Step 4 of 6
Rooms are only half a story. Your labyrinth needs inhabitants.
In step 2, you defined a rooms concern with a template. Now do the same for
characters. Open .txtscape/config.json and add a second concern:
.txtscape/config.json
{
"concerns": [
{
"folderName": "rooms",
"label": "Rooms",
"description": "Locations in the labyrinth with descriptions, exits, and items.",
"template": "# {name}\n\n## Description\n\n## Exits\n\n## Notable Features"
},
{
"folderName": "characters",
"label": "Characters",
"description": "Beings in the labyrinth with descriptions, locations, and dialogue.",
"template": "# {name}\n\n## Description\n\n## Location\n\n## Dialogue"
}
]
}
The agent creates characters/minotaur.txt following your characters
template — ## Description, ## Location,
## Dialogue.
Your project now has two kinds of knowledge, each in its own folder:
📄 characters/ariadne.txt — # Ariadne
📄 characters/minotaur.txt — # The Minotaur
📄 rooms/ariadnes-chamber.txt — # Ariadne's Chamber
📄 rooms/entrance-hall.txt — # The Entrance Hall
📄 rooms/minotaurs-lair.txt — # The Minotaur's Lair
📄 rooms/twisting-corridors.txt — # Twisting Corridors
📄 rooms/underground-pool.txt — # Underground Pool
Notice what you did: you followed the same pattern from step 2 — define a concern, give it a template, ask the agent to create content. The agent shaped its output to match, both times.
.txtscape/pages/. Your project root
still has just index.html and package.json —
no sprawling markdown files, no AI_CONTEXT.md polluting your tree.
git add .txtscape
git commit -m "add characters concern with Minotaur and Ariadne"