Dialogue Setup

This guide explains how Dialogue fits into the current module-based setup and what must exist before the Dialogue editor becomes useful.

info

Dialogue is a separate paid tool. It integrates with CIVIL-AI-SYSTEM, but it is not part of the core Civil AI package itself.

What Dialogue is responsible for

The Dialogue module owns conversation-facing systems such as:

• player-to-agent conversation
• agent conversation services
• response services
• mood and non-verbal presentation
• dialogue authoring data and editor tooling

In a combined installation, Dialogue can consume shared services provided by CivilAI.

Install combinations

The current architecture is designed around these valid combinations:

• Shared + CivilAI
• Shared + Dialogue
• Shared + CivilAI + Dialogue

For most users building a living town with interactive NPCs, Shared + CivilAI + Dialogue is the most practical setup.

Important dependency note

The Dialogue editor checks for item collections through the shared IItemCatalog service before it displays its graph view.

In practice, that means you should set up CivilAI item collections first:

• item name collection
• item type collection
• item size collection

Without those, the Dialogue editor may only show a warning state.

That dependency is why most combined projects should treat Civil AI setup as the foundation and Dialogue as a layer added on top.

Basic setup flow

1. Use the shared bootstrap

Make sure your scene contains BardTreeModuleBootstrap.

2. Create or load Dialogue settings

The Dialogue registrar will create or register module settings for Dialogue when the module is installed.

3. Configure CivilAI item collections first

If you are using both modules, open CIVIL-AI-SYSTEM -> Module Settings and assign your item collections before opening the Dialogue editor.

4. Open the Dialogue editor

Use DIALOGUE-SYSTEM -> Dialogue Editor.

From there you can begin authoring chat tree content.

What the editor supports

The Dialogue editor includes support for:

• chat tree authoring
• autosave to temporary workspace data
• loading a previous temporary workspace
• requirement widgets

That means it is suitable for iterative authoring, but users still benefit from a clear save structure and naming convention.

Practical advice

For a first pass:

• keep one conversation group per NPC role or scenario
• use simple requirements first
• test the runtime flow before creating large branching trees
• make sure your item collections and other shared data are stable before expanding conversation content

Best order for combined projects

If you are building a full medieval town experience, this is a sensible order:

1. Set up bootstrap and CivilAI module settings
2. Configure item collections
3. Build core work and need behaviours
4. Create buildings and population regions
5. Verify agents spawn and simulate
6. Add Dialogue content on top
7. Add mood and response polish later