docs: sync README and docs with main langchain merge

Co-authored-by: Simpleyyt <simpleyyt@gmail.com>

Author: cursoragent

README.md

@@ -77,7 +77,7 @@ This project primarily relies on Docker for development and deployment, requirin
 - Docker Compose
 
 Model capability requirements:
-- Compatible with OpenAI interface
+- Supports LangChain chat model providers (default `openai`)
 - Support for FunctionCall
 - Support for Json Format output
 
@@ -119,6 +119,8 @@ services:
       - API_KEY=sk-xxxx
       # LLM model name
       - MODEL_NAME=gpt-4o
+      # LLM model provider
+      - MODEL_PROVIDER=openai
       # LLM temperature parameter, controls randomness
       - TEMPERATURE=0.7
       # Maximum tokens for LLM response
@@ -278,6 +280,7 @@ API_BASE=http://mockserver:8090/v1
 
 # Model configuration
 MODEL_NAME=deepseek-chat
+MODEL_PROVIDER=openai
 TEMPERATURE=0.7
 MAX_TOKENS=2000
 
@@ -363,7 +366,7 @@ All services will run in reload mode, and code changes will be automatically rel
 
 > *Note: In Debug mode, only one sandbox will be started globally*
 
-2. When dependencies change (requirements.txt or package.json), clean up and rebuild:
+2. When dependencies change (`backend/pyproject.toml` or `frontend/package.json`), clean up and rebuild:
 ```bash
 # Clean up all related resources
 ./dev.sh down -v

README_zh.md

@@ -58,7 +58,7 @@ https://github.com/user-attachments/assets/5cb2240b-0984-4db0-8818-a24f81624b04
 - Docker Compose
 
 模型能力要求：
-- 兼容OpenAI接口
+- 支持 LangChain Chat Model（默认 `openai` 提供商）
 - 支持FunctionCall
 - 支持Json Format输出
 
@@ -101,6 +101,8 @@ services:
       - API_KEY=sk-xxxx
       # LLM model name
       - MODEL_NAME=gpt-4o
+      # LLM model provider
+      - MODEL_PROVIDER=openai
       # LLM temperature parameter, controls randomness
       - TEMPERATURE=0.7
       # Maximum tokens for LLM response
@@ -279,6 +281,7 @@ API_BASE=http://mockserver:8090/v1
 
 # Model configuration
 MODEL_NAME=deepseek-chat
+MODEL_PROVIDER=openai
 TEMPERATURE=0.7
 MAX_TOKENS=2000
 
@@ -364,7 +367,7 @@ LOG_LEVEL=INFO
 
 > *注意：在 Debug 模式全局只会启动一个沙盒*
 
-2. 当依赖变化时（requirements.txt或package.json），清理并重新构建：
+2. 当依赖变化时（`backend/pyproject.toml` 或 `frontend/package.json`），清理并重新构建：
 ```bash
 # 清理所有相关资源
 ./dev.sh down -v

backend/README.md

@@ -2,7 +2,7 @@
 
 English | [中文](README_zh.md)
 
-AI Manus is an intelligent conversation agent system based on FastAPI and OpenAI API. The backend adopts Domain-Driven Design (DDD) architecture, supporting intelligent dialogue, file operations, Shell command execution, and browser automation.
+AI Manus is an intelligent conversation agent system based on FastAPI and LangChain chat models. The backend adopts Domain-Driven Design (DDD) architecture, supporting intelligent dialogue, file operations, Shell command execution, and browser automation.
 
