assistant - ui is an open - source TypeScript/React library for quickly building production - grade AI chat interfaces, providing composable UI components, streaming responses, accessibility, etc., and supporting multiple AI backends and models.

The official MCP server of phospho, implementing the model context protocol, supporting the control of physical robots through natural language, and providing functions for performing actions and camera image streaming.

This project is a proof-of-concept (POC) that demonstrates how to implement the Model Context Protocol (MCP) through a custom host for testing agentic systems. Core features include OpenAI-compatible API, Google Gemini integration, streaming response support, and tool invocation capabilities.

FridaC2MCP is a proof - of - concept project that exposes the capabilities of the Frida dynamic instrumentation tool as a remote MCP server through HTTP streaming, aiming to achieve remote and automated mobile application security testing. The project is designed to run on a rooted Android device, allowing clients to analyze and manipulate running applications over the network.

PersonaLive is an AI-based real-time portrait animation system that supports real-time driving through a camera or offline video processing to generate vivid virtual avatars, suitable for scenarios such as live streaming.

This project is a multi - tool integration and code execution platform based on FastMCP. It supports importing MCP tools from local and SSE clients, provides Docker deployment and streaming code execution management, and is suitable for various tool invocation scenarios such as chemical information query and web page parsing.

An MCP server that provides enhanced file operation functions, supporting streaming processing, patching, and change tracking

A server that implements the Model Context Protocol (MCP) for interacting with Dust AI agents, supporting real-time streaming responses and various integration methods.

FFmpeg-MCP is a model context protocol server designed to help LLMs process various multimedia content created by humans and machines, including functions such as decoding, encoding, transcoding, mixing, separating, streaming, filtering, and playing.

The MCP Server Dashboard is a unified management platform that simplifies the management and scaling of multiple MCP servers through flexible streaming HTTP endpoints. It supports on-demand access to all servers, a single server, or a logically grouped set of servers, and provides a centralized web management interface.

A TypeScript server that converts Dify workflows into MCP tools, supporting configuration through YAML and streaming responses.

VulniCheck is an AI - driven security scanner that provides comprehensive security analysis for Python projects and GitHub repositories. It runs as a Docker - based HTTP MCP server, supports standard HTTP streaming, and offers containerized deployment and comprehensive vulnerability scanning capabilities, including dependency checking, key detection, Docker file analysis, and AI risk assessment.

The Codex MCP Tool is an open-source MCP server that connects the IDE or AI assistant to the Codex CLI, supporting non-interactive automation, secure sandbox editing, and large-scale code analysis. It provides progress streaming updates and a structured change mode.

This project provides a reference implementation of an MCP server that supports OAuth authorization and streaming HTTP/SSE transmission, allowing developers to quickly set up their own MCP services and integrate OAuth authentication.

xsMCP is a lightweight MCP SDK designed to simplify client/server construction, avoid relying on large libraries, focus on streaming HTTP communication, be compatible with Web standards, and support multiple server frameworks.

This project demonstrates a serverless MCP (Model Context Protocol) tool implementation based on AWS Lambda, including a Python server - side and a TypeScript client, supporting streaming HTTP communication. The server - side simplifies the MCP tool development process, has built - in session state management, and demonstrates integration with Amazon Bedrock.

A VSCode extension based on the FastMCP framework that transforms VSCode into an MCP server, supporting remote execution of internal VSCode commands via HTTP Streaming, querying workspace information, and asynchronous task management

A server based on the Model Context Protocol (MCP) and Server-Sent Events (SSE), integrating the Brave Search API to provide real-time streaming search functionality for AI models and clients.

WebScout MCP is a web application reverse-engineering tool based on the Model Context Protocol, focusing on automated analysis of chat interfaces and streaming APIs, providing browser automation, network traffic capture, and intelligent pattern detection capabilities

A secure pre-approved command execution MCP server that provides restricted command execution functionality and supports real-time output streaming