I craft unique cereal names, stories, and ridiculously cute Cereal Baby images.

mcp
Golang MCP Server
3 years
Works with Finder
1
Github Watches
0
Github Forks
0
Github Stars
Model Context Protocol (MCP) Implementation in Go
This repository contains a Go implementation of the Model Context Protocol (MCP), a JSON-RPC based protocol for interaction between clients and servers in LLM-powered applications.
Contents
-
mcp/schema.go
: Go struct definitions for the MCP schema, translated from the TypeScript schema -
mcp/client.go
: A client implementation for interacting with MCP servers -
examples/mcp_client_example.go
: A simple example showing how to use the client
MCP Overview
The Model Context Protocol (MCP) enables communication between model "hosts" (e.g., LLM clients or systems that interface with LLMs) and "servers" (systems that provide contextual information, tools, prompts, or resources to enhance LLM capabilities).
Key features of MCP include:
- Resources: Files, documents, and other data that a host can read
- Tools: Functions that a host can call on behalf of a model
- Prompts: Templates or pre-defined message sequences a host can use
- Roots: Filesystem or data roots that a server can operate on
Getting Started
Prerequisites
- Go 1.18 or later
Running the Example
# Set the MCP server URL (optional, defaults to http://localhost:3000/jsonrpc)
export MCP_SERVER_URL=http://your-mcp-server/jsonrpc
# Run the example
go run examples/mcp_client_example.go
Using the MCP Client
import (
"fmt"
"path/to/mcp"
)
func main() {
// Create a new client
client := mcp.NewClient("http://your-mcp-server/jsonrpc")
// Initialize the client
clientInfo := mcp.Implementation{
Name: "My MCP Client",
Version: "1.0.0",
}
capabilities := mcp.ClientCapabilities{
// Specify the capabilities your client supports
}
err := client.Initialize(clientInfo, capabilities)
if err != nil {
// Handle error
}
// Now you can use various client methods:
// List resources
resources, err := client.ListResources("")
// Read a resource
resource, err := client.ReadResource("resource-uri")
// Call a tool
result, err := client.CallTool("tool-name", map[string]interface{}{
"argument1": "value1",
"argument2": 123,
})
// Get a prompt
prompt, err := client.GetPrompt("prompt-name", map[string]string{
"argName": "argValue",
})
}
Implementation Details
Schema
The Go schema definitions in mcp/schema.go
provide type-safe structures for MCP messages. This includes:
- JSON-RPC protocol structures
- Resource, Prompt, and Tool definitions
- Content structures for text, images, and embedded resources
- Helper functions for handling polymorphic types via
json.RawMessage
Client
The client implementation in mcp/client.go
provides methods for:
- Initializing connections to MCP servers
- Sending requests and handling responses
- Converting Go structs to JSON-RPC messages
- Helper methods for common MCP operations
Contributing
Contributions are welcome! Here are some ways you can contribute:
- Implement a server
- Add WebSocket support to the client
- Improve error handling
- Add tests
- Document additional use cases
License
This project is licensed under the MIT License - see the LICENSE file for details.
References
相关推荐
Confidential guide on numerology and astrology, based of GG33 Public information
Converts Figma frames into front-end code for various mobile frameworks.
Advanced software engineer GPT that excels through nailing the basics.
Take an adjectivised noun, and create images making it progressively more adjective!
Siri Shortcut Finder – your go-to place for discovering amazing Siri Shortcuts with ease
Discover the most comprehensive and up-to-date collection of MCP servers in the market. This repository serves as a centralized hub, offering an extensive catalog of open-source and proprietary MCP servers, complete with features, documentation links, and contributors.
Micropython I2C-based manipulation of the MCP series GPIO expander, derived from Adafruit_MCP230xx
Mirror ofhttps://github.com/agentience/practices_mcp_server
Mirror ofhttps://github.com/bitrefill/bitrefill-mcp-server
An AI chat bot for small and medium-sized teams, supporting models such as Deepseek, Open AI, Claude, and Gemini. 专为中小团队设计的 AI 聊天应用,支持 Deepseek、Open AI、Claude、Gemini 等模型。
Bridge between Ollama and MCP servers, enabling local LLMs to use Model Context Protocol tools
Reviews

user_3wm64cR5
Superset MCP Integration by aptro is a game-changer for data analytics and visualization. It seamlessly integrates Superset's powerful tools with the MCP ecosystem, providing a smoother and more efficient workflow. The setup is straightforward, and the added functionalities significantly enhance productivity. Highly recommended for anyone looking to streamline their data processes! Check it out here: https://mcp.so/server/superset-mcp/aptro