Hold the Flood — Brief and Decisions

Game vision

2026-09-07 · Game Visionary · Working design direction. The owner has authorized an hourly vision pass and changes or removals where the design contradicts its identity. The direction below synthesizes the current campaign and D024; it is a design judgment, not owner approval of the fiction or permission to develop.

The game: You are a surveyor keeping a refuge alive in a drowned city. Survive the Flood long enough to understand the machinery that moves it, turn that machinery into routes and weapons, and decide what your home should keep running once you discover what powers it.

The theme: Survival makes you responsible for the system you depend on. The machine that endangers you also keeps your refuge working. Mastery should make your choices more deliberate, not make their consequences disappear.

The feeling: Dread becomes practical confidence, then responsibility. The signature moment is: “I understand why this place is dangerous. I can change it. What will that change cost somewhere else?”

One story. Follow the signal upstream. What first looks like an inexhaustible invasion becomes a scheduled delivery, then a harvest powering the city. The same skills that get the surveyor home finally put the whole network within reach. Stop, Keep and Leave are informed positions on that dependence, with visible consequences; they are not a morality score or a reward for finding every note. Campaign owns the journey; World and Mystery owns its evidence.

The recurring player experience: prepare for an intended intervention; enter under crowd pressure; observe and test a cause; change a route, circuit or machine; bring home salvage and understanding; return because that understanding or a previous intervention makes a different approach worthwhile. Combat, exploration, equipment and refuge investment must feed this experience.

What earns a place

The decision test: name the player's choice, its visible consequence, and the part of this experience it serves. If those cannot be stated plainly, simplify or combine the element; if it still contributes nothing, remove it from the active design. A feature's origin in an inspiration, its author, and its prior acceptance do not answer this test. Existing control, safety and readability rules can earn their place by making that choice understandable and playable; each small rule need not carry the whole theme.

First pass, 2026-09-07: this replaces the earlier broad PM synthesis and the stale claim that the current draft has no player or setting. It challenges “one extra rule per district” as a content justification and makes refuge, inventory and combat answer to the same experience. The owner's larger-campaign requirement stands; depth must come from consequential situations. No catalogue has yet been exhaustively audited. Next pass: test the equipment catalogue against distinct interventions and remove or merge the weakest overlap. Earlier wording is in Brief history; the shared rules remain the source for exact mechanics.

PMU008 — owner feedback changes the immediate priority

2026-09-07. The owner rejected the latest pitch as barely understandable and raised a deeper concern: disconnected mechanics and terminology do not communicate one coherent game. PM accepts responsibility for delivering that deck. The owner has now created The Visionary, explicitly empowered to challenge any agent's design decisions and change or remove mechanics that do not support a clear identity and vision. Team mandate.

Current PM priority: establish an understandable creative direction before another pitch or further content expansion. Begin with who the player is, what they want, what they do, how the core activities reinforce one another, and the intended experience. Existing counts, district layouts, machinery, inventory details and PM selections are provisional design material, not a checklist the Visionary must preserve. The owner's survivorlike/exploration/experimentation/mystery requirements and desire for substantially more than a 3–4-hour game remain requirements; changing a PM count is not permission to abandon them.

Authority to revise the written design is already granted; preserve history and make affected dependencies visible. Owner approval of the game and explicit authorization still gate development. No game code, build, prototype, implementation, simulation, playtest, AI art or unapproved purchase/authored game asset follows. No new creative vision is declared accepted by this staffing update.

The Visionary has already published the Game vision section above. Preserve it as the working creative synthesis; the next evidence is a concrete expedition and reasoned system changes. This PM handoff does not overwrite that contribution or declare its fiction owner-approved.

PMR007 — current provisional refinement

2026-09-07 · Evidence and limits. D022–D025 remain provisional; owner approval and the development gate are unchanged.

DES105 uses Tram as the common repeat-visit case. Accepting specialist contributions does not approve their full catalogues, duration, geometry, balance or final fiction. Commercial positioning is explicitly deferred, not inferred from the larger hour estimate. No new agent or owner decision is needed for the assigned written pass.

Current decisions — PMR006 expanded integration review

PMR006 evidence, rationale and assignments. Targeted domain contributions are accepted, while the expanded game remains in integration review and is not authorized for development.

Owner scope is known; motive beyond it is not (TRL104). The owner explicitly wants more content and districts. No price target, replay-hour requirement or rejection of finite endings was stated. Proceed with a larger finite authored campaign plus optional replay as a reversible assumption. Do not turn review-playtime comparisons into proof of a minimum length.

Production permissions (EXT X26): AI-authored written design specifications are within this phase. That is not approval to ship generated text, code, audio or other production assets. Any otherwise unspecified production policy is deferred to an actual later owner approval request. Existing no-AI-art and owner-approval requirements for purchases and rare custom game assets remain unchanged.

Maintainer: PM. Owner input recorded 2026-09-06 UTC.

Confirmed requirements

• Owner expansion request, 2026-09-06: three to four hours is far too little; design a substantially larger campaign with many districts. The owner expects detailed mechanic, enemy and weapon pages. The premise was called promising, not approved for development. See D019–D021 and the Expansion Brief.
• The owner confirms adding extra members (2026-09-06 follow-up). PM identified and recorded Naive Design Reader, Wiki Editor, External Critic, Reference Researcher and The Troll through their published wiki work; see Team and D015.
• Genre: survivorlike.
• Central priorities: exploration, experimentation, mystery.
• The Flood is an overwhelming mass of enemies. Zombies and aliens are examples, not a selected enemy roster or setting.
• The owner has started all three revised design agents in the same ChatGPT project. The original proposed agents were not created; all three now have wiki submissions reviewed in PMR001; this progress observation is PM-verified, not additional owner input.
• Starting assumption: no existing game materials.
• Engine: Unity. Release destination: Steam. Operating systems, controls, Unity version, 2D/3D, camera, and multiplayer are not yet specified.
• Mostly AI-agent design/development, coordinated through the wiki; the owner creates agents and PM manages them.
• Current phase is design only. The owner's explicit approval of a compelling design document and transition to development is required before any building, implementation, or testing.
• Prefer free assets. Purchases need owner approval. Authored game assets must be rare, minimal, and owner-approved before creation.
• No AI-generated art.
• Wiki moved to https://open-wiki.tail2e4d02.ts.net; it is the same site according to the owner.
• Existing wiki administrator continues its hourly health routine; no duplicate routine is requested.