 ## Project Architecture
 
@@ -25,9 +25,7 @@ backend/
 │   ├── infrastructure/  # Infrastructure layer: provides technical implementation
 │   └── main.py          # Application entry
 ├── Dockerfile           # Docker configuration file
-├── run.sh               # Production environment startup script
-├── dev.sh               # Development environment startup script
-├── requirements.txt     # Project dependencies
+├── pyproject.toml       # Project dependencies and metadata
 └── README.md            # Project documentation
 ```
 
@@ -45,33 +43,33 @@ backend/
 
 ## Requirements
 
-- Python 3.9+
+- Python 3.12+
 - Docker 20.10+
 - MongoDB 4.4+
 - Redis 6.0+
 
 ## Installation and Configuration
 
-1. **Create a virtual environment**:
+1. **Install uv**:
 ```bash
-python -m venv .venv
-source .venv/bin/activate
+pip install uv
 ```
 
 2. **Install dependencies**:
 ```bash
-pip install -r requirements.txt
+uv sync
 ```
 
 3. **Environment variable configuration**:
 Create a `.env` file and set the following environment variables:
 ```
 # Model provider configuration
-API_KEY=your_api_key_here                # API key for OpenAI or other model providers
-API_BASE=https://api.openai.com/v1       # Base URL for the model API, can be replaced with other model provider API addresses
+API_KEY=your_api_key_here                # API key for model providers
+API_BASE=https://api.openai.com/v1       # Base URL for model API (optional for some providers)
 
 # Model configuration
 MODEL_NAME=gpt-4o                        # Model name to use
+MODEL_PROVIDER=openai                    # Model provider for LangChain
 TEMPERATURE=0.7                          # Model temperature parameter
 MAX_TOKENS=2000                          # Maximum output tokens per model request
 
@@ -86,9 +84,11 @@ SANDBOX_TTL_MINUTES=30                   # Sandbox container time-to-live (minut
 SANDBOX_NETWORK=manus-network            # Docker network name for communication between sandbox containers
 
 # Database configuration
-MONGODB_URL=mongodb://localhost:27017    # MongoDB connection URL
+MONGODB_URI=mongodb://localhost:27017    # MongoDB connection URL
 MONGODB_DATABASE=manus                   # MongoDB database name
-REDIS_URL=redis://localhost:6379/0       # Redis connection URL
+REDIS_HOST=localhost                     # Redis host
+REDIS_PORT=6379                          # Redis port
+REDIS_DB=0                               # Redis DB index
 
 # Log configuration
 LOG_LEVEL=INFO                           # Log level, options: DEBUG, INFO, WARNING, ERROR, CRITICAL
@@ -99,7 +99,7 @@ LOG_LEVEL=INFO                           # Log level, options: DEBUG, INFO, WARN
 ### Development Environment
 ```bash
 # Start the development server (with hot reload)
-uvicorn app.main:app --host 0.0.0.0 --port 8000 --reload
+uv run uvicorn app.main:app --host 0.0.0.0 --port 8000 --reload
 ```
 
 The service will start at http://localhost:8000.

backend/README_zh.md

@@ -2,7 +2,7 @@
 
 [English](README.md) | 中文
 
-AI Manus 是一个基于 FastAPI 和 OpenAI API 的智能对话代理系统。该后端采用领域驱动设计(DDD)架构，支持智能对话、文件操作、Shell命令执行以及浏览器自动化等功能。
+AI Manus 是一个基于 FastAPI 和 LangChain Chat Model 的智能对话代理系统。该后端采用领域驱动设计(DDD)架构，支持智能对话、文件操作、Shell命令执行以及浏览器自动化等功能。
 
