Building Community Worlds: Tips from EchoQuest Creators
The EchoQuest community library exists because players want to share their worlds. Since we launched creator tools, dozens of worlds have been published — from gritty political thrillers to cozy mystery towns to cosmic horror epics, from solarpunk anarchies to gothic faerie courts to noir-tinged retro-futures. The community is small enough that creators talk to each other directly, learn from each other's successes and stumbles, and tend to converge on a shared craft. This post is the distillation of what those creators have learned, drawn from interviews with the most-played community worlds and from the patterns we see ourselves when reviewing newly-submitted campaigns.
If you're thinking about publishing your first community world, the advice below is the closest thing we have to a battle-tested playbook. None of it is mandatory; some of the best worlds have broken every rule. But they broke the rules deliberately, with a reason, and that's the difference between intentional choice and accidental rough edge. Read these as defaults to deviate from with purpose.
Start With the Opening Scene, Not the Lore
The instinct when building a world is to start with history — the ancient wars, the founding myths, the timeline of major events, the cosmology of the gods. Resist this. Players don't experience your world through its history. They experience it through a specific moment: the opening scene. The opening is the only paragraph guaranteed to land. Everything else is optional.
Build the opening first. Who is the player character? Where are they? What's immediately happening? What's the first decision they need to make? A vivid, grounded opening scene does more for player engagement than pages of backstory. You can always add lore later — and the lore lands harder when it's delivered in response to player questions, in chunks shaped by what the player wants to know.
A useful rule: if your opening scene's first paragraph could appear in another, completely different world, the opening isn't specific enough. The opening should announce this world. Re-read it through the eyes of a player who has never heard of your setting. Are they oriented? Are they tempted? Do they want to know what happens next? If yes, ship it. If no, rewrite.
Write for Listening, Not Reading
Community world text gets narrated aloud — either by the browser TTS or by ElevenLabs voices. This changes how you should write. The cadence of prose that reads beautifully on the page often falls flat when spoken; rhythms that work in audio sometimes look strange in print. Play to the medium.
- Use shorter sentences than you would in prose fiction. The voice doesn't have your eye's ability to skip ahead and parse a long sentence at a glance
- Avoid complex nested clauses that are hard to parse when heard rather than read. Rewrite sentences with three sub-clauses into two sentences with two each
- Favor concrete sensory detail over abstract description. "The smell of woodsmoke and roasted onions" beats "the homey smell" every time
- Read your opening scenario aloud before publishing — if it sounds awkward spoken, rewrite it. This is the single highest-leverage editing pass available to a community creator
- Watch out for proper nouns that are easy to read but hard to say. Aeryndel-Tael-Khorin will trip every TTS engine; consider a shorter alternative or a phonetic note
The best community world descriptions have a rhythm to them when read aloud. That's not an accident.
Give the AI Specific Constraints
The AI Game Master is powerful but needs guardrails to stay consistent with your world. The clearest, most specific constraints produce the best results.
Vague: "Magic is limited in this world." Specific: "Magic requires spoken incantations and physical components. It is rare, feared, and associated with the heretical old religion. No character casts magic publicly. The Church executes practitioners. Petitioners sometimes claim to have witnessed miracles; these are usually frauds, sometimes mass hysteria, and rarely something more disturbing."
Specific constraints let the AI make confident, consistent calls when magic-adjacent situations arise in play. They also rule out generic-fantasy defaults the model would otherwise fall into. The model is happy to generate "a wizard with a staff casts a spell" if you let it; if you've told it magic is illegal and feared, it produces something much more interesting instead.
The same principle applies to technology, social structure, religion, and economics. The more your constraints differ from generic fantasy defaults, the more specific you have to be. A world with no kings needs to say so explicitly. A world with universal literacy needs to say so. A world where commerce isn't conducted in coins needs to spell out what is.
Design for Replayability
Community worlds get played by many different players with different approaches. Design scenarios that work whether the player is aggressive or cautious, political or action-oriented, suspicious or trusting, lawful or chaotic, sociable or solitary. The opening should accommodate any of these starting energies and let the player play to their preference.
The best community campaigns have a central tension that creates interesting choices regardless of the approach — because the approach changes which choices are available and what their costs are, but the fundamental tension remains. A campaign about "a city on the verge of revolution" will play differently for a sympathetic player and a counter-revolutionary player and a player who just wants to keep their head down, but all three will find interesting things to do because the tension itself doesn't depend on which side they pick.
If your campaign requires the player to pursue a specific goal in a specific way to be interesting, it isn't designed for replayability. It's designed for one play-through. That's a legitimate choice — some short, focused campaigns are great as one-shots — but be honest about it in the description so players know what they're getting.
Let the AI Improvise
Some creators try to script every outcome — pre-writing branches, anticipating every player choice, attempting to control the AI's responses in detail. This doesn't work with an AI GM. The model improvises by nature, and trying to constrain it into a pre-written script produces stilted, generic output. Instead of trying to control what happens, focus on the furniture: who the characters are, what they want, what the world feels like, what's at stake. The AI will fill in the rest.
Players consistently report that the best community sessions feel like the world was responding intelligently to their specific choices — not following a script. That feeling comes from good furniture, not from scripted outcomes. A world with detailed factions, rich NPCs, vivid sensory texture, and clear constraints will produce wildly different sessions for different players, all of which feel like authentic stories in that world. A world with a pre-scripted plot will feel like a railroaded video game even when the AI is doing the narration.
Iterate After Publication
Your first version is a draft. Publish, watch how players actually engage with the world (we provide play telemetry to creators about which scenes get the most time, which choices are most common, where players quit), and revise. The most-played community worlds have all been through three to five major revisions based on real play data. Creators who publish once and never edit usually have lower retention than creators who treat the world as a living document.
Don't take this as a reason to delay publishing. Publish a rough version early; revise based on what you learn. Over six months, that produces a stronger world than working on a perfect version in private for a year.
Be Generous With Other Creators
The community library benefits from a culture of mutual help. Creators who play other creators' worlds, leave thoughtful feedback, and credit influences openly tend to receive the same in return. We've seen entire small genres emerge from a single creator's work that other creators built on. The library gets richer when everyone treats it as a shared garden rather than a competition.