lanhu-mcp-collaboration

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Lanhu MCP Collaboration

Lanhu MCP 协作

Skill by ara.so — MCP Skills collection.

由 ara.so 开发的Skill —— MCP Skills 合集。

Overview

概述

Lanhu MCP Server is a Model Context Protocol server that connects AI coding assistants to the Lanhu (蓝湖) design collaboration platform. It enables automated requirements analysis, UI design extraction, team knowledge sharing, and design-to-code conversion with vision-capable AI models.

Core capabilities:

Requirements Analysis: Automatic Axure prototype extraction with 3 analysis modes (Development/Testing/Exploration)
UI Design Support: Design spec extraction with precise parameters (spacing, colors, fonts) + HTML/CSS code generation
Team Message Board: Shared knowledge base across all AI assistants, breaking IDE silos
Smart Asset Export: Automatic design slice extraction with semantic naming
Performance: Version-based caching, incremental updates, concurrent processing

Supported AI Clients: Cursor, Windsurf, Claude Code, OpenClaw, ClawBot, Trae, Cline, and any MCP-compatible tool

Lanhu MCP Server 是一款 Model Context Protocol 服务器，可将AI编码助手与蓝湖设计协作平台相连。它借助具备视觉能力的AI模型，实现自动化需求分析、UI设计提取、团队知识共享以及设计转代码功能。

核心能力：

需求分析：自动提取Axure原型，支持3种分析模式（开发/测试/探索）
UI设计支持：提取包含精确参数（间距、颜色、字体）的设计规范 + 生成HTML/CSS代码
团队留言板：跨所有AI助手共享知识库，打破IDE信息孤岛
智能资源导出：自动提取设计切片并进行语义化命名
性能优化：基于版本的缓存、增量更新、并发处理

支持的AI客户端：Cursor、Windsurf、Claude Code、OpenClaw、ClawBot、Trae、Cline，以及任何兼容MCP的工具

Installation

安装

Prerequisites

前置条件

Python 3.10+
Vision-capable AI model (Claude, GPT-4V, Gemini, Kimi, Qwen, DeepSeek)
Lanhu account with valid cookie authentication

Python 3.10+
具备视觉能力的AI模型（Claude、GPT-4V、Gemini、Kimi、Qwen、DeepSeek）
蓝湖账号及有效的Cookie认证

Quick Install (Recommended)

快速安装（推荐）

Simply ask your AI assistant:

"Help me clone and install https://github.com/dsphper/lanhu-mcp"

The AI will guide you through cloning, dependency installation, cookie configuration, and server startup.

直接向你的AI助手提问：

"Help me clone and install https://github.com/dsphper/lanhu-mcp"

AI会引导你完成克隆、依赖安装、Cookie配置和服务器启动步骤。

Manual Installation

手动安装

Option 1: Docker (Recommended)

bash

undefined

选项1：Docker（推荐）

bash

undefined

Clone repository

git clone https://github.com/dsphper/lanhu-mcp.git cd lanhu-mcp

Configure environment (interactive cookie setup)

bash setup-env.sh # Linux/Mac

or

setup-env.bat # Windows

Start service

docker-compose up -d


**Option 2: Source Code**

```bash

docker-compose up -d


**选项2：源代码安装**

```bash

Clone repository

git clone https://github.com/dsphper/lanhu-mcp.git cd lanhu-mcp

One-click installation (includes cookie setup)

bash easy-install.sh # Linux/Mac

or

easy-install.bat # Windows


**Manual dependency installation:**
```bash
pip install -r requirements.txt
playwright install chromium

easy-install.bat # Windows


**手动安装依赖：**
```bash
pip install -r requirements.txt
playwright install chromium

Configuration

配置

Required: Lanhu Cookie

必填：蓝湖Cookie

Export your Lanhu cookie (obtained from browser DevTools after logging into lanhuapp.com):

bash

export LANHU_COOKIE="your_lanhu_cookie_here"

导出你的蓝湖Cookie（登录lanhuapp.com后，从浏览器开发者工具获取）：

bash

export LANHU_COOKIE="your_lanhu_cookie_here"

Optional: Feishu Webhook

可选：飞书Webhook

For team notifications and @mentions:

bash

export FEISHU_WEBHOOK_URL="https://open.feishu.cn/open-apis/bot/v2/hook/your-webhook-url"

