daemons-engine 0.18.1

A modern MUD engine - headless game engine for real-time, text-based multiplayer worlds

💀 Daemons Engine

A Python framework for building text-based multiplayer RPGs with real-time WebSocket communication. Includes a batteries-included headless engine server, a debugging client, example content and a CMS.

⚠️ Pre-Release Beta Breaking package updates may still occur regularly and unpredictably. There is no timeline yet for a stable Alpha release.

🤖 Clanker Alert: Extensive use of LLM generated code This project is an exercise in testing the limits of agentic development processes. Ideologically opposed developers should skip it. As an experimental artifact, the existence of this project should not be mistaken for a positive or negative statement about agentic development. Although we can say there has been a rigorous cybersecurity review and the engine has over 800 tests, this software is provided "as-is" and "without warranty of any kind", per the MIT license.

Latest Release: version = "0.18.1" - documentation cleanup and Daemonswright CMS area editor progress

---

Quickstart

Prerequisites

• Python 3.11 or higher
• pip (Python package manager)

Step 1: Create Project Directory

mkdir my-game
cd my-game

Step 2: Create Virtual Environment

# Windows
python -m venv .venv
.\.venv\Scripts\Activate

# macOS/Linux
python3 -m venv .venv
source .venv/bin/activate

Step 3: Install Daemons Engine

pip install daemons-engine

Step 4: Initialize Your Game

daemons init <new_game>

This creates:

• world_data/ YAML content (rooms, NPCs, items)
• config.py Server configuration
• main.py Application entry point
• alembic.ini Database migration config
• alembic/ Migration scripts

Step 5: Set Up Database

cd <new_game>
daemons db upgrade

Step 6: Run the Server

From /<new_game>...

# Development (auto-reload on code changes)
daemons run --reload

# Production (requires JWT secret key)
daemons run --production

The --production flag enables security hardening:

• Requires JWT_SECRET_KEY environment variable
• Enforces JWT issuer/audience validation
• Runs without auto-reload

If JWT_SECRET_KEY is not set, you'll be prompted to generate one.

You should see:

INFO:     Uvicorn running on http://127.0.0.1:8000 (Press CTRL+C to quit)
INFO:     Application startup complete.

Step 7: Connect a Client

Option A: Reference Client

pip install flet
daemons client

Option B: Build Your Own

Connect via WebSocket to ws://127.0.0.1:8000/ws/game/auth.

Test Credentials:

• Username: testplayer1 / Password: testpass1
• Username: testplayer2 / Password: testpass2

---

Create Game Content

All content is defined in YAML files under world_data/.

Example Room (world_data/rooms/tavern.yaml)

id: tavern_main
name: "The Rusty Tankard"
description: |
  A cozy tavern with a crackling fireplace.
room_type: indoor
area_id: starter_town
exits:
  north: town_square

Example NPC (world_data/npcs/barkeeper.yaml)

id: npc_barkeeper
name: "Greta the Barkeeper"
description: "A stout woman with a warm smile."
level: 5
behaviors:
  - merchant
spawn_room: tavern_main

---

Project Structure

my-game/
 world_data/              # Your game content
    rooms/               # Room definitions
    items/               # Items and equipment
    npcs/                # NPC templates
    quests/              # Quest definitions
    dialogues/           # NPC dialogue trees
 alembic/                 # Database migrations
 config.py                # Server configuration
 main.py                  # Application entry point
 dungeon.db               # SQLite database

---

Quick Reference

Task	Command
Initialize project	daemons init
Run migrations	daemons db upgrade
Start server (dev)	daemons run --reload
Start server (prod)	daemons run --production
Run client	daemons client

URL	Description
http://127.0.0.1:8000/docs	Swagger API docs
http://127.0.0.1:8000/redoc	ReDoc API docs

---

Documentation

• Longform Readme
• Roadmap
• Architecture
• Protocol
• Operations
• Contributions

LLM Context Files

The docs/build_docs/ directory contains context articles and implementation plans designed for AI coding assistants (GitHub Copilot, Claude, Cursor, etc.). When we ask an LLM agent to make a detailed plan before implementation, a best practice is to capture the plan in a markdown file so that we can keep the LLM on task and hold it accountable to stated objectives.

For your own game: Run the build script to generate context files for your customizations:

python docs/build_docs/build_context.py

This will produce llm_context_architecture.md, llm_context_content.md, llm_context_index.md, and llm_context_protocol.md.

Delete these files and run build_context.py to rebuild the context after a major change.

---

License

MIT License see LICENSE for details.