Cover image
Try Now
2025-04-14

使用script_starter.py将文件组织到正确结构化项目中的MCP服务器

3 years

Works with Finder

0

Github Watches

0

Github Forks

0

Github Stars

CodePortal MCP Server

This server helps you organize code files into properly structured projects and provides direct access to AI models like OpenAI and Google.

Features

  • Project Organization - Structure projects using your existing Script_starter.py
  • Security First - All settings are local and secure
  • API Key Management - Store your OpenAI and Google keys securely
  • Direct AI Access - Generate text from OpenAI or Google directly on your computer
  • Auto-shutdown - Server closes after 15 minutes to keep your system clean

Getting Started

  1. Clone this repository to your computer

    git clone https://github.com/TylerALofall/code-portal-mcp.git
    cd code-portal-mcp
    
  2. Double-click start_server.bat to start the server

  3. Open your browser to http://localhost:8001/setup

    • Complete the initial setup
    • Configure your Script_starter.py paths

Using the AI Features

  1. Visit http://localhost:8001/ai/ui after starting the server
  2. Add your API keys in the "API Keys" tab:
    • OpenAI key (starts with sk-)
    • Google Generative AI key (starts with AIza)
  3. Use the "Generate Text" tab to:
    • Choose your provider (OpenAI or Google)
    • Enter your prompt
    • Get AI responses directly on your computer

Project Management

The server can create properly organized projects from multiple files:

import requests

# Files to organize
files = {
    "main.py": "# Your Python code here",
    "utils.py": "# Utility functions",
    "README.md": "# Project Documentation"
}

# Send to CodePortal
response = requests.post(
    "http://localhost:8001/createProject",
    json={
        "project_path": "C:/Users/tyler/Projects/my_project",
        "project_name": "My Project",
        "files": files
    }
)

print(response.json())

Security

  • All API keys are stored locally on your computer
  • The server only runs when you start it
  • Auto-shutdown after 15 minutes
  • Access restricted to localhost by default

For Advanced Users

You can enable external access by:

  1. Going to http://localhost:8001/setup
  2. Enabling "Public API Mode"
  3. Setting a secure API key

This allows you to send files from other computers or AI services.

Using with Other AI Models

See external_api.md for instructions on using with Claude or other external models.

相关推荐

  • av
  • 毫不费力地使用一个命令运行LLM后端,API,前端和服务。

  • 1Panel-dev
  • 🔥1Panel提供了直观的Web接口和MCP服务器,用于在Linux服务器上管理网站,文件,容器,数据库和LLMS。

  • WangRongsheng
  • 🧑‍🚀 llm 资料总结(数据处理、模型训练、模型部署、 o1 模型、mcp 、小语言模型、视觉语言模型)|摘要世界上最好的LLM资源。

  • Byaidu
  • PDF科学纸翻译带有保留格式的pdf -基于ai完整保留排版的pdf文档全文双语翻译

  • rulego
  • ⛓️Rulego是一种轻巧,高性能,嵌入式,下一代组件编排规则引擎框架。

  • sigoden
  • 使用普通的bash/javascript/python函数轻松创建LLM工具和代理。

  • hkr04
  • 轻巧的C ++ MCP(模型上下文协议)SDK

  • RockChinQ
  • 😎简单易用、🧩丰富生态 -大模型原生即时通信机器人平台| 适配QQ / 微信(企业微信、个人微信) /飞书 /钉钉 / discord / telegram / slack等平台| 支持chatgpt,deepseek,dify,claude,基于LLM的即时消息机器人平台,支持Discord,Telegram,微信,Lark,Dingtalk,QQ,Slack

  • dmayboroda
  • 带有可配置容器的本地对话抹布

  • paulwing
  • 使用MCP服务创建的测试存储库

    Reviews

    4 (1)
    Avatar
    user_kxtKITG8
    2025-04-24

    I've been a dedicated user of code-portal-mcp and it has significantly streamlined my coding workflows. TylerALofall has done an excellent job in making an efficient and user-friendly product. The comprehensive features and intuitive interface have made my tasks much easier. Highly recommend to all developers!