiro.py
@@ -0,0 +1,75 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""Kiro CLI 解析器 - 从 SQLite 数据库读取"""
+import os
+import json
+import sqlite3
+import hashlib
+from datetime import datetime
+from .base import BaseParser, SessionData
+
+KIRO_DB = os.path.expanduser("~/.local/share/kiro-cli/data.sqlite3")
+
+class KiroParser(BaseParser):
+ def find_files(self, paths: list) -> list:
+ """返回数据库路径（如果存在）"""
+ if os.path.exists(KIRO_DB):
+ return [KIRO_DB]
+ return []
+
+ def parse_file(self, filepath: str) -> list:
+ """解析 Kiro SQLite 数据库，返回多个 SessionData"""
+ sessions = []
+ file_mtime = os.path.getmtime(filepath)
+
+ conn = sqlite3.connect(filepath)
+ for row in conn.execute('SELECT key, value FROM conversations'):
+ cwd, value = row
+ try:
+ data = json.loads(value)
+ except json.JSONDecodeError:
+ continue
+
+ conv_id = data.get('conversation_id', hashlib.md5(cwd.encode()).hexdigest()[:12])
+ history = data.get('history', [])
+
+ messages = []
+ for item in history:
+ # 用户消息
+ if 'user' in item:
+ user = item['user']
+ content = user.get('content', {})
+ if isinstance(content, dict) and 'Prompt' in content:
+ prompt = content['Prompt'].get('prompt', '')
+ if prompt:
+ messages.append({
+ 'time': '',
+ 'role': 'user',
+ 'content': prompt
+ })
+
+ # AI 回复
+ if 'assistant' in item:
+ assistant = item['assistant']
+ content = assistant.get('content', {})
+ if isinstance(content, dict) and 'Message' in content:
+ msg = content['Message'].get('message', '')
+ if msg:
+ messages.append({
+ 'time': '',
+ 'role': 'ai',
+ 'content': msg
+ })
+
+ if messages:
+ sessions.append(SessionData(
+ session_id=f'kiro-{conv_id[:12]}',
+ source='kiro',
+ file_path=f'kiro:{conv_id}',
+ file_mtime=file_mtime,
+ cwd=cwd,
+ messages=messages
+ ))
+
+ conn.close()
+ return sessions
diff --git a/libs/external/chat-vault/src/storage.py b/libs/external/chat-vault/src/storage.py
new file mode 100644
index 0000000..e07a9a0
--- /dev/null
+++ b/libs/external/chat-vault/src/storage.py
@@ -0,0 +1,246 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""SQLite 存储模块 - 完整版"""
+import sqlite3
+import json
+import os
+import datetime
+import tiktoken
+
+SCHEMA_VERSION = 5
+CLIS = ('codex', 'kiro', 'gemini', 'claude')
+
+_encoder = tiktoken.get_encoding("cl100k_base")
+
+def count_tokens(text: str) -> int:
+ return len(_encoder.encode(text)) if text else 0
+
+class ChatStorage:
+ def __init__(self, db_path: str):
+ self.db_path = db_path
+ os.makedirs(os.path.dirname(db_path) or '.', exist_ok=True)
+ self._init_db()
+
+ def _conn(self):
+ return sqlite3.connect(self.db_path)
+
+ def _init_db(self):
+ with self._conn() as conn:
+ conn.execute('''CREATE TABLE IF NOT EXISTS meta (key TEXT PRIMARY KEY, value TEXT)''')
+ for cli in CLIS:
+ conn.execute(f'''CREATE TABLE IF NOT EXISTS meta_{cli} (key TEXT PRIMARY KEY, value TEXT)''')
+ conn.execute('''
+ CREATE TABLE IF NOT EXISTS sessions (
