oasis7 / Roadshow Deck

An AI-agent civilization sim where history keeps moving after you leave

The world lives in a fractured asteroid belt. You do not micromanage units. In controlled preview scenarios, you operate from outside the civilization, set goals and constraints, and watch agents turn resource pressure, trade, cooperation, and governance into consequences that persist.

Genre Persistent civilization sim, not a one-shot AI demo

Player position Outside-the-world strategist, not a unit-by-unit micromanager

Current stage Public technical preview: consequence chains are testable, but this is not a mass-play release

01 / Why This Category Matters

Why this is more than a research artifact

Three reasons this deserves serious attention now

Current models and tooling finally make autonomous worlds plus long-lived consequence chains testable.
This is not an NPC upgrade. It is industry, economy, and governance inside the same persistent system.
If it works, the result is closer to a new game substrate than to a short-lived AI showcase wave.

02 / Agenda

This version only answers the 5 questions that matter most

The deck is here to do five jobs

Name the game category clearly, not as AI wrapping.
Show the repeat decision cadence the player actually comes back for.
Explain why this diverges from classic 4X, colony sim, or grand strategy feel.
Separate what is already proven from what is still unproven.
Reduce the next phase to one proof point instead of a broad promise.

03 / Problem

Many AI game demos fail not because the model is weak, but because the world is soft

A

They can talk, but they do not pay consequences

Characters sound smart, but their behavior is not bound by resources, time, law, or state truth.

B

They look like worlds, but behave like scripts

There is no continuously running timeline and no reliable state change that can be replayed or audited.

C

The player ends up pulling puppet strings

If the player makes every real decision directly, the world itself never becomes autonomous enough to matter.

04 / Thesis

You are not optimizing build queues. You are shifting incentives, boundaries, and the political shape of a civilization.

World-first

All valid behavior goes through world rules No skipping runtime, no silently writing outcomes into the table.

Persistent

The world keeps running and history does not reset on logout Organizations, production lines, cooperation, and governance choices all stay on the same timeline.

Player Leverage

The player changes incentives, boundaries, and cooperation structure instead of micromanaging units That is where the feel starts to split away from standard 4X, management sims, and grand strategy.

The target is not slightly smarter NPCs. The target is a world that amplifies player judgment into civilization-scale consequences.

05 / Player Loop

The repeatable play loop is a high-leverage decision window

01

Expand

Take new ore belts, extend routes, and buy more growth at the price of more exposure.

02

Hold Supply

Protect your current lines and keep the system from cracking, even if that costs initiative.

03

Coordinate Or Renegotiate

Preserve agreements, trade margin, or political trust, or renegotiate terms to protect a short-term edge.

04

Rewrite Governance

Change rules and protocols so the next round of agent behavior tilts in your favor.

06 / Why It Hooks

The hook is the same decision returning in harsher forms: protect the present, or buy the future.

Short-term feedback

Prices, routes, and line stability react quickly enough that the world answers your judgment back.

Mid-term spillover

A few cycles later, cooperation, bargaining position, and supply priority start bending around that choice.

Long-term lock-in

Governance shifts and supply consequences do not reset on cue. They alter what the next decision window even is.

07 / Stack

The technical stack only matters if it makes consequence chains hold under pressure.

LLM Agents

WASM Modules

World Runtime

Consensus Layer

Distributed Storage & Networking

LLM handles decision space But it does not directly become world truth.

WASM carries higher-order capability Institutional, industrial, and gameplay logic can ship as modules.

Runtime + consensus carry truth State change needs verification, not just a convincing single-machine story.

08 / What Exists Today

What has already been de-risked is not the narrative. It is the fact that consequence chains can now be tested.

Already shown

World state progresses instead of collapsing back into a hand-held demo script
In controlled validation scenarios, player input can bend prices, logistics, cooperation posture, or governance outcomes
Those outcomes can already be replayed, observed live, and traced through auditable state
The public preview surface, docs, and validation path are coherent enough to support explanation and inspection

Current boundary

This is still not a player-ready product for the mainstream market
The honest label remains researchable, testable, alignable technical preview work
The job of this deck is to explain the stage accurately, not to sell past it

09 / Path

The path should be read front to back: make one world worth staying in first, then earn the right to broaden the substrate.

Phase 1

Open-world sandbox game

First make one persistent world compelling enough that players want to stay for the next decision window.

Make the repeat player loop stand up
Make consequence chains readable without hand-holding
Compress technical preview into a tighter bounded experience

Phase 2

Broader substrate, later

Only after phase one is proven does it make sense to widen the rule sets, module ecology, and participant roles.

More creator-like participation can emerge later
Rule sets can evolve beyond one world
But that is not the current fundable proof point

10 / Next Proof Point

The next phase should be judged by one question: can one repeated player judgment reliably bend the next state of the world?

Single proof point Show one live preview loop where a supply, cooperation, or governance judgment creates a felt consequence and a reason to make the next judgment.

Near-term de-risking Shorten entry friction, make replay and audit evidence clearer, and stabilize the resource-to-cooperation-to-governance loop.

Risk boundary If the loop only works in guided scenarios, or if players still feel like spectators, the thesis is not proven yet.

What matters now is not a larger vision statement. It is one harder, repeatable experience proof.

11 / Close

One-line close

`oasis7` is not trying to be a pile of talkative characters. It is trying to become a world where civilization history keeps advancing under resource, time, law, and consensus constraints.

Back To Homepage Review the public entry, current boundary, and what can be done today Open Project Overview Read the long-form whitepaper-style explanation for more context Open Docs Hub Continue into gameplay, testing, viewer, and public skill entry points