mcpField
A

Ai Develop Assistant

Local
github.com/jieshao11AI & Agents131Updated 69w ago

About Ai Develop Assistant

AI Develop Assistant is an MCP server that helps AI developers clarify requirements, design modules and draft technical architecture. It tracks requirement branches such as goals, features, tech preferences and UI design, and stores results for later export.

Key features

  1. 01Guided requirement clarification
  2. 02Branch-based requirement status tracking
  3. 03Architecture generation from requirements
  4. 04Persistent storage of analysis results
  5. 05Export of complete project documents

Use cases

  1. 01Scoping requirements for a new AI project
  2. 02Drafting module breakdown and technical architecture
  3. 03Exporting requirement and architecture documents for review

MCP support

This is an MCP server that exposes tools such as requirement_clarifier and export_final_document to MCP clients for requirement analysis and architecture design.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"mcpforStudyAIDevelop": {
"command": "uvx",
"args": [
"ai-develop-assistant@latest"
],
"env": {
"MCP_STORAGE_DIR": "<MCP_STORAGE_DIR>"
}
}
}
}
Run commanduvx ai-develop-assistant@latest
Environment variables to set
MCP_STORAGE_DIR

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
pypiai-develop-assistant@latest

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

🚀 MCP AI开发助手

协助AI开发者进行智能化需求分析与架构设计的MCP工具

✨ 核心特性

  • 智能需求澄清: 自动识别项目类型,生成针对性问题
  • 分支感知管理: 跟踪项目目标、功能设计、技术偏好、UI设计等维度
  • 架构自动生成: 基于完整需求生成技术架构方案
  • 持久化存储: 自动保存分析结果,支持导出文档

📁 快速配置

旧版本配置
  1. 克隆代码

    git clone https://github.com/jiemobasixiangcai/ai-develop-assistant.git
    
  2. 推荐虚拟环境

    python -m venv venv
    source venv/bin/activate  # Unix/Linux/MacOS
    venv\Scripts\activate  # Windows
    
  3. 安装依赖

    pip install -r requirements.txt
    
  4. 配置文件位置

    Windows: %APPDATA%\Claude\claude_desktop_config.json
    macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
    
  5. 添加配置

    {
      "mcpServers": {
        "ai-develop-assistant": {
          "command": "python",
          "args": ["path/to/AIDevlopStudy.py"],
          "env": {
            "MCP_STORAGE_DIR": "./mcp_data"
          }
        }
      }
    }
    
  6. 重启Claude Desktop

新版本配置
🔧 核心工具
  1. start_new_project - 开始新项目
  2. create_requirement_blueprint - 创建需求蓝图
  3. requirement_clarifier - 获取需求澄清提示
  4. save_clarification_tasks - 保存澄清任务
  5. update_branch_status - 更新分支状态
  6. requirement_manager - 需求文档管理器
  7. check_architecture_prerequisites - 检查架构前置条件
  8. get_architecture_design_prompt - 获取架构设计提示
  9. save_generated_architecture - 保存生成的架构设计
  10. export_final_document - 导出完整文档
  11. view_requirements_status - 查看需求状态
配置(远程直连复制到你的工具中,将MCP_STORAGE_DIR替换为你的本地目录)
{
  "mcpServers": {
    "ai-develop-assistant": {
      "command": "uvx",
      "args": ["ai-develop-assistant@latest"],
      "env": {
        "MCP_STORAGE_DIR": "/path/to/your/storage"
      }
    }
  }
}

🎯 使用流程

基本步骤
  1. 需求澄清

    requirement_clarifier("我要做一个在线教育平台")
    
  2. 需求管理

    requirement_manager("目标用户:学生和教师", "项目概述")
    
  3. 查看状态

    view_requirements_status()
    
  4. 架构设计

    architecture_designer("在线教育平台架构")
    
  5. 导出文档

    export_final_document()
    

🚀 开始使用

快速上手
  1. 配置Claude Desktop (参考上面的配置方法)
  2. 重启Claude Desktop
  3. 开始智能需求分析:
    requirement_clarifier("描述你的项目想法")
    
  4. 跟随AI的智能引导,逐步完善各个需求分支
  5. 导出完整文档:
    export_final_document()
    
最佳实践
  • 💬 信任AI的分支管理:让AI引导你完成所有需求分支
  • 🎯 明确表达偏好:对技术选型、UI风格等明确表达偏好
  • 📊 定期查看状态:使用 view_requirements_status 了解进度
  • 🤖 适当授权AI:对不确定的部分可以说"用常规方案"

🎯 现在您拥有了一个真正智能的AI开发助手,它会记住每个细节,引导您完成完整的需求分析!

💬 交流群

交流群
交流群

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.