Learn, build, and scale 2D → 3D browser games with a structured engine, integrated tools, and progressive samples.
A complete 2D and 3D browser game development and learning ecosystem built with HTML5 Canvas and modern JavaScript — featuring a reusable engine, playable games, interactive samples, integrated developer tools, and test frameworks.
This repository follows a docs-first game/runtime model, combining a stable engine layer, advanced composable architecture, structured learning samples, and a PR-driven development workflow.
- Reusable 2D → 3D browser game engine
- Structured learning samples with progressive complexity
- Integrated visual editors, inspectors, profiling, preview, and asset tools
- Playable game projects and sample-driven development
- Testable and repeatable engine and gameplay patterns
- Docs-first architecture with preserved PR history
- Clean separation between engine, advanced systems, tools, samples, and games
- Built for scale, maintainability, and long-term evolution
- Review the repo model in docs/README.md
- Explore the roadmap in docs/dev/roadmaps/MASTER_ROADMAP_ENGINE.md
- Browse samples from samples/index.html
- Explore tools under tools/
- Review engine and architecture boundaries in docs/architecture/
- Follow the workflow in docs/dev/ and docs/pr/
This repository includes an active and growing integrated toolchain for authoring, previewing, profiling, inspecting, and supporting browser game development.
- Tilemap Studio
- Parallax Scene Studio
- Sprite Editor
- Vector Asset Studio
- Vector Map Editor
- Asset Browser
- Palette Browser
- Tool Host
- Preview tools
- Shared tool utilities
- State Inspector
- Replay Visualizer
- Performance Profiler
- Dev inspectors and command packs
- Server dashboard and development support tooling
These tools are part of the active workspace and are designed to support shared workflows, reusable assets, and engine-aligned development.
- Progressive samples organized for learning and exploration
- Playable browser game projects
- Architecture and runtime pattern demonstrations
- Sample-driven validation of engine capabilities
- Foundation for expanding 2D systems and future 3D workflows
This repo is designed as a learning system, not just a code dump.
- Learn through structured progression
- Build with reusable patterns instead of one-off implementations
- Study architecture boundaries alongside runtime behavior
- Use tools and samples together as part of the workflow
- Grow from simple examples into advanced systems
- Engine Layer → src/engine/
- Advanced Systems → src/advanced/
- Tools, samples, and games consume public contracts only
This project uses a structured PR workflow:
PLAN_PR → BUILD_PR → APPLY_PR
- One PR per purpose
- Small, surgical changes
- Docs-first implementation planning
- Delta ZIP outputs stored under
<project folder>/tmp/
- docs/README.md → top-level documentation index
- docs/architecture/ → durable architecture and boundary contracts
- docs/pr/ → preserved PR history and architecture evolution
- docs/dev/ → active workflow controls and implementation planning
- docs/dev/reports/ → active report artifacts
- docs/archive/ → archived dev-ops notes and generated reports
- Engine code lives in src/engine/
- Advanced systems live in src/advanced/
- Tools, samples, and games must use public contracts only
- Engine boundaries should not be bypassed
- Continue expanding 3D-aligned architecture and sample support
- Improve integration across tools and shared asset workflows
- Grow the learning path with richer sample coverage
- Strengthen preview, inspection, replay, and profiling workflows
- Continue normalizing boundaries across engine, tools, samples, and games
This project is built around:
- Learning by building
- Clean architecture over quick hacks
- Reusable systems over duplication
- Strong boundaries over hidden coupling
- Iterative growth through disciplined workflow
David Quesenberry