In this post, I want to explore an idea I’ve been experimenting with: common memory for AI agents. I’ll explain what I mean by this term, how such memory can be implemented, and why I believe it's worth exploring.

What Is Agent's “Common” Memory?

I’m not sure whether “common memory” is already a widely accepted term in the AI space, or even the most accurate label for the concept I have in mind — but I’ll use it for now until a better one emerges (or someone suggests one).

By common memory, I mean:

A shared repository of memories formed by a single AI agent from interactions with multiple other agents — including both humans and other AI agents. For eample, AI Chat can retain information learned from conversations with different users, and selectively reference that information in future interactions.

This is distinct from related terms:

• Shared memory usually refers to memory shared across different AI systems or agents — not across users of the same assistant.
• Collaborative memory comes closer, but often implies more structured cooperation and might be too narrow for what I’m describing.

So for now, I’ll stick with common memory to describe a memory system that allows an AI assistant to retain and selectively reference information learned across interactions with multiple users.

AI Chats do not use Common Memory

When we interact with AI chats like ChatGPT, they typically do not retain information across different users. Each conversation is isolated, and the AI does not remember past interactions with other users. This means that if you ask the AI about something you discussed with another user, it won’t have any context or memory of that conversation. Only the current user's context and history are considered.

With the recent addition of A2A (Agent-to-Agent) protocol support in CleverChatty, it’s now possible to build powerful, intelligent applications—without writing any custom logic. In this blog post, we’ll walk through how to build an Agentic RAG (Retrieval-Augmented Generation) system using CleverChatty.

🤖 What is Agentic RAG?

The term agentic refers to an agent's ability to reason, make decisions, use tools, and interact with other agents or humans intelligently.

In the context of RAG, an Agentic RAG system doesn’t just retrieve documents based on a user’s prompt. Instead, it:

• Preprocesses the user’s query,
• Executes a more contextually refined search,
• Postprocesses the results, summarizing and formatting them,
• And only then returns the final answer to the user.

This kind of intelligent behavior is made possible by using a Large Language Model (LLM) as the core reasoning component.

The goal of a RAG system is to enrich the user’s query with external context, especially when the required information is not available within the LLM itself. This typically involves accessing an organization’s knowledge base—structured or unstructured—and providing relevant data to the LLM to enhance its responses.

Recently, I released a new version of CleverChatty with built-in support for the A2A (Agent-to-Agent) protocol. This addition enables AI agents to call each other as tools, opening the door to more dynamic, modular, and intelligent agent systems.

🔄 What Is the A2A Protocol?

The A2A protocol defines a standard for communication and collaboration between AI agents. It allows one agent to delegate tasks to another, much like how humans might assign work to collaborators with specific expertise.

Many blog posts and articles describe the A2A protocol and provide examples of an A2A client calling an A2A server. However, few explain how an AI agent decides when and why to call another agent in a real scenario.

Let’s consider an example: Imagine there's a specialized AI agent called "Document Summarizer", exposed via the A2A protocol. Another agent — a general-purpose chat assistant with access to an LLM — receives this user query:

In this post, I’m excited to announce a new version of CleverChatty that introduces server mode — unlocking powerful new capabilities for building AI assistants and agents that can interact over the network.

Previously, CleverChatty functioned only as a command-line interface (CLI) for interacting with LLM-based assistants. A typical use case involved a single user chatting with an AI model via the terminal. With this latest update, CleverChatty can now run as a server, enabling:

• Concurrent communication with multiple clients
• Background operation on local or cloud environments
• Integration into distributed agent systems

But that’s not all. The biggest leap forward? Full support for A2A (Agent-to-Agent) protocol.

CleverChatty

This is the third part of the novella "The End of the Holocene". The second part

"He who fights with monsters should see to it that he does not become a monster himself" — Friedrich Nietzsche, "Beyond Good and Evil"

In May 2030, a strange event occurred.

At dawn, a rocket was launched from a desert area in Mexico into space. The incident caused a stir among military forces and politicians around the world—most of all, in the United States. The rocket was large, and based on its trajectory, it was heading into space, not toward any target on Earth.

Immediately after the launch, the U.S. military activated defensive protocols and went into full alert. Several other countries did the same. Within minutes, it became clear that the rocket wasn’t aimed at any earthly location—it was continuing on, further into space. Calculations confirmed it was headed toward Mars.

Emergency meetings were convened between governments and military leaders to determine what was going on. The Mexican government and armed forces also issued statements saying they had no knowledge of the launch and were just as confused as everyone else. They agreed to full cooperation in the investigation.

This is the second part of the novella "The End of the Holocene". The first part

"Life, although it may only be an accumulation of suffering, is dear to me, and I will defend it." — Mary Shelley, Frankenstein

April 2030. San Francisco.

Michael Kravchenko returned to his place of power on the ocean shore near San Francisco. A light mist had almost completely swallowed the Golden Gate Bridge. Michael missed this view, these scents. He hadn’t been here in almost half a year. A cascade of events that followed the launch of the general artificial intelligence Suffragium, developed with his participation, had brought a dark streak into his life.

