Simple MCP Agent Application
Simple AI application for implementing MCP server and specialized agents (prompt chaining).
Ask AI about Simple MCP Agent Application
Powered by Claude · Grounded in docs
I know everything about Simple MCP Agent Application. Ask me about installation, configuration, usage, or troubleshooting.
0/500
Reviews
Documentation
Simple MCP agent application
This project was created to put hands-on in the development of AI applications.
With this project I try to put in practice two concepts:
- Create MCP servers
- Create specialized agents that interacts one each other resulting in the desired behavior.
What is the project about?
Just for giving a context I decided to create a simple store system.
To keep the things simple, the system was divided into two main areas:
- sellings: All related to selling and products
- system_management: Users and privileges.
Structure of the project
The project is composed of three main modules:
- agent_app: Host and MCP client server. Application
- management_server: MCP server
- selling_server: MCP server
Experiments
Just for experimenting
Validation through the model.
It was insteresting for me to use a model for validating who are you and then
Commands
Run servers
Command to run the application
Note: Notice that we pass the path to the mcp server through the command
uv run .\agent_app\mcp_client.py D:\DOCUMENTS\self_study\Agents\mcp_server\management_server\mcp_management_server.py D:\DOCUMENTS\self_study\Agents\mcp_server\selling_server\mcp_selling_server.py
Run web servers
Command to run the webservers
Management server
uvicorn management_server.management_server:app --reload --port 8000
Selling server
uvicorn selling_server.selling_server:app --reload --port 8001