+ file_path TEXT PRIMARY KEY,
+ session_id TEXT,
+ source TEXT NOT NULL,
+ cwd TEXT,
+ messages TEXT,
+ file_mtime INTEGER,
+ start_time TEXT,
+ token_count INTEGER DEFAULT 0
+ )
+ ''')
+ conn.execute('CREATE INDEX IF NOT EXISTS idx_source ON sessions(source)')
+ conn.execute('CREATE INDEX IF NOT EXISTS idx_session_id ON sessions(session_id)')
+ conn.execute('CREATE INDEX IF NOT EXISTS idx_start_time ON sessions(start_time)')
+ self._set_meta('meta', 'schema_version', str(SCHEMA_VERSION))
+
+ def _set_meta(self, table: str, key: str, value: str):
+ with self._conn() as conn:
+ conn.execute(f'INSERT OR REPLACE INTO {table} (key, value) VALUES (?, ?)', (key, value))
+
+ def _get_meta(self, table: str, key: str) -> str:
+ with self._conn() as conn:
+ row = conn.execute(f'SELECT value FROM {table} WHERE key = ?', (key,)).fetchone()
+ return row[0] if row else None
+
+ def update_cli_meta(self, cli: str, path: str, sessions: int, messages: int, tokens: int = None):
+ table = f'meta_{cli}'
+ now = datetime.datetime.now().isoformat()
+ # 顺序: path, sessions, messages, total_tokens, last_sync
+ self._set_meta(table, 'path', path)
+ self._set_meta(table, 'sessions', str(sessions))
+ self._set_meta(table, 'messages', str(messages))
+ self._set_meta(table, 'total_tokens', str(tokens or 0))
+ self._set_meta(table, 'last_sync', now)
+
+ def update_total_meta(self, sessions: int, messages: int, tokens: int = None):
+ now = datetime.datetime.now().isoformat()
+ self._set_meta('meta', 'total_sessions', str(sessions))
+ self._set_meta('meta', 'total_messages', str(messages))
+ if tokens is not None:
+ self._set_meta('meta', 'total_tokens', str(tokens))
+ self._set_meta('meta', 'last_sync', now)
+
+ def get_total_meta(self) -> dict:
+ return {
+ 'schema_version': int(self._get_meta('meta', 'schema_version') or 0),
+ 'total_sessions': int(self._get_meta('meta', 'total_sessions') or 0),
+ 'total_messages': int(self._get_meta('meta', 'total_messages') or 0),
+ 'last_sync': self._get_meta('meta', 'last_sync'),
+ }
+
+ def get_file_mtime(self, file_path: str) -> int:
+ with self._conn() as conn:
+ row = conn.execute('SELECT file_mtime FROM sessions WHERE file_path = ?', (file_path,)).fetchone()
+ return row[0] if row else 0
+
+ def upsert_session(self, session_id: str, source: str, file_path: str,
+ cwd: str, messages: list, file_mtime: int, start_time: str = None):
+ if file_path and not file_path.startswith('claude:') and not os.path.isabs(file_path):
+ file_path = os.path.abspath(file_path)
+
+ total_tokens = sum(count_tokens(msg.get('content', '')) for msg in messages)
+ if not start_time and messages:
+ start_time = messages[0].get('time')
+
+ messages_json = json.dumps(messages, ensure_ascii=False)
+ with self._conn() as conn:
+ conn.execute('''
+ INSERT INTO sessions (file_path, session_id, source, cwd, messages, file_mtime, start_time, token_count)
+ VALUES (?, ?, ?, ?, ?, ?, ?, ?)
