What transports does SwiftMCP support?
SwiftMCP supports Standard I/O (stdio), HTTP+SSE (Server‑Sent Events), and TCP+Bonjour for local discovery. These can even be run in parallel.
@Cocoanetics
基本信息
Model Context Protocol Server for Swift
工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
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。
SwiftMCP is a Swift implementation of the Model Context Protocol (MCP) for JSON-RPC communication over multiple transports. It is intended for Swift developers who want to build MCP servers or clients that expose tools, resources, and prompts to AI models and other MCP-aware applications.
Add SwiftMCP as a dependency in your Package.swift, then declare a server with the @MCPServer macro, define tools with @MCPTool, and configure a transport such as stdio, HTTP+SSE, or TCP+Bonjour. The included demo provides commands like SwiftMCPDemo stdio and SwiftMCPDemo httpsse --port 8080.
@MCPTool), resources (@MCPResource), and prompts (@MCPPrompt)SwiftMCP supports Standard I/O (stdio), HTTP+SSE (Server‑Sent Events), and TCP+Bonjour for local discovery. These can even be run in parallel.
The HTTP+SSE transport can require a bearer token via the `--token` option, or validate tokens through an OAuth provider using introspection or JWKS. Custom authorization handlers can also be set.
Yes. The `SwiftMCPUtility generate-proxy` command creates a typed Swift proxy from any MCP server (including non‑SwiftMCP servers) by inspecting its tool schemas. Optionally, return types can be enriched from an OpenAPI document.
When the `--openapi` option is enabled, the server exposes an OpenAPI specification at `/openapi.json` and an AI plugin manifest at `/.well-known/ai-plugin.json`, enabling integration with AI models that understand OpenAPI.
SwiftMCP is written in Swift and works on macOS, Linux, and Windows. By disabling the `Server` trait (which drops swift‑nio), the package builds on Windows and reduces dependencies for client‑only consumers.

See and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.

Connect your AI assistant to your own YOUB endurance training account. Read recorded activities with full analysis, planned sessions, race goals, wellness (HRV, sleep, resting HR) and your athletic profile. Create or mov
Painless E2E Automation for Mobile and Web

Yuppsy MCP: build and publish real estate websites from your AI assistant. Yuppsy lets your AI assistant create live websites for real estate agents and individual property listings. You don't need a designer or any code

DeepMark helps teachers deliver rapid, consistent marking with meaningful feedback for every student — in a fraction of the time. What once took a week, now takes one free period.
评论