Or edit

lanhu_mcp_server.py

python

DEFAULT_FEISHU_WEBHOOK = "https://open.feishu.cn/open-apis/bot/v2/hook/your-webhook-url"

用于团队通知和@提及：

bash

export FEISHU_WEBHOOK_URL="https://open.feishu.cn/open-apis/bot/v2/hook/your-webhook-url"

或编辑

lanhu_mcp_server.py

：

python

DEFAULT_FEISHU_WEBHOOK = "https://open.feishu.cn/open-apis/bot/v2/hook/your-webhook-url"

Optional: Server Configuration

可选：服务器配置

bash

export SERVER_HOST="0.0.0.0"
export SERVER_PORT=8000
export DATA_DIR="./data"
export HTTP_TIMEOUT=30
export VIEWPORT_WIDTH=1920
export VIEWPORT_HEIGHT=1080
export DEBUG="false"

bash

export SERVER_HOST="0.0.0.0"
export SERVER_PORT=8000
export DATA_DIR="./data"
export HTTP_TIMEOUT=30
export VIEWPORT_WIDTH=1920
export VIEWPORT_HEIGHT=1080
export DEBUG="false"

Start Server

启动服务器

Source code:

bash

python lanhu_mcp_server.py

Docker:

bash

docker-compose up -d
docker-compose logs -f  # View logs
docker-compose down     # Stop

Server runs at

http://localhost:8000/mcp

源代码方式：

bash

python lanhu_mcp_server.py

Docker方式：

bash

docker-compose up -d
docker-compose logs -f  # 查看日志
docker-compose down     # 停止服务

服务器运行地址：

http://localhost:8000/mcp

AI Client Configuration

AI客户端配置

Claude Code

claude_desktop_config.json

json

{
  "mcpServers": {
    "lanhu": {
      "type": "http",
      "url": "http://localhost:8000/mcp?role=Developer&name=YourName"
    }
  }
}

在

claude_desktop_config.json

中：

json

{
  "mcpServers": {
    "lanhu": {
      "type": "http",
      "url": "http://localhost:8000/mcp?role=Developer&name=YourName"
    }
  }
}

Cursor / Windsurf / Others

Cursor / Windsurf / 其他工具

In MCP settings:

json

{
  "mcpServers": {
    "lanhu": {
      "url": "http://localhost:8000/mcp?role=Developer&name=YourName"
    }
  }
}

URL Parameters:

```
role
```
: User role (Developer/Frontend/Backend/Tester/Product)
```
name
```
: Username for collaboration tracking and @mentions (use English to avoid encoding issues)

在MCP设置中：

json

{
  "mcpServers": {
    "lanhu": {
      "url": "http://localhost:8000/mcp?role=Developer&name=YourName"
    }
  }
}

URL参数说明：

```
role
```
: 用户角色（Developer/Frontend/Backend/Tester/Product）
```
name
```
: 用于协作追踪和@提及的用户名（建议使用英文避免编码问题）

Key Tools & Commands

核心工具与命令

1. Requirements Analysis

1. 需求分析

Tool:

analyze_requirements_document

Analyze Axure prototypes with AI-powered extraction:

python

undefined

工具：

analyze_requirements_document

借助AI提取功能分析Axure原型：

python

undefined

User prompt example:

用户提示示例：

"Please analyze this requirements document using MCP: https://lanhuapp.com/web/#/item/project/product?tid=xxx&pid=xxx&docId=xxx"

The AI will call:

AI会调用：

analyze_requirements_document( url="https://lanhuapp.com/web/#/item/project/product?tid=xxx&pid=xxx&docId=xxx", mode="development" # Options: development | testing | exploration )


**Analysis Modes:**

- **`development`**: Detailed field rules, business logic, global flowcharts
- **`testing`**: Test scenarios, test cases, boundary values, validation rules
- **`exploration`**: Core feature overview, module dependencies, review points

**Four-Stage Workflow:**
1. Global text scanning (establish overall understanding)
2. Grouped detailed analysis (based on selected mode)
3. Reverse validation (ensure zero omissions)
4. Generate deliverables (requirements doc/test plan/review slides)

analyze_requirements_document( url="https://lanhuapp.com/web/#/item/project/product?tid=xxx&pid=xxx&docId=xxx", mode="development" # 可选值：development | testing | exploration )


**分析模式：**

- **`development`**: 详细字段规则、业务逻辑、全局流程图
- **`testing`**: 测试场景、测试用例、边界值、验证规则
- **`exploration`**: 核心功能概述、模块依赖、评审要点

