modelcontextprotocol/swift-sdk

The official Swift SDK for Model Context Protocol servers and clients. Maintained in collaboration with Apple

What it solves

The Model Context Protocol (MCP) standardizes how applications communicate with AI models. This SDK gives Swift developers official building blocks to create both clients (apps that connect to MCP servers) and servers (apps that expose capabilities to clients). Without it, developers would have to implement the protocol from scratch.

How it works

The SDK includes a Client class for connecting to MCP servers over stdio or HTTP transports, and a Server class for hosting capabilities. Both support core MCP features: tools, resources, prompts, completions, sampling (server→client LLM requests), elicitation (server→client user input requests), roots (filesystem boundaries), logging, and progressive updates. The client supports strict or non-strict capability checking, request batching, and cancellation. The server includes graceful shutdown and HTTP request context handling.

Who it’s for

Swift developers building applications that need to interact with AI/ML models via MCP, or creating MCP servers to expose tools, data, and prompts. Requires Swift 6.0+ (Xcode 16+).

Highlights

  • Official SDK aligned with the latest MCP specification (2025-11-25).
  • Both client and server implementations in one package.
  • Supports multiple transports: stdio for local subprocesses and streaming HTTP for remote communication.
  • Comprehensive client features: tools, resources, prompts, completions, sampling, elicitation, roots, logging, cancellation, progress tracking, batching, and authentication flows.
  • Server features: same capabilities plus initialize hook, HTTP request context, graceful shutdown.
  • Easy integration via Swift Package Manager.

Related

  • Project
    hujjun/mcd-saver-strategist

    A WorkBuddy skill that fetches McDonald’s China menu and coupons via the MCP API, then runs a deterministic Python optimizer to find the cheapest legal combination of items, respecting budget, calorie and delivery constraints. It can be used in chat or via a command‑line demo, and includes a self‑test suite and an online interactive demo.

  • Project
    e2b-dev/desktop

    An open-source secure virtual desktop designed for AI agents to perform 'Computer Use' tasks via programmatic GUI control and screen streaming.

  • Project
    VersusControl/devops-ai-guidelines

    A comprehensive educational repository providing roadmaps, technical tutorials, and governance frameworks to help DevOps professionals and teams integrate AI into their infrastructure and workflows.

  • Project
    langwatch/better-agents

    A CLI tool and set of standards for building production-ready AI agents with standardized project structures, prompt versioning, and integrated testing.