Mcphost Demo
What is Mcphost Demo
mcphost-demo is a demonstration project for accessing Large Language Models (LLM) using an OpenAI-compatible protocol, featuring both client and server components.
Use cases
Use cases for mcphost-demo include retrieving weather forecasts for specific locations, calculating distances between cities, and providing real-time navigation assistance.
How to use
To use mcphost-demo, set up your environment by adding your API keys to a .env file, activate the virtual environment, and run the client with either standard input or server-sent events (SSE) mode.
Key features
Key features include compatibility with OpenAI protocols, support for both standard input and SSE, and the ability to query for various types of information, such as weather forecasts and geographical distances.
Where to use
mcphost-demo can be utilized in various fields such as weather forecasting, geographical information systems, and any application requiring natural language processing capabilities.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Overview
What is Mcphost Demo
mcphost-demo is a demonstration project for accessing Large Language Models (LLM) using an OpenAI-compatible protocol, featuring both client and server components.
Use cases
Use cases for mcphost-demo include retrieving weather forecasts for specific locations, calculating distances between cities, and providing real-time navigation assistance.
How to use
To use mcphost-demo, set up your environment by adding your API keys to a .env file, activate the virtual environment, and run the client with either standard input or server-sent events (SSE) mode.
Key features
Key features include compatibility with OpenAI protocols, support for both standard input and SSE, and the ability to query for various types of information, such as weather forecasts and geographical distances.
Where to use
mcphost-demo can be utilized in various fields such as weather forecasting, geographical information systems, and any application requiring natural language processing capabilities.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Content
MCP Demo
mcp host demo, include mcp client and mcp server(sse and stdio).
Access LLM using OpenAI-compatible protocol
install
echo "DEEPSEEK_API_KEY=xxx" >> .env echo "\nAMAP_KEY=xxx" >> .env uv venv source .venv/bin/activate
run
stdio
uv run client.py mcpserver-weather.py
sse
uv run client.py
example
query:
ca forecast -------------- [调用工具 get_forecast，参数 {'latitude': 36.7783, 'longitude': -119.4179}] Here’s the current weather forecast for California (based on coordinates near the state's center): ### **Today** ☀️ **Sunny** - **High:** 62°F - **Wind:** 0-5 mph (WSW) ### **Tonight** 🌤️ **Mostly Clear** - **Low:** 42°F - **Wind:** 5 mph (NE) ### **Friday** ☀️ **Sunny** - **High:** 68°F - **Wind:** 5 mph (SSE) ### **Friday Night** 🌙 **Clear** - **Low:** 45°F - **Wind:** 5 mph (NNE) ### **Saturday** ☀️ **Sunny & Warmer** - **High:** 73°F - **Wind:** 5 mph (W) *Note: This is a general forecast; conditions may vary by region (e.g., coastal, inland, mountainous). Let me know if you'd like details for a specific city!*
query:
北京到武汉有多远 -------------- [调用工具 maps_geo，参数 {'address': '北京'}] 北京到武汉的直线距离约为**1,150公里**，实际交通距离会因路线不同有所差异： 1. **高铁**：最快约4-5小时（京广高铁，约1,200公里） 2. **驾车**：约12小时（京港澳高速，约1,200公里） 3. **飞行**：直飞约1.5小时（空中距离约1,050公里） 建议使用导航软件（如百度地图/高德地图）获取实时路线规划。需要更具体的出行信息可随时告知！
Dev Tools Supporting MCP
The following are the main code editors that support the Model Context Protocol. Click the link to visit the official website for more information.
