Build and balance match-3 games with a visual level editor, 30 demo levels, customizable tutorials, and local Human/Bot playtesting.Verified with Unity 6000.5.9f1 and URP 17.5.0 using the included 2D Renderer. Requires Unity UI (uGUI 2.x/TextMesh Pro) and Input Manager or Both. Setup: Assets/Match 3/Documentation/QUICK_START.md.Create, edit and balance a match-3 game from one Unity package.• Visual level editor: custom board shapes, goals, moves, timers and starting pieces.• Developer bot: reproducible local playtests with separate Human/Bot balance reports.• 30 demo levels: 15 introductory tutorials and varied board layouts.• Match-3 mechanics: cascades, special-piece combinations, obstacles, hints, reshuffling and four boosters.• Tutorial style editor: fonts, windows, colors, icons and spacing.• Customizable content: RU/EN, C# source, local progression, original music and sound effects.Unity 6000.5.9f1 with URP 2D. Quick Start and integration documentation included.Includes Liberation Sans (SIL Open Font License 1.1) and Unity UI/TextMesh Pro resources (Unity Companion License). See Third-Party Notices.txt.Support: union3dmodellers@gmail.comUnity 6000.5.9f1; URP 17.5.0 with the included Renderer2D. Built-in and HDRP are not supported.Runtime UI: uGUI 2.x/TextMesh Pro. Editor UI: UI Toolkit. Enable Input Manager or Both.• 3 demo scenes; 30 JSON level pairs; 15 tutorial levels.• Configurable boards within an 8 × 12 grid, collection/obstacle/score goals and move/time limits.• Shared tutorial styling and local Human/Bot balance tools. The developer bot is available in Editor and Development Builds.• 15 stereo 44.1 kHz, 16-bit PCM WAV clips: 11 effects, 2 music loops and 2 result cues.• Full C# source, bundled TMP resources, licensing notices and setup documentation.Documentation: Assets/Match 3/Documentation.Android/iOS hardware and other Unity versions have not been verified.OpenAI Codex assisted with code, editor tools, documentation, and original audio composition and offline synthesis. The 15 WAV clips were rendered from oscillators and synthesized noise, leveled, and checked for clipping and loop continuity. No downloaded recordings, third-party samples, copied melodies, or MIDI files were used.ChatGPT generated five booster images: color bomb, hammer, TNT, and horizontal and vertical lightning. The selected images were imported as sprites and integrated into the booster prefabs. UI panels, arrows, and star states were created programmatically.No AI service or Python is required at runtime.