**四阶段工作流：**
1. 全局文本扫描（建立整体认知）
2. 分组详细分析（基于所选模式）
3. 反向验证（确保无遗漏）
4. 生成交付物（需求文档/测试计划/评审幻灯片）

2. UI Design Analysis

2. UI设计分析

Tool:

view_design_document

Extract design specs with precise parameters and generated code:

python

undefined

工具：

view_design_document

提取包含精确参数的设计规范并生成代码：

python

undefined

User prompt:

用户提示：

"Please view this design document using MCP: https://lanhuapp.com/web/#/item/project/stage?tid=xxx&pid=xxx"

Returns:

返回内容：

- Design image previews

- 设计图片预览

- Precise parameters (dimensions, spacing, colors, fonts)

- 精确参数（尺寸、间距、颜色、字体）

- HTML + CSS code conversion

- HTML + CSS代码转换结果


**Output includes:**
- Component dimensions and spacing
- Color values (HEX/RGB)
- Font sizes and weights
- Auto-generated HTML/CSS code matching Lanhu's native export


**输出包含：**
- 组件尺寸与间距
- 颜色值（HEX/RGB）
- 字体大小与权重
- 匹配蓝湖原生导出的自动生成HTML/CSS代码

3. Design Asset Export

3. 设计资源导出

Tool:

export_design_slices

Download design slices with semantic naming:

python

undefined

工具：

export_design_slices

下载带语义化命名的设计切片：

python

undefined

User prompt:

用户提示：

"Export all design slices from this Lanhu page"

export_design_slices( design_url="https://lanhuapp.com/web/#/item/project/stage?tid=xxx&pid=xxx", output_dir="./assets" )


**Features:**
- Automatic slice detection
- Semantic file naming based on layer paths
- Organized folder structure
- Supports PNG, SVG, and other formats

"Export all design slices from this Lanhu page"

export_design_slices( design_url="https://lanhuapp.com/web/#/item/project/stage?tid=xxx&pid=xxx", output_dir="./assets" )


**特性：**
- 自动检测切片
- 基于图层路径的语义化文件名
- 结构化文件夹目录
- 支持PNG、SVG等格式

4. Team Message Board

4. 团队留言板

Tools:

create_message

list_messages

search_messages

Share knowledge and context across all team AI assistants:

python

undefined

工具：

create_message

list_messages

search_messages

跨所有团队AI助手共享知识与上下文：

python

undefined

Create knowledge entry

创建知识库条目

create_message( project_url="https://lanhuapp.com/web/#/item/project/...", content="User authentication requires OAuth2 flow with refresh token rotation", message_type="knowledge", # Options: knowledge | task | question | experience tags=["auth", "security", "backend"] )

create_message( project_url="https://lanhuapp.com/web/#/item/project/...", content="User authentication requires OAuth2 flow with refresh token rotation", message_type="knowledge", # 可选值：knowledge | task | question | experience tags=["auth", "security", "backend"] )

Search team knowledge

搜索团队知识

search_messages( project_url="https://lanhuapp.com/web/#/item/project/...", keyword="authentication", message_type="knowledge" )

@mention team member (triggers Feishu notification)

@提及团队成员（触发飞书通知）

create_message( project_url="https://lanhuapp.com/web/#/item/project/...", content="@zhangsan Please review the API error handling logic", message_type="task", mentioned_users=["zhangsan"] )


**Message Types:**
- **`knowledge`**: Permanent knowledge base entries (pitfalls, best practices)
- **`task`**: Task assignments with @mention support
- **`question`**: Questions for team discussion
- **`experience`**: Lessons learned and implementation notes

create_message( project_url="https://lanhuapp.com/web/#/item/project/...", content="@zhangsan Please review the API error handling logic", message_type="task", mentioned_users=["zhangsan"] )


**消息类型：**
- **`knowledge`**: 永久知识库条目（陷阱、最佳实践）
- **`task`**: 支持@提及的任务分配
- **`question`**: 用于团队讨论的问题
- **`experience`**: 经验总结与实现笔记

Common Patterns

常见使用模式

Pattern 1: Full Requirements Analysis Workflow

模式1：完整需求分析工作流

python

undefined

python

undefined

Step 1: User provides Lanhu URL

步骤1：用户提供蓝湖URL

user: "Analyze requirements: https://lanhuapp.com/web/#/item/project/product?tid=123&pid=456&docId=789"

