# AI tools

This project demonstrates how to build a remote MCP server on Cloudflare Workers, including the OAuth login function, and supports connection and use through the MCP Inspector and Claude desktop client.

An AI agent forwarding solution based on the MCP protocol, realizing unified HTTP interfaces and fast integration of large model services.

A lightweight screenshot microservice based on FastAPI, compatible with the MCP protocol, supporting full - screen, region, and window screenshots, suitable for AI agents and automation tools.

FastMCP is a TypeScript-based framework for building MCP servers that support client sessions. It provides simple definitions for tools, resources, and prompts, supports functions such as authentication, session management, returning image and audio content, logging, error handling, SSE communication, etc., and includes CLI tools for testing and debugging.

This project demonstrates how to build a remote MCP server on Cloudflare Workers and implement the OAuth login function. It supports local development and debugging, connection testing via the MCP inspector, integration with the Claude desktop client, and can ultimately be deployed to the Cloudflare cloud service.

Steel MCP Server is a server based on the Model Context Protocol. Through the integration of the Puppeteer tool and the Steel browser, it enables LLM models like Claude to perform automated operations such as web page navigation, clicking, and input. It supports both local and cloud running modes and provides web page interaction and screenshot functions.

A Wikipedia search service based on the MCP protocol, providing a structured interface for AI to access Wikipedia content.

Airtable MCP is a service that connects AI tools with Airtable, supporting querying, creating, updating, and deleting records through natural language. It includes functions such as basic management, table operations, schema operations, record filtering, and data migration.

Memory Bank MCP is a server based on the Model Context Protocol (MCP) for managing the memory bank, enabling AI assistants to store and retrieve information across different sessions.

A simple MCP client-server project implemented in Python, supporting direct tool invocation and LLM interaction based on OpenAI.

An MCP protocol server that provides access to the tools on the Toolhouse platform

The Ragie Model Context Protocol server provides knowledge base retrieval capabilities

This project demonstrates how to deploy a remote MCP server on Cloudflare Workers and implement the OAuth login function. It supports local development and debugging, can be connected through the MCP inspector or the Claude desktop application, and is finally deployed to the Cloudflare cloud service.

This project demonstrates how to deploy a remote MCP server on Cloudflare Workers and implement the OAuth login function. It supports local development and debugging, can connect to the service through the MCP inspector or Claude desktop application, and is finally deployed to the Cloudflare cloud.

A weather data server based on the Model Context Protocol that connects to the OpenWeatherMap API via SSE to provide real-time weather query services for AI clients.

GitHub Repo MCP is an open-source MCP server that allows AI assistants to browse GitHub repositories, explore directory structures, and view file contents. It supports MCP clients such as Cursor.

This is an MCP server that connects to the Coreflux MQTT broker, providing Coreflux and MQTT operation tools for AI assistants such as Claude.

The YouTube MCP Server is a tool for obtaining and extracting subtitles from YouTube videos, enabling AI language models to access and process video content.

A server based on the Model Context Protocol that integrates the ntfy push notification service, enabling LLMs and AI agents to send highly customizable notifications to devices.

An MCP server implemented in TypeScript that provides access and operation support for shadcn/ui components to AI assistants, including obtaining source code, sample code, and installation guides.

This project demonstrates how to build a remote MCP server on Cloudflare Workers, including the OAuth login function, and supports connection and use via the MCP inspector and Claude desktop client.

This project is a server based on the MCP protocol, specifically designed to check domain name availability through WHOIS queries and support the integration of domain name checking functionality into tools such as AI assistants.

Supergateway is a tool used to convert MCP servers based on standard input and output into services accessible via SSE or WebSocket, facilitating remote debugging and integration.

This project demonstrates a disk usage query service based on the MCP protocol, allowing Claude AI to obtain real-time disk information of a Mac system through a local server and present it in a structured and user-friendly manner.

Quick-start Auto MCP is a tool that helps users quickly register the Anthropic Model Context Protocol (MCP) in Claude Desktop and Cursor. It provides functions such as RAG document search, Dify workflow, real-time web search, and supports one-click generation of JSON configuration files.

This project demonstrates how to deploy a remote MCP server on Cloudflare Workers and implement OAuth login, supporting connection and use via the MCP Inspector and Claude Desktop.

A Python service implementing the MCP protocol, integrating the Jira API for AI models to call

A Python template project for creating a VS Code MCP server, following the best practices of production-level MCP servers, providing infrastructure for configuration management, logging, and tool implementation.

The MCP project is a collection of message control protocol services for local AI applications, including a server and client modules for Git operation management.

This project demonstrates how to deploy a remote MCP server on Cloudflare Workers and implement OAuth login functionality. It provides a local development guide, a method for connecting with the MCP Inspector, a solution for integrating with the Claude desktop client, and a complete process for deploying to Cloudflare.