Rework Getting started and agent types tutorials

[azhebel]

Move prerequisites to a snippet library and reuse

Rename Complex workflow agents -> Graph-based agents

Rename Agent types -> Agents

Document what agents are and what are the core components, reference different types, document agent configuration via AIAgentConfig

Core components/concepts are: Prompts, Strategies, Tools, Features (instead of Events, because event handling is one of the features)

Also move abbreviations to snippets

Motivation and Context

Breaking Changes

---

Type of the changes

• New feature (non-breaking change which adds functionality)
• Bug fix (non-breaking change which fixes an issue)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update
• Tests improvement
• Refactoring
• CI/CD changes
• Dependencies update

Checklist

• The pull request has a description of the proposed change
• I read the Contributing Guidelines before opening the pull request
• The pull request uses develop as the base branch
• Tests for the changes have been added
• All new and existing tests passed

Additional steps for pull requests adding a new feature

• An issue describing the proposed change exists
• The pull request includes a link to the issue
• The change was discussed and approved in the issue
• Docs have been added / updated

[innateteniuk]

Maybe it would be better to omit the details about the invoke() operator here and instead just write that users can instantiate the AIAgent interface as a class? For example: “Although AIAgent is an interface, you can instantiate it using a constructor-like syntax.”
If a user needs more details, they can follow the link to the API reference provided in the previous sentence.

[innateteniuk]

I'd suggest using "system prompt" in the first paragraph to keep consistency with the parameter name, and then introducing "system message" for users who want to learn more about prompts in Koog.
For example: “A system prompt defines the agent's role and behavior, as well as the purpose, context, and instructions related to the task. Use the systemPrompt parameter to provide it. This creates a single system message in the prompt. To learn more about the prompt structure, see Creating prompts.”

[innateteniuk]

IMO, even though these pages are basic, it makes sense to keep them in their own section. We planned to add an overview for the “Agent types” section that explains the differences between the types and how to choose the right one for the user’s needs. So, a separate section seems even more appropriate. I would only rename it from “Agent types” to “Agents”.

I’d also suggest renaming “Getting started” to “Quickstart”, as this title better reflects the guide’s purpose. In the “Next steps” chapter at the end of the guide, we could add a link to the Agents overview page to help users choose the agent type and get started with implementation.

[innateteniuk]

We need to somehow connect the listed concepts to the actual creation of agents. So I think it would still be better to give a brief description of how Koog agents are created.
I’d use this paragraph from basic-agents.md, since it’s true for all Koog agents:

The AIAgent interface is the primary starting point for creating Koog agents. The overloaded invoke() operator functions on its companion object enable you to instantiate this interface with a constructor-like syntax.

And then add something like

You provide it with the core components that handle the agent’s execution loop:

val agent = AIAgent(
   promptExecutor = promptExecutor,
   toolRegistry = toolRegistry,
   strategy = agentStrategy,
   agentConfig = agentConfig,
   installFeatures = features
)

We can still have the Getting started link below, as it explains how to install Koog and set up a key.

[innateteniuk]

Should we rename Getting started to Quickstart?

[innateteniuk]

Here, I’d start with an intro about what the agent configuration does:

Agent configuration defines the agent's execution parameters, including the initial prompt, language model, and iteration limits.

[innateteniuk]

Let's add

This enables you to define complex prompts with multiple messages, conversation history, LLM parameters, and additional execution parameters.

[innateteniuk]

Let's add

Can be a string or a complex prompt with multiple messages created using the Kotlin DSL.