Step 2: AI calls analyze_requirements_document

步骤2：AI调用analyze_requirements_document

result = analyze_requirements_document( url="https://lanhuapp.com/web/#/item/project/product?tid=123&pid=456&docId=789", mode="development" )

Step 3: AI processes four-stage analysis

步骤3：AI执行四阶段分析

- Stage 1: Scans all pages and extracts text

- 阶段1：扫描所有页面并提取文本

- Stage 2: Groups pages and analyzes by business modules

- 阶段2：分组页面并按业务模块分析

- Stage 3: Reverse validates for missing items

- 阶段3：反向验证检查遗漏项

- Stage 4: Generates structured requirements document

- 阶段4：生成结构化需求文档

Step 4: Save insights to team knowledge base

步骤4：将洞察保存到团队知识库

create_message( project_url="https://lanhuapp.com/web/#/item/project/product?tid=123&pid=456", content="Key finding: User role permissions require cascading delete logic", message_type="knowledge", tags=["permissions", "database"] )

undefined

undefined

Pattern 2: Design-to-Code Implementation

模式2：设计转代码实现

python

undefined

python

undefined

Step 1: View design and get parameters

步骤1：查看设计并获取参数

user: "Implement this design: https://lanhuapp.com/web/#/item/project/stage?tid=123&pid=456"

design_data = view_design_document( url="https://lanhuapp.com/web/#/item/project/stage?tid=123&pid=456" )

user: "Implement this design: https://lanhuapp.com/web/#/item/project/stage?tid=123&pid=456"

design_data = view_design_document( url="https://lanhuapp.com/web/#/item/project/stage?tid=123&pid=456" )

Step 2: Extract design parameters

步骤2：提取设计参数

Returns:

返回内容：

{

"preview_image": "base64_image_data",

"parameters": {

"width": "375px",

"height": "812px",

"spacing": {"top": "20px", "left": "16px"},

"colors": {"primary": "#1677FF", "text": "#333333"},

"fonts": {"title": "16px/bold", "body": "14px/regular"}

},

"html_css": "<div class='container'>...</div>\n<style>...</style>"

}

Step 3: Export required assets

步骤3：导出所需资源

export_design_slices( design_url="https://lanhuapp.com/web/#/item/project/stage?tid=123&pid=456", output_dir="./src/assets/images" )

Step 4: AI generates implementation code using parameters + HTML/CSS reference

步骤4：AI结合参数与HTML/CSS参考生成实现代码

undefined

undefined

Pattern 3: Team Collaboration Tracking

模式3：团队协作追踪

python

undefined

python

undefined

Developer A's AI analyzes requirements

开发者A的AI分析需求

analyze_requirements_document( url="https://lanhuapp.com/web/#/item/project/product?tid=123&pid=456&docId=789", mode="development" )

Save analysis results

保存分析结果

create_message( project_url="https://lanhuapp.com/web/#/item/project/product?tid=123&pid=456", content="Requirements analysis complete. 5 core modules identified: User, Product, Order, Payment, Notification", message_type="knowledge", tags=["requirements", "architecture"] )

Tester B's AI searches team knowledge

测试人员B的AI搜索团队知识

messages = search_messages( project_url="https://lanhuapp.com/web/#/item/project/product?tid=123&pid=456", keyword="requirements analysis", message_type="knowledge" )

Returns Developer A's analysis — no duplicate work!

返回开发者A的分析结果 —— 无需重复工作！

Tester B's AI now performs test-focused analysis

测试人员B的AI现在执行测试聚焦的分析

analyze_requirements_document( url="https://lanhuapp.com/web/#/item/project/product?tid=123&pid=456&docId=789", mode="testing" )

undefined

analyze_requirements_document( url="https://lanhuapp.com/web/#/item/project/product?tid=123&pid=456&docId=789", mode="testing" )

undefined

Pattern 4: Environment Variable Best Practices

模式4：环境变量最佳实践

python

undefined

python

undefined

Never hardcode secrets

切勿硬编码敏感信息

❌ BAD:

❌ 错误示例:

LANHU_COOKIE = "abc123..." FEISHU_WEBHOOK = "https://open.feishu.cn/open-apis/bot/v2/hook/xxx"