 ## 项目架构
 
@@ -25,9 +25,7 @@ backend/
 │   ├── infrastructure/  # 基础设施层：提供技术实现
 │   └── main.py          # 应用入口
 ├── Dockerfile           # Docker配置文件
-├── run.sh               # 生产环境启动脚本
-├── dev.sh               # 开发环境启动脚本
-├── requirements.txt     # 项目依赖
+├── pyproject.toml       # 项目依赖与元数据
 └── README.md            # 项目文档
 ```
 
@@ -45,33 +43,33 @@ backend/
 
 ## 环境要求
 
-- Python 3.9+
+- Python 3.12+
 - Docker 20.10+
 - MongoDB 4.4+
 - Redis 6.0+
 
 ## 安装配置
 
-1. **创建虚拟环境**:
+1. **安装 uv**:
 ```bash
-python -m venv .venv
-source .venv/bin/activate
+pip install uv
 ```
 
 2. **安装依赖**:
 ```bash
-pip install -r requirements.txt
+uv sync
 ```
 
 3. **环境变量配置**:
 创建 `.env` 文件并设置以下环境变量:
 ```
 # Model provider configuration
-API_KEY=your_api_key_here                # OpenAI 或其他模型供应商的 API 密钥
-API_BASE=https://api.openai.com/v1       # 模型 API 的基础 URL，可替换为其他模型供应商的 API 地址
+API_KEY=your_api_key_here                # 模型供应商 API 密钥
+API_BASE=https://api.openai.com/v1       # 模型 API 基础 URL（部分供应商可选）
 
 # Model configuration
 MODEL_NAME=gpt-4o                        # 使用的模型名称
+MODEL_PROVIDER=openai                    # LangChain 模型供应商
 TEMPERATURE=0.7                          # 模型温度参数
 MAX_TOKENS=2000                          # 模型单次请求最大输出 token 数量
 
@@ -86,9 +84,11 @@ SANDBOX_TTL_MINUTES=30                   # 沙盒容器生存时间（分钟）
 SANDBOX_NETWORK=manus-network            # Docker 网络名称，用于沙盒容器间通信
 
 # Database configuration
-MONGODB_URL=mongodb://localhost:27017    # MongoDB 连接 URL
+MONGODB_URI=mongodb://localhost:27017    # MongoDB 连接 URL
 MONGODB_DATABASE=manus                   # MongoDB 数据库名称
-REDIS_URL=redis://localhost:6379/0       # Redis 连接 URL
+REDIS_HOST=localhost                     # Redis 主机地址
+REDIS_PORT=6379                          # Redis 端口
+REDIS_DB=0                               # Redis 数据库编号
 
 # Log configuration
 LOG_LEVEL=INFO                           # 日志级别，可选: DEBUG, INFO, WARNING, ERROR, CRITICAL
@@ -99,7 +99,7 @@ LOG_LEVEL=INFO                           # 日志级别，可选: DEBUG, INFO, W
 ### 开发环境
 ```bash
 # 启动开发服务器（带热重载功能）
-uvicorn app.main:app --host 0.0.0.0 --port 8000 --reload
+uv run uvicorn app.main:app --host 0.0.0.0 --port 8000 --reload
 ```
 
 服务将在 http://localhost:8000 启动。

docs/configuration.md

@@ -7,12 +7,13 @@
 | 配置项 | 默认值 | 是否必需 | 说明 |
 |--------|--------|----------|------|
 | `API_KEY` | - | 是 | LLM 模型的 API 密钥 |
-| `API_BASE` | `http://mockserver:8090/v1` | 是 | API 基础地址，用于指定模型服务的端点 |
+| `API_BASE` | `http://mockserver:8090/v1` | 否 | API 基础地址，用于指定模型服务的端点 |
 
 ### 模型配置
 
 | 配置项 | 默认值 | 是否必需 | 说明 |
 |--------|--------|----------|------|
+| `MODEL_PROVIDER` | `openai` | 否 | 模型提供商（如 `openai`、`anthropic`、`google_genai`、`ollama`） |
 | `MODEL_NAME` | `deepseek-chat` | 是 | 要使用的模型名称 |
 | `TEMPERATURE` | `0.7` | 否 | 模型响应的随机性程度，范围 0-1 |
 | `MAX_TOKENS` | `2000` | 否 | 模型响应的最大 token 数量 |
@@ -57,7 +58,7 @@
 
 | 配置项 | 默认值 | 是否必需 | 说明 |
 |--------|--------|----------|------|
-| `SEARCH_PROVIDER` | `baidu` | 否 | 搜索引擎提供商 (`baidu`、`google` 或 `bing`) |
+| `SEARCH_PROVIDER` | `bing` | 否 | 搜索引擎提供商 (`baidu`、`google` 或 `bing`) |
 
