# MCP-Go

> A Go implementation of the Model Context Protocol (MCP), enabling seamless integration between LLM applications and external data sources and tools.

## Docs

- [Core Concepts](/core-concepts): Understanding the fundamental concepts of MCP and how MCP-Go implements them is essential for building effective MCP servers and clients.
- [Getting Started](/getting-started): The Model Context Protocol (MCP) is an open standard that enables secure, controlled connections between AI applications and external data sources and tools. It provides a standardized way for Large Language Models (LLMs) to access and interact with external systems while maintaining security and user control.
- [Quick Start](/quick-start): Get up and running with MCP-Go in minutes. This guide walks you through creating your first MCP server and client.
- [Sampling](/clients/advanced-sampling): Learn how to implement MCP clients that can handle sampling requests from servers, enabling bidirectional communication where clients provide LLM capabilities to servers.
- [Client Basics](/clients/basics): Learn the fundamentals of creating and managing MCP clients, including lifecycle management, initialization, and error handling.
- [Building MCP Clients](/clients): Learn how to build MCP clients that connect to and interact with MCP servers. This section covers client creation, operations, and transport-specific implementations.
- [Client Operations](/clients/operations): Learn how to use MCP clients to interact with servers through tools, resources, prompts, and subscriptions.
- [Client Transports](/clients/transports): Learn about transport-specific client implementations and how to choose the right transport for your use case.
- [StreamableHTTP Transport](/transports/http): StreamableHTTP transport provides traditional request/response communication for MCP servers, perfect for REST-like interactions, stateless clients, and integration with existing web infrastructure.
- [Transport Options](/transports): MCP-Go supports multiple transport methods to fit different deployment scenarios and integration patterns. Choose the right transport based on your use case, performance requirements, and client capabilities.
- [In-Process Transport](/transports/inprocess): In-process transport enables direct integration of MCP servers within the same process, eliminating network overhead and providing seamless integration for embedded scenarios.
- [SSE Transport](/transports/sse): Server-Sent Events (SSE) transport enables real-time, web-friendly communication between MCP clients and servers. Perfect for web applications that need live updates and multi-client support.
- [STDIO Transport](/transports/stdio): STDIO (Standard Input/Output) transport is the most common MCP transport method, perfect for command-line tools, desktop applications, and local integrations.
- [Sampling](/servers/advanced-sampling): Learn how to implement MCP servers that can request LLM completions from clients using the sampling capability.
- [Advanced Server Features](/servers/advanced): Icons provide visual identifiers for servers, tools, resources, and prompts. All icon URIs must use HTTPS or data URI schemes for security.
- [Server Basics](/servers/basics): Learn how to create, configure, and start MCP servers with different transport options.
- [Auto-Completion](/servers/completions): The `completion/complete` MCP method enables servers to provide argument completions for prompts and resource templates. This helps clients offer autocomplete suggestions when users are filling in prompt arguments or resource template parameters.
- [Elicitation](/servers/elicitation): Learn how to implement MCP servers that can request additional information from users during interactions.
- [Building MCP Servers](/servers): Learn how to build powerful MCP servers with MCP-Go. This section covers everything from basic server setup to advanced features like typed tools and session management.
- [Middleware](/servers/middleware): Middleware lets you wrap tool handlers with cross-cutting logic such as logging, authentication, and rate limiting. Middlewares execute in registration order (first registered runs first) and form a chain around the underlying tool handler.
- [Implementing Prompts](/servers/prompts): Prompts are reusable interaction templates that help structure conversations between users and LLMs. They provide context, instructions, and can include dynamic content from resources.
- [Implementing Resources](/servers/resources): Resources expose data to LLMs in a read-only manner. Think of them as GET endpoints that provide access to files, databases, APIs, or any other data source.
- [Task-Augmented Tools](/servers/tasks): Task-augmented tools allow long-running operations to execute asynchronously. Instead of blocking until completion, the server creates a task and the client can poll for status updates. This is ideal for operations like batch processing, data analysis, or any work that may take more than a few seconds.
- [Implementing Tools](/servers/tools): Tools provide functionality that LLMs can invoke to take actions or perform computations. Think of them as function calls that extend the LLM's capabilities.