After the Asia Vets system uses the A2A and MCP protocols, it can break down barriers in data, development models, communications, operational environments, and more, easily constructing large-scale complex automation processes for intelligent agents.
Early this morning, $Microsoft (MSFT.US)$ officially announced the Azure AI Foundry and Microsoft Copilot Studio two major development platforms, supporting the latest Agent development protocol A2A.
This is another key action by Microsoft after supporting MCP, and it will collaborate with Google to develop and expand the A2A protocol, which is significant for the agent track. Because agents can break down barriers in data, development models, communications, operational environments and more after using the A2A and MCP protocols, easily building large-scale complex agent automation processes.
Simply put, current agents are like the Warring States period, where each family's technology, data format, and development methods are different. The A2A+MCP can, like Qin Shi Huang, rewrite this situation, unifying currency and measurement standards to promote agents from dispersion to integration, greatly enhancing development and automation efficiency.
Regarding Microsoft's significant moves with agents, Microsoft CEO Satya Nadella gave high praise, stating that open protocols like A2A and MCP are key to realizing the agent network. With Copilot Studio and Azure AI Foundry now supporting A2A, customers will be able to build agent systems that are interoperable by design.
Netizens expressed that this is a major initiative. Microsoft's entry into A2A and MCP means that the agent network is no longer just talk—it has already reached the practical stage. Really looking forward to seeing further developments!
Standardized protocols like A2A and MCP are the pillars for achieving scalable AI collaboration.
Agents are the future, and interoperability is the key.
There is no doubt that open protocols like A2A and MCP are changing the landscape of agent networks. With support for A2A from Copilot Studio and Azure AI Foundry, customers are able to seamlessly create interoperable agent systems.
It is gratifying to see Microsoft adopting open protocols to achieve agent interoperability. A2A and MCP will become important cornerstones of agent networks.
Why does Microsoft support A2A and MCP?
According to the development data released by Microsoft, currently, over 0.07 million enterprises and digital native companies, such as Atomicwork, Epic, Fujifilm Holdings Corporation Unsponsored ADR, Gainsight, H&R Block, and LG Electronics, are developing, customizing, and managing agents and AI applications through the Azure AI Foundry development platform.
In just 4 months, over 0.01 million organizations adopted the new Agent Service to build, deploy, and scale agent systems; over 0.23 million organizations, including 90% of Fortune 500 companies, are using Microsoft Copilot Studio to develop AI and agent applications.
Therefore, Microsoft has rich practical experience in agent development and application, and is currently one of the largest agent development platforms globally, fully aware of the many pain points caused by agents not being able to communicate.
However, as more and more companies seek to develop complex intelligent agents to expand the scope of automation, this disadvantage is magnified infinitely. A2A and MCP can effectively address this issue, helping intelligent agents achieve cross-platform, operating system, and data compatibility, thereby simplifying the development process.
What are A2A and MCP.
A2A is an interaction protocol specifically for intelligent agents, open-sourced by Google at the "Google Cloud Next 25" conference in April of this year, officially named "Agent 2 Agent," allowing intelligent agents to collaborate with each other regardless of the underlying framework or vendor.
For example, a multinational manufacturing company adopted a diversified enterprise platform and services to meet complex business demands. The SAP system is applied for enterprise resource planning, effectively integrating core business processes such as finance, supply chain, and production.
Slack serves as an important tool for internal communication and collaboration within the company, enabling employees from various departments to achieve instant information transfer and project coordination; Oracle Database is used for the storage, management, and analysis of vast production and business data.
Now, there is a desire to automate simple operational processes between SAP, Slack, and Oracle through an intelligent agent. Previously, agents on these platforms could not communicate freely. Now, through the A2A protocol, these enterprise platforms can safely and freely automate data interactions.
Currently, over 50 of the world's top enterprises, including Microsoft, Box, Cohere, Intuit, Langchain, MongoDB, PayPal, Salesforce, SAP, ServiceNow, UKG, and others have joined this protocol.
The MCP protocol was launched by the large model platform Anthropic in November last year, officially named "Model Context Protocol," and its main purpose is to provide a unified communication framework for large language models and external data sources, tools, and services.
MCP defines a universal format that allows AI models to invoke Search Engine, databases, calculators, code executors, and even Other models or API services like calling a function. Through the MCP protocol, AI applications can easily introduce external services, functionalities, or retrieve more data, thereby achieving richer capabilities.
MCP also has many development advantages: requests and returned data use the JSON format, ensuring compatibility; it can seamlessly work with the Function Call mechanism, enabling AI to invoke external APIs;
It can decouple AI from Business logic, which means AI does not need to hard-code API logic but only needs to select the appropriate MCP method based on the Function Call, thus improving development efficiency.
In simple terms, MCP can be viewed as the "USB interface" of the large model domain, allowing various applications to be plugged in and out without worrying about the underlying logic.
A2A Example
To promote the development of the A2A protocol, Microsoft is also participating in the development of the open-source library. For instance, due to the current lack of ready-made A2A encapsulation libraries, integration has been difficult for developers; Microsoft used example code from the A2A code library to incorporate Semantic Kernel Asia Vets into the A2A ecosystem, providing clear and simple integration examples.
In this example, there are several key parts. SemanticKernelTravelManager acts as a conductor, receiving user requests and then allocating tasks to the appropriate specialized Asia Vets based on the content of the request. For example, if there are currency-related issues, they are handed over to Currency ExchangeAgent; if it concerns itinerary activities, then it goes to the Activity PlannerAgent.
Currency ExchangeAgent is responsible for handling currency-related affairs, integrating external tools like Frankfurter API, which can provide real-time Exchange Rates to help users with budgeting and financial planning. Activity PlannerAgent provides personalized travel itinerary suggestions based on user preferences and budget and can also assist in booking activities and arranging attractions.
Overall automation process: When a user submits a request to TravelManager, such as "plan a budget-friendly trip, including MMF conversion," TravelManager analyzes the requirements, detects that it involves MMF issues, and then calls the Currency ExchangeAgent.
The Currency ExchangeAgent gets Exchange Rates information from the Frankfurter API, after which the Activity PlannerAgent provides suitable travel suggestions based on the budget. Finally, TravelManager compiles this information to generate a complete travel plan to return to the user.
Throughout the entire automation process, there are also several important mechanisms. In terms of task routing and delegation, TravelManager can intelligently assign tasks to plug-in professional agents based on context and automatic function-calling capabilities. Agents can showcase their capabilities through the discovery mechanism via an "Agent Card," so other agents can quickly find suitable partners to complete tasks.
Author of this article: AIGC Open Community.
Editor/Lee