 #### Google 搜索配置
 

docs/en/configuration.md

@@ -7,12 +7,13 @@
 | Configuration | Default Value | Required | Description |
 |---------------|---------------|----------|-------------|
 | `API_KEY` | - | Yes | API key for the LLM model |
-| `API_BASE` | `http://mockserver:8090/v1` | Yes | Base API address for specifying model service endpoint |
+| `API_BASE` | `http://mockserver:8090/v1` | No | Base API address for specifying model service endpoint |
 
 ### Model Configuration
 
 | Configuration | Default Value | Required | Description |
 |---------------|---------------|----------|-------------|
+| `MODEL_PROVIDER` | `openai` | No | Model provider (e.g. `openai`, `anthropic`, `google_genai`, `ollama`) |
 | `MODEL_NAME` | `deepseek-chat` | Yes | Name of the model to use |
 | `TEMPERATURE` | `0.7` | No | Randomness level of model responses, range 0-1 |
 | `MAX_TOKENS` | `2000` | No | Maximum number of tokens in model response |
@@ -57,7 +58,7 @@
 
 | Configuration | Default Value | Required | Description |
 |---------------|---------------|----------|-------------|
-| `SEARCH_PROVIDER` | `baidu` | No | Search engine provider (`baidu`, `google`, or `bing`) |
+| `SEARCH_PROVIDER` | `bing` | No | Search engine provider (`baidu`, `google`, or `bing`) |
 
 #### Google Search Configuration
 

docs/en/quick_start.md

@@ -9,7 +9,7 @@ This project mainly relies on Docker for development and deployment, requiring a
 
 Model capabilities required:
 
- * Compatible with OpenAI API
+ * Supports LangChain chat models (default provider is `openai`)
  * Supports Function Call
  * Supports JSON Format output
 
@@ -27,7 +27,7 @@ Install Docker Engine according to official requirements: https://docs.docker.co
 
 ## Deployment
 
-Deploy using Docker Compose, you need to modify the `API_BASE` and `API_KEY` environment variables:
+Deploy using Docker Compose: at minimum set `API_KEY`, and adjust `API_BASE` and `MODEL_PROVIDER` for your model service:
 
 <!-- docker-compose-example.yml -->
 ```yaml
@@ -61,6 +61,8 @@ services:
       - API_KEY=sk-xxxx
       # LLM model name
       - MODEL_NAME=gpt-4o
+      # LLM model provider
+      - MODEL_PROVIDER=openai
       # LLM temperature parameter, controls randomness
       - TEMPERATURE=0.7
       # Maximum tokens for LLM response

docs/quick_start.md

@@ -9,7 +9,7 @@
 
 模型能力要求：
 
- * 兼容 OpenAI 接口
+ * 支持 LangChain Chat Model（默认 `openai` 提供商）
  * 支持 FunctionCall
  * 支持 Json Format 输出
 
@@ -28,7 +28,7 @@
 
 ## 部署
 
-使用Docker Compose进行部署，需要修改`API_BASE`与`API_KEY`环境变量：
+使用 Docker Compose 进行部署，至少需要修改 `API_KEY`，并根据模型服务调整 `API_BASE` 与 `MODEL_PROVIDER`：
 
 <!-- docker-compose-example.yml -->
 ```yaml
@@ -62,6 +62,8 @@ services:
       - API_KEY=sk-xxxx
       # LLM model name
       - MODEL_NAME=gpt-4o
+      # LLM model provider
+      - MODEL_PROVIDER=openai
       # LLM temperature parameter, controls randomness
       - TEMPERATURE=0.7
       # Maximum tokens for LLM response