During that time, Michael had to justify himself a thousand times before various committees, proving that there had been no malicious intent in his actions. That the responsibility couldn’t be laid on the engineers. Sometimes, science encounters failures. Ultimately, it’s all experience. And there hadn’t been any serious problems—aside from the financial losses. Yes, the global network was unstable for a while. But everything was resolved eventually.

I'm excited to introduce a new package for Go developers: CleverChatty.
CleverChatty implements the core functionality of an AI chat system. It encapsulates the essential business logic required for building AI-powered assistants or chatbots — all while remaining independent of any specific user interface (UI).

In short, CleverChatty is a fully working AI chat backend — just without a graphical UI. It supports many popular LLM providers, including OpenAI, Claude, Ollama, and others. It also integrates with external tools using the Model Context Protocol (MCP).

Model Context Protocol (MCP) is now popular subject in discussions around AI and LLMs. It was designed to add a standard way to connect "external" tools to LLMs to make them more useful. Classic example is the "what is the weather in ..." too. Each AI chat tool could do this with own way. Now there is a standard and a plugin made for one Ai Chat system can work with others.

We can se burst of enthusiasm in implementig of MCP servers for everything. I expect this trend will grow. Especially usage of MCP servers with SSE transport. Implementing of MCP server with Server-Sent Events make it similar to SaaS server designed for LLM/AI tool as a client.

There are to reason i decided to write this artcile.

• First. It is reported that internet users now often go to AI chat (often ChatGPT) to find something instead of going to google
• Second. OpenAI anounced they will add support of MCP to ChatGPT Desktop soon. And they will add both STDIO and SSE transport protocols for MCP

Based on this i expect we will see some interesting changes soon.

As large language models (LLMs) find real-world use, the need for flexible ways to connect them with external tools is growing. The Model Context Protocol (MCP) is an emerging standard for structured tool integration.

Most current tutorials focus on STDIO-based MCP servers (Standard Input/Output), which must run locally with the client. But MCP also supports SSE (Server-Sent Events), allowing remote, asynchronous communication over HTTP—ideal for scalable, distributed setups.

In this article, we'll show how to build an SSE-based MCP server to enable real-time interaction between an LLM and external tools.

For this example, I've chosen the "Execute any command on my Linux" tool as the backend for the MCP server. Once connected to an LLM, this setup enables the AI to interact with and manage a Linux instance directly.

Additionally, I'll demonstrate how to add a basic security layer by introducing authorization token support for interacting with the MCP server.

This is the final version of the part 1. V2 (the first version is here)

“But the woman opened the large lid of the chest and released all the misfortunes upon the human race.”
The myth of Pandora’s Box

October 2, 2029. San Francisco.

The morning over the San Francisco Bay was fresh and cool. Scientist Michael Kravchenko often came here before work. The gentle whisper of ocean waves always helped him focus his thoughts and find solutions in his scientific pursuits. This morning, Michael spent more time here than usual. Today was destined to be a landmark in his life and in the history of science—the scheduled launch of the artificial intelligence project he had led as chief architect. It was the work of his lifetime, his greatest achievement.

Surrounded by the peace of nature, Michael felt doubts creeping in. He thought about how the launch of artificial intelligence would change people’s lives. On the one hand, it could mark the beginning of a new era—an era where technology would help solve the most complex problems and improve quality of life. On the other hand, there were fears and uncertainties about safety. Was he about to accidentally create a monster—uncontrollable and unpredictable?

His thoughts drifted between possibilities and risks, between hope and fear. He knew that today would be a milestone—not only for his career, but possibly for all of humanity.

Finally gathering his thoughts, Michael turned away from the shore and headed toward the office. The path to the first launch of artificial intelligence was not just a commute to work—it was the journey toward realizing his deepest dream, a journey on which he had to decide whether the dream was worth all the risks it carried.

At the Yerba Buena Center in San Francisco, an incredible atmosphere reigned that day. Among the attendees, emotions were often compared to those felt during Steve Jobs’s legendary presentations. But many expected something even bigger today. NovusAI had grown from a relatively unknown startup into a global player in the IT industry in just a few years. Their AI-powered technologies had taken over the market and were visibly transforming many areas of everyday life.

Today, NovusAI announced the presentation of something new and revolutionary. But there wasn’t much mystery left. Yes—what everyone had long awaited was finally going to happen. The company planned to announce the launch of the world’s first artificial general intelligence.

Joseph Mayer, the CEO of NovusAI, despite his young age, had already become a symbol of the AI tech revolution. Everyone expected real results and a breakthrough from him. This was supposed to be his day. In the past, startup founders tried to emulate Steve Jobs. But that era had passed—this was the era of Joseph Mayer. And after today’s presentation, if it delivered what everyone hoped for, Joseph Mayer would solidify his status as the number one figure and a living legend in the IT world.

The presentation was being meticulously prepared, as company leadership believed the significance of this event was hard to overstate. Stage lighting, audio effects—everything was set up to highlight the technological leap being officially launched here.