آموزش MCP و A2A - پروتکل کانتکست مدل و پروتکل عامل به عامل - آخرین آپدیت

درس ها Lessons

• Course Introduction

• SECTION 1 MCP Overview - What is Model Context Protocol?

• Get Code on Github

• Install Claude for Desktop

• Install Python, UV and VS Code

• SECTION 2. Build your own mcpserver

• SECTION 3 Build Your Own MCP Client

• SECTION 4 - Containerize your MCP Server Using Docker

• SECTION 5 - Simplify client code with LangGraph & LangChain

• SECTION 6 START Build MCP Client with Multiple Server Support: 6.1 Introduction

• 6.2 Let's look at the config.json file

• 6.3 Demo - MCP Client with Multiple MCP Servers

• 6.4 MCP Client (with json config) Code Walk Through - Part 1

• 6.5 Why Choose Gemini 2.0 Flash and not the Pro Models

• 6.6 MCP Client (with json config) Code Walk Through - Part 2 (continued)

• 6.7 How to use existing MCP servers from MCP Github (example uses "fetch" server)

• SECTION 7 START - Server Sent Events - MCP Server and Clients using SSE - 7.1 Introduction

• 7.2 Quick Recap, What is STDIO and SSE?

• 7.3 Setup Directories, Clone GitHub Code (git pull only, if done at course start)

• 7.4 Setup Virtual Environment and Dependencies

• 7.5 MCP SSE Server Code Walkthrough

• 7.6 MCP SSE Client Code Walkthrough

• 7.6 Dockerfile Code (for MCP Server) Walkthrough

• 7.7 Test your MCP SSE Server and Client Locally

• SECTION 8 START - Deploying MCP Server to Google Cloud Platform - 8.1 Create a new Gmail Account (if

• 8.2 Create a Google Cloud Project

• 8.3 Install and Setup Google Cloud Command Line Interface (gcloud CLI)

• 8.4 Build Docker Image for Google Cloud

• 8.5 Deploy MCP SSE Server to Google Cloud Run

• 8.6 Test MCP SSE Server on Google Cloud

• SECTION 9 START - STREAMABLE HTTP MCP SERVER - 9.1 Quick MCP Recap and What is Streamable HTTP

• 9.2 Overview of Streamable HTTP using Sequence Diagram

• 9.3 Initialisation Phase

• 9.4 Client Requests Phase of Streamable HTTP

• 9.5 MCP Client Notifications and Responses

• 9.6 Client Listening to Messages from the Server in Streamable HTTP

• 9.7 Session handling in Streamable HTTP

• 9.8 External Resource for keeping up to date

• SECTION 9.2 START Streamable HTTP MCP Server - Step-by-step demo

• SECTION 9.3 START Streamable HTTP MCP Client using Gemini and Google ADK - Introduction

• 9.3.2 Code overview

• 9.3.3 MCP Client User Interface - cmd.py

• 9.3.4 MCP Client Implementation - client.py

• 9.3.5 MCP Agent Implementation - agent.py

• 9.3.6 Code - MCP config json file

• 9.3.7 Code for Utilities & STDIO Server + Environment Setup

• SECTION 10 START - Streamlit User Interface for MCP Client - 10.1 Streamlit UI MCP Client Overview

• 10.2 Streamlit UI Demo

• 10.3 Comparing our UI with Claude Desktop!

• 10.4 Setup Directories (skip if done earlier)

• 10.5 Setup Google Gemini API Key (verify again if done earlier)

• 10.6 Create Virtual Environment and Install Dependencies

• 10.7 Get Streamlit UI Code

• 10.8 Streamlit App Imports and State Initialisation

• 10.9 Code for Utility Functions

• 10.10 Streamlit App Sidebar Code

• 10.11 Building the Main Chat UI for the App

• 10.12 Core Logic For Query Handling

• 10.13 Trigger Logic for Send and Other Chat Buttons

• 10.14 Streamlit App MCP Client Code

• 10.15 theailanguage_config.json and other files

• 10.16 Running the Streamlit UI

• SECTION 11 START - A2A or Agent to Agent Protocol - Lifecycle - 11.1 Introduction

• 11.2 Why A2A Protocol - A2A vs MCP

• 11.3 Discovery - A2A Client, A2A Server, Agent Card

• 11.4 Initiation - Tasks, Messages, Parts

• 11.5 Processing - Artifacts, Streaming, Push Notifications, Non-streaming

• 11.6 Interaction - input-required state

• 11.7 Completion Flow & Summary of A2A

• SECTION 12 START - Building your own A2A Client and A2A Server

• SECTION 13 START - Build your own A2A Agent with Google Agent Development Kit (ADK) - Introduction

• 13.2 Quick Demo of the A2A Agent

• 13.3 Code and Gemini API Key Setup

• 13.4 Run your A2A Agent Server and Client

• 13.5 Agent Code Walkthrough

• 13.6 Agent Taskmanager Code Walkthrough

• 13.7 Agent main.py code walkthrough

• 13.8 Client code walkthrough - part 1

• 13.9 Client code walkthrough - part 2 - Manual Agent Discovery

• 13.10 Client code walkthrough - part 3

• 13.11 App code walkthrough

• 13.12 Quick Recap of Code Architecture

• 13.13 Models code walkthrough

• SECTION 14 START - A2A with Multiple Agents

• SECTION 15 START - Connect 3 Agents with A2A

• 15.2 A2A Architecture Recap

• 15.3 Agent Discovery and Registry

• 15.4 Orchestrator Host Agent

• 15.5 Orchestrator Taskmanager

• 15.6 Orchestrator entry python script

• 15.7 Greeting agent

• 15.8 Summary and other files, folders

• 15.9 Setup the code & API Key

• 15.10 Running the agents and client

• SECTION 16 START - Introduction to the A2A Python SDK

• 16.2 What is the A2A Python SDK

• 16.3 Demo 1 - Understanding Single turn A2A Agent Streaming Task Updates

• 16.4 Demo 2 - Understanding Multi turn A2A Agent Streaming Task Updates

• 16.5 Setting up and Running the Code

• 16.6 Code Walkthrough: A2A Client

• 16.7 Code Walkthrough: Agent

• 16.8 Code Walkthrough AgentExecutor

• 16.9 Code Walkthrough Agent main py

• SECTION 17 START - Multi-agent A2A + MCP Architecture