1 What are LLM Agents and Multi-Agent Systems?

Large language models can describe what they intend to do, but by themselves they cannot actually carry out actions in the world. The chapter introduces LLM agents as the orchestration layer that turns those intentions into real work by managing tool calls, passing results back into the model, and repeating this process until a task is complete. It also explains why this matters in practice: modern agents become far more useful when they can search the web, query knowledge bases, run code, or control software and even operating systems.

To make this concrete, the chapter surveys common use cases such as report generation, web search and deep research, retrieval-augmented generation, coding assistance, and computer-use personal assistants. It then defines an LLM agent as an autonomous system built around a backbone LLM and tools, and shows that the backbone model needs two key capabilities to be effective: planning and tool calling. The processing loop is the core mechanism that ties these together, letting the agent form an initial plan, take actions, inspect results, revise its approach, and continue until the task is done, while optionally involving memory or humans to improve quality and reduce errors.

The chapter also broadens the view from single agents to multi-agent systems, where multiple specialized agents collaborate to handle larger or more complex tasks, sometimes through subagents or interoperable protocols. It highlights important standards and patterns such as MCP for tool and resource access, Agent Skills for reusable workflows, and A2A for agent-to-agent communication, along with design enhancements like memory and human-in-the-loop review. Finally, it lays out the book’s educational goal: to build an entire LLM agent framework from scratch, learning the underlying mechanics of tools, models, memory, skills, subagents, and multi-agent coordination by implementing them directly.

The applications for LLM agents are many, including agentic RAG, report generation, deep search and computer use, all of which can benefit from MAS.

An LLM agent is comprised of a backbone LLM and its equipped tools.

LLM agents utilize the planning capability of backbone LLMs to formulate initial plans for tasks, as well as to adapt current plans based on the results of past steps or actions taken towards task completion.

An illustration of the tool-equipping process, where a textual description of the tool that contains the tool’s name, description and its parameters is provided to the LLM agent.

The tool-calling process, where any equipped tool can be used.

A mental model of an LLM agent performing a task through its processing loop, where tool calling and planning are used repeatedly. The task is executed through a series of sub-steps, a typical approach for performing tasks.

An LLM agent that has access to memory modules where it can store key information of task executions and load this back into its context for future tasks.

A mental model of the LLM agent processing loop that has memory modules for saving and loading important information obtained during task execution.

An LLM agent processing loop with access to human operators. The processing loop is effectively paused each time a human operator is required to provide input.

Multiple LLM agents collaborating to complete an overarching task. The outcomes of each LLM agent’s processing loop are combined to form the overall task result.

The difference between LLM agent framework developers and application developers

A first look at the llm-agents-from-scratch framework that we’ll build together.

A simple UML class diagram that shows two classes from the llm-agents-from-scratch framework. The BaseTool class lives in the base module, while the ToolCallResult lives in the data_structures module. The attributes and methods of both classes are indicated in their respective class diagrams and the relation between them is also described.

A UML sequence diagram that illustrates how the flow of a tool call. First, an LLM agent prepares a ToolCall object and invokes the BaseTool, which initiates the processing of the tool call. Once completed, the BaseTool class constructs a ToolCallResult which then gets sent back to the LLM agent.

The build plan for our llm-agents-from-scratch framework

• LLMs have become very powerful text generators that have been applied successfully to tasks like text summarization, question-answering, and text classification, but they have a critical limitation in that they cannot act; they can only express an intent to act (such as making a tool call) through text. That’s where LLM agents come in to bring in the ability to carry out the intended actions.
• Applications for LLM agents are many, such as report generation, deep research, computer use and coding.
• With MAS, individual LLM agents collaborate to collectively perform tasks.
• Many applications for LLM agents can further benefit from MAS. In principle, MAS excel when complex tasks can be decomposed into smaller subtasks, where specialized LLM agents outperform general-purpose LLM agents.
• LLM agents are systems comprised of an LLM and tools that can act autonomously to perform tasks.
• LLM agents use a processing loop to execute tasks. Tool calling and planning capabilities are key components of that processing loop.
• Protocols like MCP, Agent Skills, and A2A have helped to create a vibrant LLM agent ecosystem that is powering the growth of LLM agents and their applications.
• MCP is a protocol developed by Anthropic that has paved the way for LLM agents to use third-party provided tools.
• Agent Skills is standard that originated from Anthropic that standardizes how reusable workflows are documented and used by LLM agents
• In one form of MAS, subagents can be dispatched by an LLM agent to work on subtasks that contribute to the overall task execution
• A2A is a protocol introduced by Google to standardize agent-to-agent interactions in MAS
• In this book, we’ll primarily be framework developers, building our own LLM agent framework to learn the internals of LLM agents more deeply.
• Supplementary materials in the form of additional example notebooks and capstone projects are provided in the book’s GitHub repository to deepen your learning.

FAQ