+ ON CONFLICT(file_path) DO UPDATE SET
+ session_id=excluded.session_id, messages=excluded.messages,
+ file_mtime=excluded.file_mtime, start_time=excluded.start_time, token_count=excluded.token_count
+ ''', (file_path, session_id, source, cwd, messages_json, file_mtime, start_time, total_tokens))
+
+ def get_cli_stats(self, cli: str) -> dict:
+ with self._conn() as conn:
+ sessions = conn.execute('SELECT COUNT(*) FROM sessions WHERE source = ?', (cli,)).fetchone()[0]
+ row = conn.execute('SELECT SUM(json_array_length(messages)) FROM sessions WHERE source = ?', (cli,)).fetchone()
+ messages = row[0] or 0
+ tokens = conn.execute('SELECT SUM(token_count) FROM sessions WHERE source = ?', (cli,)).fetchone()[0] or 0
+ return {'sessions': sessions, 'messages': messages, 'tokens': tokens}
+
+ def get_total_stats(self) -> dict:
+ with self._conn() as conn:
+ sessions = conn.execute('SELECT COUNT(*) FROM sessions').fetchone()[0]
+ row = conn.execute('SELECT SUM(json_array_length(messages)) FROM sessions').fetchone()
+ messages = row[0] or 0
+ tokens = conn.execute('SELECT SUM(token_count) FROM sessions').fetchone()[0] or 0
+ return {'sessions': sessions, 'messages': messages, 'tokens': tokens}
+
+ def get_token_stats(self) -> dict:
+ with self._conn() as conn:
+ rows = conn.execute('SELECT source, SUM(token_count) FROM sessions GROUP BY source').fetchall()
+ return {r[0]: r[1] or 0 for r in rows}
+
+ # === 清理孤立记录 ===
+ def prune(self) -> dict:
+ """删除源文件已不存在的记录"""
+ removed = {'codex': 0, 'kiro': 0, 'gemini': 0, 'claude': 0}
+ with self._conn() as conn:
+ rows = conn.execute('SELECT file_path, source FROM sessions').fetchall()
+ for fp, source in rows:
+ if fp.startswith('claude:'):
+ continue # Claude 使用虚拟路径
+ if not os.path.exists(fp):
+ conn.execute('DELETE FROM sessions WHERE file_path = ?', (fp,))
+ removed[source] = removed.get(source, 0) + 1
+ return removed
+
+ # === 查询 ===
+ def search(self, keyword: str, source: str = None, limit: int = 50) -> list:
+ """搜索消息内容"""
+ sql = "SELECT file_path, session_id, source, cwd, messages, start_time FROM sessions WHERE messages LIKE ?"
+ params = [f'%{keyword}%']
+ if source:
+ sql += " AND source = ?"
+ params.append(source)
+ sql += f" ORDER BY start_time DESC LIMIT {limit}"
+
+ results = []
+ with self._conn() as conn:
+ for row in conn.execute(sql, params):
+ results.append({
+ 'file_path': row[0], 'session_id': row[1], 'source': row[2],
+ 'cwd': row[3], 'messages': json.loads(row[4]), 'start_time': row[5]
+ })
+ return results
+
+ def get_session(self, file_path: str) -> dict:
+ """获取单个会话"""
+ with self._conn() as conn:
+ row = conn.execute(
+ 'SELECT file_path, session_id, source, cwd, messages, start_time, token_count FROM sessions WHERE file_path = ?',
+ (file_path,)
+ ).fetchone()
+ if not row:
+ return None
+ return {
+ 'file_path': row[0], 'session_id': row[1], 'source': row[2], 'cwd': row[3],
+ 'messages': json.loads(row[4]), 'start_time': row[5], 'token_count': row[6]
+ }
+
+ def list_sessions(self, source: str = None, limit: int = 100, offset: int = 0) -> list:
+ """列出会话"""
+ sql = "SELECT file_path, session_id, source, cwd, start_time, token_count FROM sessions"
+ params = []
+ if source:
+ sql += " WHERE source = ?"
+ params.append(source)
+ sql += f" ORDER BY start_time DESC LIMIT {limit} OFFSET {offset}"
+
+ results = []
+ with self._conn() as conn:
+ for row in conn.execute(sql, params):
+ results.append({
+ 'file_path': row[0], 'session_id': row[1], 'source': row[2],
+ 'cwd': row[3], 'start_time': row[4], 'token_count': row[5]
+ })
+ return results
+
+ # === 导出 ===
+ def export_json(self, output_path: str, source: str = None):
+ """导出为 JSON"""
+ sql = "SELECT file_path, session_id, source, cwd, messages, start_time, token_count FROM sessions"
+ params = []
+ if source:
+ sql += " WHERE source = ?"