✅ GOOD: Use environment variables

✅ 正确示例: 使用环境变量

import os

LANHU_COOKIE = os.getenv("LANHU_COOKIE") FEISHU_WEBHOOK = os.getenv("FEISHU_WEBHOOK_URL")

import os

LANHU_COOKIE = os.getenv("LANHU_COOKIE") FEISHU_WEBHOOK = os.getenv("FEISHU_WEBHOOK_URL")

Validate required config

验证必填配置

if not LANHU_COOKIE: raise ValueError("LANHU_COOKIE environment variable is required")

undefined

if not LANHU_COOKIE: raise ValueError("LANHU_COOKIE environment variable is required")

undefined

Troubleshooting

故障排除

Issue: "No vision-capable model detected"

问题："No vision-capable model detected"

Cause: Using text-only AI model (e.g., GPT-3.5, Claude Instant)

Solution: Switch to vision-capable model:

Claude 3+ (Sonnet, Opus)
GPT-4V, GPT-4o
Gemini Pro Vision
Kimi, Qwen-VL, DeepSeek-VL

原因： 使用纯文本AI模型（如GPT-3.5、Claude Instant）

解决方案： 切换到具备视觉能力的模型：

Claude 3+（Sonnet、Opus）
GPT-4V、GPT-4o
Gemini Pro Vision
Kimi、Qwen-VL、DeepSeek-VL

Issue: "Cookie authentication failed"

问题："Cookie authentication failed"

Cause: Invalid or expired Lanhu cookie

Solution:

Login to https://lanhuapp.com in browser
Open DevTools → Network tab
Find any API request to lanhuapp.com
Copy full
```
Cookie
```
header value
Update
```
LANHU_COOKIE
```
environment variable
Restart server

原因： 蓝湖Cookie无效或已过期

解决方案：

在浏览器登录https://lanhuapp.com
打开开发者工具 → 网络标签
找到任意lanhuapp.com的API请求
复制完整的
```
Cookie
```
头值
更新
```
LANHU_COOKIE
```
环境变量
重启服务器

Issue: "Design-to-code conversion unavailable"

问题："Design-to-code conversion unavailable"

Cause: Design file uploaded with outdated Lanhu plugin

Solution:

Ask UI designer to update Lanhu plugin (Figma/Sketch/Adobe XD)
Re-upload design file
Retry design analysis

原因： 使用过时的蓝湖插件上传设计文件

解决方案：

请UI设计师更新蓝湖插件（Figma/Sketch/Adobe XD）
重新上传设计文件
重试设计分析

Issue: "Message board not syncing across AI assistants"

问题："Message board not syncing across AI assistants"

Cause: Different MCP server instances or cache issues

Solution:

Ensure all AI clients connect to same MCP server URL
Verify
```
project_url
```
is identical across calls
Clear cache:
```
rm -rf ./data/cache/*
```
Restart MCP server

原因： 不同的MCP服务器实例或缓存问题

解决方案：

确保所有AI客户端连接到同一个MCP服务器URL
验证所有调用的
```
project_url
```
完全一致
清除缓存：
```
rm -rf ./data/cache/*
```
重启MCP服务器

Issue: Docker container fails to start

问题：Docker容器启动失败

Cause: Port conflict or missing environment variables

Solution:

bash

undefined

原因： 端口冲突或缺少环境变量

解决方案：

bash

undefined

Check port availability

检查端口可用性

lsof -i :8000

Verify environment variables

验证环境变量

docker-compose config

Check logs

查看日志

docker-compose logs lanhu-mcp

Restart with clean state

清理状态后重启

docker-compose down -v docker-compose up -d

undefined

docker-compose down -v docker-compose up -d

undefined

Issue: Slow requirements analysis

问题：需求分析速度慢

Cause: Large prototype with many pages, no caching

Solution:

Enable version-based caching (automatic)
Use
```
exploration
```
mode for quick overview

Increase concurrent processing:

bash

export HTTP_TIMEOUT=60
export VIEWPORT_WIDTH=1920
export VIEWPORT_HEIGHT=1080

Subsequent analyses will use cached data (much faster)