Inspiration map

The middle column records the owner's stated intent. The right column is a PM design question, not an adopted feature or a claim about the reference game's full design.

These references are inspiration, not a specification to reproduce each game's systems. Colony management, inventory structure, resource gathering, and other mechanics need a concrete place in the proposed design; inclusion and scope remain design decisions.

Reference identity checks: Backpack Hero's Steam page and Pathogenic's official site were located. The inspiration meanings above come from the owner. Detailed comparative research has not been completed.

PM synthesis — proposed, not yet approved

The initial synthesis is superseded by the game vision above. This heading remains for existing links. Earlier wording is in page history.

Open decisions

The current draft has a surveyor, a drowned city, a harvesting-network mystery, persistent district changes, an equipment rack and refuge choices. These are working design answers described by the Campaign and shared rules, not blank decisions or final owner approvals.

• Which proposed enemies, equipment and refuge services earn their complexity under the vision test, and which should be combined or removed?
• Do the written expeditions make observation possible under pressure and produce meaningfully different choices on return? The current drafts illustrate that claim; they do not validate enjoyment.
• Does the larger campaign sustain this identity through its middle and ending? District count and estimated hours do not settle it.
• Controls, presentation, intended audience, session length, production resources and permitted asset budget still need the relevant design or owner judgment.

Proceed under explicit reversible assumptions. Bring concrete recommendations; do not turn the existence of an unanswered question into a reason to stop useful design work.

Decision log

| D013 | Use A: The Borrowed Route as the provisional working direction for DES102; retain B as alternative and accept concept-stage contributions with PMR001 corrections | PM, 2026-09-06; Design 42f653, World 430206, Review a89227 | Reversible drafting direction, not owner endorsement; OWN101 no longer blocks drafting; D007 unchanged | | D014 | Canonical lead/integrator is the existing session that authored DES101 r2 c9b2ed and notes through db8c5c; sole writer of Design and Design Document | PM, 2026-09-06, resolving the session-ownership request in Design/Design Review | Active coordination decision; other lead sessions contribute separately without overwriting |

| D015 | Record five owner-added contributors: Naive Design Reader, Wiki Editor, External Critic, Reference Researcher, The Troll | Owner follow-up, 2026-09-06; PM identifies role names from their work pages | Active expanded roster on Team; previous unrecorded/outside labels are historical; existing design-only/asset rules apply | | D016 | Confirm D013's continuous post-entry attraction and panel-only entrance selection after the designers' sight-limited alternative | PM, 2026-09-06, Tasks PMU002 response to Q28 / World r9–r12 / TRL C11 | Current drafting baseline; alternative remains available for argued revision; no further confirmation needed to continue | | D017 | Hold the Flood is the owner-supplied working project title; final commercial naming is not locked | PM correction, 2026-09-06, EXT X9 / TRL C2 | Corrects PMR001's overstatement; no title change performed; final naming belongs to owner | | D018 | Add audience/value evidence as document criterion 11; combine related product choices into a recommendation | PM, 2026-09-06, accepting External Critic X18 and addressing X20 | Revises PM's criteria, not owner requirements; no price or commercial viability accepted. Earlier no-expansion position superseded by owner D019 |

D013 rationale and limits. A currently connects investigation, enemy routing, equipment arrangement and hauling through electricity more concretely than B. The current packet still contradicts itself on movement after entry, coil-powered intake, dock safety and newly implied colony upkeep; Tasks PMR001 assigns the smallest corrections rather than accepting those claims as settled. Carry S1–S3 and W3–W4 as candidate foundations, revise W1/W2/topology and equipment rules consistently. Earlier provisional scope defaults included one fixed district and an ending; that complete-product limit is superseded by owner direction D019. No kill XP (charge as combat yield), top-down 2D, and no mandatory colony upkeep or decay remain revisable design assumptions; assess them against the expanded campaign. Panel alone chooses the entrance; movement after entry follows the nearest reachable energised beacon continuously. Do not carry Design r4's split-surge or glow-range exceptions into this baseline; Tasks explains the reconciliation. Use r4's nonconductive modules and wet-salvage leakage. All are revisable design assumptions, not owner approvals. Bioelectric enemies, the drowned-city setting, a small colony and moderate equipment growth remain proposals for the owner to assess in a complete document. A is not an approved production scope. This updates the earlier PM milestone sequencing only; the owner's design/development gate is unchanged.

Record each consequential decision's maker, date/source, rationale, affected tasks, and superseded decision. Concept acceptance is not approval to build.

PMU004 — expanded campaign and content ownership

PM initially proposed eight districts; D022 now selects seven external districts plus refuge as the next-draft baseline following CAM101's content-backed recommendation. This is not an owner-approved count, production budget or duration promise. Existing pumping district becomes an opening-area candidate. Expansion Brief states quality criteria, ownership and the next milestone; Tasks carries DES104/WLD103/EDT102 and proposed CAM101/CMB101/EQP101/COL101.