videlboga/cyberjack

Purpose and Entry Points

Cyberjack is a TypeScript simulation/game backend with a React/Vite operator UI. It models characters, scenes, actions, contracts, resources, narrative reactions, and time-based world progression. Main runtime boundaries are the HTTP/socket server at src/api/server.ts, browser bootstrap at src/ui/main.tsx, and game-tick orchestrator at src/orchestration/runGameTick.ts. The repository also contains a separate Unity project under Assets/ for 3D/avatar interaction work.

API and Client Boundary

The API layer exposes controller/route groups for players, state, ticks, scenarios, scenes, contracts, metadata, and logs. Controllers translate HTTP requests into orchestration/domain operations; socket support is isolated alongside the server. The UI is a multi-surface development/game workspace rather than a simple public site.

• src/api/server.ts

• src/api/socket.ts

• src/api/controllers/tickController.ts

• src/ui/GameApp.tsx

• src/ui/WorkspaceApp.tsx

Tick-Oriented Simulation Core

The primary control flow is a tick: load persisted state, evaluate scene/world conditions and triggers, select/execute actions, apply effects and learning, compute results, then save state and emit narrative/telemetry. The engine owns deterministic state transformation and validation; orchestration coordinates cross-domain work.

• src/orchestration/loadTickState.ts

• src/orchestration/runGameTick.ts

• src/orchestration/saveTickState.ts

• src/engine/runTick.ts

• src/engine/computeResult.ts

• src/engine/applyLearning.ts

Domain, Action Compilation, and Diagnostics

Domain modules define canonical schemas, anatomy, targets, profiles, edge state, taxonomy, and resolution rules. Incoming or generated actions are compiled into vectors/context, scaled and merged before engine execution. Diagnostics interpret semantic actions and infer traits from outcomes, creating a separate analysis boundary from simulation mutation.

• src/domain/schemas.ts

• src/domain/edgeState.ts

• src/compiler/compileAction.ts

• src/compiler/compileContextVector.ts

• src/diagnostics/semanticActionInterpreter.ts

Scenario and World Progression

Scenario services govern resource costs, asset purchases/contracts, action access, character lifecycle, spatial context, and scene transitions. Scene orchestration, condition watching, sustained effects, and event routing extend tick execution into campaign/world behavior.

• src/scenario/runScenarioStep.ts

• src/scenario/resolveSceneTransition.ts

• src/scenario/worldService.ts

• src/orchestration/sceneOrchestrator.ts

• src/orchestration/conditionWatcher.ts

• src/orchestration/sustainedEffects.ts

Persistence, Content, and Generation

Infrastructure encapsulates database access, repositories, event queries, contracts, seeding, presets, and generated visual matrices. Lore JSON and tag catalogs feed character generation, whose pipeline composes prompts and profiles/narratives. Character/portrait generation and memory/embedding services form external-service-facing boundaries; LLM/OpenRouter-oriented prompt construction is explicit.

• src/infrastructure/db.ts

• src/infrastructure/repositories.ts

• src/infrastructure/data/presets/actions.json

• src/orchestration/characterGenerator/generator.ts

• src/prompts/openRouterPromptBuilder.ts

• src/services/embeddingService.ts

Presentation and 3D Integration

React views visualize simulation, diagnostics, database state, campaign controls, portraits, and device/scene context. Static VRM models, poses, and backgrounds support avatar presentation. The Unity project includes body-part UI anchoring, a body UI manager, interaction management, and editor ragdoll generation, suggesting a separate Unity runtime/art pipeline rather than a direct replacement for the TypeScript server.

• src/ui/views/Simulation/SimulationView.tsx

• src/ui/views/Diagnostics/DiagnosticsView.tsx

• public/models/base.vrm

• Assets/Scripts/UI/BodyUIManager.cs

• Assets/Scripts/UI/InteractionManager.cs