MCP.so
登录

MCP-Mealprep

@JoshuaRL

关于 MCP-Mealprep

This project takes a number of MCP servers from GitHub locations, packages them together with this repo's GHCR container, and launches them with docker-compose to run as a stack for ML/AI resources.

基本信息

分类

版本控制

许可证

Apache-2.0

运行时

dockerfile

传输方式

stdio

发布者

JoshuaRL

配置

暂无标准配置

该服务器的 README 中没有可解析的 MCP 配置块,请前往代码仓库查看安装说明。

代码仓库

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is MCP-Mealprep?

MCP-Mealprep is a Docker-based project that packages multiple Model Context Protocol (MCP) servers from GitHub into a single stack using supergateway and optionally mcpo. It is intended for developers and AI enthusiasts who want to quickly deploy a collection of MCP‑compatible services — such as web search, GitHub, Slack, and database tools — for use with LLM agents and external applications like OpenWebUI, n8n, or Claude.

How to use MCP-Mealprep?

Deploy the stack using Portainer (upload the docker-compose.yml or paste it into the web editor, configure environment variables, and click “Deploy”) or directly via docker-compose. Each server container runs a startup script that exposes the MCP server on SSE; adding USE_MCPO=true in the environment will also create an OpenAPI‑compatible HTTP endpoint for each server. Connect your MCP client to the IP/domain and port of the running stack.

Key features of MCP-Mealprep

  • Curated collection of MCP servers ready to deploy as a Docker stack
  • Uses supergateway for safe SSE exposure and optional mcpo for OpenAPI endpoints
  • Supports npx, uv, and pip installs inside each container
  • Includes security scanning with mcp-shield at startup
  • Easily add or remove servers by editing the docker-compose file

Use cases of MCP-Mealprep

  • Run a private set of MCP servers for local AI agent tasks
  • Expose MCP services securely to external tools like OpenWebUI, n8n, or Flowise
  • Prototype and test multiple MCP tools without manual container orchestration
  • Bundle custom MCP servers with a consistent deployment pattern

FAQ from MCP-Mealprep

Which MCP servers are included in the stack?

The stack includes servers for web search (SearXNG, Brave), GitHub, Slack, Google Maps, web crawl (Fetch, Puppeteer), sequential thinking, Postgres/vector retrieval (Supabase, Qwant, Pinecone, Memory), a calculator, Everart, and more. The list can be customized.

How do I add a new MCP server to this project?

Add a new service block to docker-compose.yml following the existing syntax — the server must have a clean install command via npx, uv, or pip, and all known environment variables and the startup.sh command must be included. Submit a pull request or issue with the proven configuration.

Is MCP-Mealprep secure?

The startup.sh script runs npx mcp-shield to scan for common security issues, and traffic is exposed over SSE. The project does not vet individual MCP server functionality or safety; you are responsible for securing your underlying system, proxy, and encryption settings.

What are the prerequisites to run MCP-Mealprep?

You need Docker and optionally Portainer. Some servers require API tokens or keys (e.g., Slack Bot Token, Brave Search API Key, Google Maps API Key) which must be provided as environment variables.

Can I connect MCP-Mealprep to OpenWebUI or Claude?

Yes. Once the stack is running, connect your MCP client to the server’s IP and port. If you enable mcpo (USE_MCPO=true), each server also exposes an OpenAPI endpoint that integrates with OpenWebUI and provides documentation at /docs. Integration notes for specific clients (n8n, Flowise, Claude, Cursor) are currently marked as TBD.

评论

版本控制 分类下的更多 MCP 服务器