liveBook · Manning

Planning

As explained in Spring AI in Action

Agent planning is the capability of selecting a path through available actions to achieve a desired goal.

Agent planning selects a path through available actions to achieve a goal, instead of following a workflow fixed in advance. section 12.1

In Embabel, the agent determines which action methods to use from the user’s goal. Different questions can therefore produce different execution paths without explicitly defining the workflow. section 12.3.6

Planning through an Azul rules query

A shell request asking how many people can play Azul follows actions that extract the game title, locate and read the rules, and determine the player count. A mechanics question follows a different path. section 12.3.6

@SpringBootApplication
@EnableAgentShell
public class EmbabelGamesAgentApplication {
  // ...
}

Enable the agent shell in the Spring Boot application so planned action paths can be exercised interactively. section 12.3.6

How Spring AI in Action covers it

How deep it goes
Demonstrates planning through execution traces rather than formal planning algorithms.
Examples
Uses Azul questions about player count and game mechanics to show different action paths.
What it assumes
Introduces agents and workflows as the vocabulary for relating planning to LLM-based systems.
What you can do after
Evaluate an agent’s autonomy by observing which actions it selects for a goal.
How it connects
Connects planning with agents, workflows, action methods, tool use, and LLMs.

Also explained in

Compare how 6 books explain planning →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Spring AI in Action. The section links point to the parts of the book it draws on.