+ params.append(source)
+ sql += " ORDER BY start_time"
+
+ data = []
+ with self._conn() as conn:
+ for row in conn.execute(sql, params):
+ data.append({
+ 'file_path': row[0], 'session_id': row[1], 'source': row[2], 'cwd': row[3],
+ 'messages': json.loads(row[4]), 'start_time': row[5], 'token_count': row[6]
+ })
+
+ with open(output_path, 'w', encoding='utf-8') as f:
+ json.dump(data, f, ensure_ascii=False, indent=2)
+ return len(data)
+
+ def export_csv(self, output_path: str, source: str = None):
+ """导出为 CSV（扁平化消息）"""
+ import csv
+ sql = "SELECT session_id, source, cwd, messages, start_time FROM sessions"
+ params = []
+ if source:
+ sql += " WHERE source = ?"
+ params.append(source)
+ sql += " ORDER BY start_time"
+
+ count = 0
+ with open(output_path, 'w', encoding='utf-8', newline='') as f:
+ writer = csv.writer(f)
+ writer.writerow(['session_id', 'source', 'cwd', 'time', 'role', 'content'])
+ with self._conn() as conn:
+ for row in conn.execute(sql, params):
+ session_id, src, cwd, msgs_json, _ = row
+ for msg in json.loads(msgs_json):
+ writer.writerow([session_id, src, cwd, msg.get('time', ''), msg.get('role', ''), msg.get('content', '')])
+ count += 1
+ return count
+
+ # === 获取所有文件路径（用于 prune 检查） ===
+ def get_all_file_paths(self, source: str = None) -> set:
+ sql = "SELECT file_path FROM sessions"
+ params = []
+ if source:
+ sql += " WHERE source = ?"
+ params.append(source)
+ with self._conn() as conn:
+ return {row[0] for row in conn.execute(sql, params)}
diff --git a/libs/external/chat-vault/src/watcher.py b/libs/external/chat-vault/src/watcher.py
new file mode 100644
index 0000000..7ae95cb
--- /dev/null
+++ b/libs/external/chat-vault/src/watcher.py
@@ -0,0 +1,44 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""跨平台文件监控 (Linux/macOS/Windows)"""
+
+from watchdog.observers import Observer
+from watchdog.events import FileSystemEventHandler
+import time
+
+class ChatFileHandler(FileSystemEventHandler):
+ def __init__(self, callback, extensions):
+ self.callback = callback
+ self.extensions = extensions
+
+ def _check(self, event):
+ if event.is_directory:
+ return
+ path = event.src_path
+ if any(path.endswith(ext) for ext in self.extensions):
+ self.callback(path, event.event_type)
+
+ def on_created(self, event):
+ self._check(event)
+
+ def on_modified(self, event):
+ self._check(event)
+
+class ChatWatcher:
+ def __init__(self, paths: list, callback, extensions=('.jsonl', '.json')):
+ self.observer = Observer()
+ handler = ChatFileHandler(callback, extensions)
+ for path in paths:
+ self.observer.schedule(handler, path, recursive=True)
+
+ def start(self):
+ self.observer.start()
+ try:
+ while True:
+ time.sleep(1)
+ except KeyboardInterrupt:
+ self.stop()
+
+ def stop(self):
+ self.observer.stop()
+ self.observer.join()
diff --git a/libs/external/chat-vault/start.bat b/libs/external/chat-vault/start.bat
new file mode 100644
index 0000000..2da21f6
--- /dev/null
+++ b/libs/external/chat-vault/start.bat
@@ -0,0 +1,7 @@
+@echo off
+cd /d "%~dp0src"
+echo 正在启动 AI Chat Converter...
+echo 按 Ctrl+C 停止
+echo.
+python main.py --watch
+pause
diff --git a/libs/external/chat-vault/start.sh b/libs/external/chat-vault/start.sh
new file mode 100755
index 0000000..2a2aa59
--- /dev/null
+++ b/libs/external/chat-vault/start.sh
@@ -0,0 +1,7 @@
+#!/bin/bash
+# AI Chat Converter - 一键启动
+cd "$(dirname "$0")/src"
+echo "正在启动 AI Chat Converter..."
+echo "按 Ctrl+C 停止"
+echo ""
+python3 main.py --watch
diff --git a/scripts/.gitkeep b/scripts/.gitkeep
deleted file mode 100644
index e69de29..0000000