MCP Studio
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MCP StudioRun a SQL query to list all student records"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP Studio Windows 64 位便携版
完整解压 ZIP 到可写的全英文路径(例如 C:\MCP-Studio,可包含空格,不支持中文路径),双击 Start.cmd。不需要安装 Python 或 MySQL,不需要管理员权限。首次初始化数据库需要一些时间;程序会将本地生成的 HTTPS 证书加入当前 Windows 用户的信任库,然后打开 https://localhost:9440。
启动窗口保持打开。在启动窗口按 Enter 正常停止全部服务后再移动目录。不要在压缩包内直接运行,也不要强制关机或删除正在使用的数据库。
首次使用请自行注册账号,第一个注册者是管理员。模型对话需要填写自己的 API Key 并联网;本地工具可离线使用。API Key 仅保存在进程内存,重启后重新填写。
包含 SQL 查询、计算统计、文件沙箱、聊天、工具调试、账号和文件共享功能。网页在系统浏览器运行。数据库中只有项目自带的虚构课程示例;没有开发者账号、聊天、上传文件、证书或密钥。
所有数据保存在解压目录的 .env、deploy/mysql/data、deploy/caddy/data、uploads 和 logs。请勿将使用过的整个目录发给别人;分享原始 ZIP。备份时先正常停止,再复制整个目录。
固定使用本机端口 3307、2019、5000、8001–8003、9440–9443;冲突时启动器会报错,不会终止其他软件。仅监听本机;一次运行一个实例。启动失败查看 logs。
运行环境包含 Python 3.11、MySQL 8.4、Caddy 及 requirements.txt 中依赖。第三方许可随各运行时与 Python 包附带,MySQL 许可见 runtime/mysql/LICENSE。Windows 10/11 x64;其他系统不可直接运行此包。
卸载:先正常停止,然后删除解压目录。证书可在 Windows 当前用户证书管理器的“受信任的根证书颁发机构”中按本包 root.crt 指纹核对删除;不要批量删除其他 Caddy 证书。
下载
在 Releases 下载 MCP-Studio-v1.0.0-Windows-x64.zip,不要下载 GitHub 自动生成的 Source code ZIP 作为运行包。
Related MCP server: Windows Scoped Remote MCP Server
验证
在独立英文含空格目录中,使用包内 Python/MySQL/Caddy 完成首次初始化、重启、HTTPS Host 检查、三个 MCP 服务共 9 个工具发现、231 项 pytest 测试及数据库正常关闭。没有调用付费模型 API。
开发源码
源码包括全部应用与工具、数据库结构与示例、前端和便携启动器。portable/build.py 使用构建机器已有的 Python、项目虚拟环境、MySQL 和 Caddy 来组装运行包;构建输入路径可按机器修改。运行时、个人部署配置、账号、数据库与凭据不进入 Git 源码。
This server cannot be deployed
Maintenance
Related MCP Connectors
AI workspace for you, your team, and every agent. Tables, docs (images, 4K video), formulas.
Your AI builds, deploys, and runs full-stack apps on a hosted workspace created at first sign-in.
Zero-Ops deploy of a private AI workspace to your own VPS — from your AI chat. Free and open-source.
Cross-agent artifact workspace with provenance across Claude Code, Codex, Cursor, LangGraph.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceA local sandbox that provides AI agents with code execution, filesystem access, and a full GUI desktop environment for 'computer use' capabilities. It exposes tools for shell commands, multi-language code interpretation, and remote desktop automation via the Model Context Protocol.214GPL 3.0
- FlicenseNot gradedqualityBmaintenanceEnables secure remote development on Windows by allowing LLM clients to manage files, run commands, and automate browsers within sandboxed workspaces, with multi-project switching and OAuth authentication.-
- AlicenseNot gradedqualityAmaintenanceEnables AI agents to execute local commands and multi-step workflows with durable checkpoints, operator controls, and safe resume/revise capabilities. It also provides local web and YouTube research tools, with optional Windows UI automation.21 npmMIT
- AlicenseNot gradedqualityCmaintenanceEnables AI assistants to perform local development tasks on Windows by reading and editing files, running commands, and controlling browser and desktop tools, all within isolated workspaces. Supports secure remote access via an optional tunnel for ChatGPT clients.MIT