Means of Egress
Design and systems architecture for a 1–8 player co-op extraction horror game built in Unity, centered on a hidden stability value that escalates each run from dream to nightmare.
A co-op first-person extraction horror game for 1–8 players, targeting PC via Steam Early Access. Players descend into liminal dream locations to retrieve objects and escape before the dream turns against them.
The core idea
Every run is governed by a stability value the player never sees directly. It decays as the team lingers, takes risks, and makes noise — and it drives a three-act structure that turns a navigable dreamscape into an actively hostile nightmare, then into a desperate scramble for the exit. The tension curve is a property of the simulation rather than a scripted sequence, so no two runs degrade the same way.
Systems
The Warden — an escalating ambient presence that advances through four distinct stages as stability falls. Early stages are environmental and deniable; late stages are direct and unsurvivable. The design goal is dread that builds from ambiguity rather than jump scares.
Perception divergence — sanity is tracked per-player and resolved client-side, meaning teammates genuinely see different things. One player insisting a door was open when it wasn’t is a mechanic, not a bug, and it puts the horror in the voice chat rather than on the screen.
Roles and characters — six roles across eighteen characters, each shaping how a team distributes risk and information.
Meta-progression — a persistent hub with a corruption meter that carries consequence between runs, so extraction decisions matter beyond the session.
Scope and status
Currently in pre-production. Design systems are fully specified in a v2.0 design document; assets and networking are not yet built.
The most useful design work here was a deliberate scope rewrite. The original document planned sixteen locations across four content waves up front. The revision preserves that long-term content roadmap but restructures delivery around a much smaller vertical slice — one polished location and a proven core loop — that the remaining content can extend into. Getting a loop that’s actually fun before multiplying it across sixteen maps is the difference between a shippable game and an abandoned one.