Report Abuse

Basic Information

This repository demonstrates a working pattern for Server-Sent Events (SSE) based MCP servers and standalone MCP clients. It provides example code that implements an SSE server (weather.py) which exposes tools based on the National Weather Service APIs and a client (client.py) that connects to that SSE endpoint to discover and call tools. The examples are adapted from the MCP documentation's STDIO examples and illustrate how agents can connect to a long-running MCP server, use exposed tools such as get_forecast and get_alerts, and disconnect. The README includes usage commands, a sample interactive session, and a note about supplying ANTHROPIC_API_KEY via an environment variable. The project emphasizes decoupled, cloud-native deployments where servers and clients can run as independent processes.

Links

Categorization

App Details

Features
Provides an SSE-based MCP server implementation (weather.py) that serves weather-related tools adapted from MCP STDIO examples. Includes a standalone MCP client (client.py) that connects to an SSE endpoint, lists available tools, and invokes tools such as get_forecast and get_alerts. Default server host and port are 0.0.0.0:8080 and are configurable via command line arguments --host and --port. Shows example commands for running the server and client using uv. Notes required environment setup such as ANTHROPIC_API_KEY. Offers an installation hint via Smithery for automatic installation and highlights sample interaction logs demonstrating tool calls and responses.
Use Cases
The repo is helpful as a practical reference for developers building MCP-based agents and services that need decoupled communication via SSE. It demonstrates a cloud-native pattern where the MCP server can run independently and multiple clients can connect and use its tools without spawning subprocesses. The weather example provides a concrete toolset and shows how a client discovers tools and calls them with arguments. Configuration and usage examples accelerate onboarding, and the Smithery installation option offers an automated install path. The README also clarifies environment requirements and shows how to customize host and port for different deployment scenarios.

Please fill the required fields*