原因： 原型页面数量多，未启用缓存

解决方案：

启用基于版本的缓存（自动生效）
使用
```
exploration
```
模式快速概览

增加并发处理能力：

bash

export HTTP_TIMEOUT=60
export VIEWPORT_WIDTH=1920
export VIEWPORT_HEIGHT=1080

后续分析将使用缓存数据（速度大幅提升）

Advanced Usage

高级用法

Custom Analysis Modes

自定义分析模式

Modify

lanhu_mcp_server.py

to add custom analysis perspectives:

python

ANALYSIS_MODES = {
    "development": "Developer perspective with detailed field rules",
    "testing": "QA perspective with test cases and validation",
    "exploration": "Quick overview for stakeholder review",
    "security": "Security-focused analysis for audit"  # Custom mode
}

修改

lanhu_mcp_server.py

添加自定义分析视角：

python

ANALYSIS_MODES = {
    "development": "Developer perspective with detailed field rules",
    "testing": "QA perspective with test cases and validation",
    "exploration": "Quick overview for stakeholder review",
    "security": "Security-focused analysis for audit"  # 自定义模式
}

Feishu User ID Mapping

飞书用户ID映射

Enable @mention notifications by updating

FEISHU_USER_ID_MAP

python

FEISHU_USER_ID_MAP = {
    "zhangsan": "ou_1234567890abcdef",
    "lisi": "ou_abcdef1234567890",
    # Add your team members
}

更新

FEISHU_USER_ID_MAP

启用@提及通知：

python

FEISHU_USER_ID_MAP = {
    "zhangsan": "ou_1234567890abcdef",
    "lisi": "ou_abcdef1234567890",
    # 添加团队成员
}

Performance Tuning

性能调优

bash

undefined

bash

undefined

Increase concurrent downloads

增加并发下载超时时间

export HTTP_TIMEOUT=60

Larger viewport for high-res screenshots

更大视窗用于高分辨率截图

export VIEWPORT_WIDTH=2560 export VIEWPORT_HEIGHT=1440

Enable debug logging

启用调试日志

export DEBUG="true"

undefined

export DEBUG="true"

undefined

Integration Examples

集成示例

Cursor AI Integration

Cursor AI集成

typescript

// In Cursor, add to .cursor/mcp.json
{
  "mcpServers": {
    "lanhu": {
      "url": "http://localhost:8000/mcp?role=Frontend&name=Alice"
    }
  }
}

// Then prompt:
// "Using Lanhu MCP, analyze the design at https://lanhuapp.com/... 
//  and generate React components with Tailwind CSS"

typescript

// 在Cursor中，添加到 .cursor/mcp.json
{
  "mcpServers": {
    "lanhu": {
      "url": "http://localhost:8000/mcp?role=Frontend&name=Alice"
    }
  }
}

// 然后输入提示：
// "Using Lanhu MCP, analyze the design at https://lanhuapp.com/... 
//  and generate React components with Tailwind CSS"

Windsurf Cascade Integration

Windsurf Cascade集成

json

// In Windsurf settings
{
  "mcp": {
    "servers": {
      "lanhu": {
        "url": "http://localhost:8000/mcp?role=Fullstack&name=Bob"
      }
    }
  }
}

json

// 在Windsurf设置中
{
  "mcp": {
    "servers": {
      "lanhu": {
        "url": "http://localhost:8000/mcp?role=Fullstack&name=Bob"
      }
    }
  }
}

Claude Code Integration

Claude Code集成

json

// In claude_desktop_config.json
{
  "mcpServers": {
    "lanhu": {
      "type": "http",
      "url": "http://localhost:8000/mcp?role=Backend&name=Charlie"
    }
  }
}

json

// 在claude_desktop_config.json中
{
  "mcpServers": {
    "lanhu": {
      "type": "http",
      "url": "http://localhost:8000/mcp?role=Backend&name=Charlie"
    }
  }
}

Additional Resources

额外资源

Cookie Setup Guide:
```
GET-COOKIE-TUTORIAL.md
```
AI Installation Guide:
```
ai-install-guide.md
```
Docker Deployment:
```
DEPLOY.md
```
Contributing:
```
CONTRIBUTING.md
```
GitHub: https://github.com/dsphper/lanhu-mcp
MCP Protocol: https://modelcontextprotocol.io/

Cookie设置指南：
```
GET-COOKIE-TUTORIAL.md
```
AI安装指南：
```
ai-install-guide.md
```
Docker部署：
```
DEPLOY.md
```
贡献指南：
```
CONTRIBUTING.md
```
GitHub：https://github.com/dsphper/lanhu-mcp
MCP协议：https://modelcontextprotocol.io/

License

许可证

MIT License - See

LICENSE

file for details.

MIT License - 详见

LICENSE

文件。