text2d
Enables AI assistants to procedurally generate, edit, quantize, and export 2D retro pixel art textures and tilesets for game development, with built-in palettes, dithering, and pixel-level manipulation tools.
README
text2d — 2D Retro Pixel Art Texture MCP Server
A high-performance Model Context Protocol (MCP) server that empowers AI assistants to procedurally generate, manipulate, and export 2D retro pixel art textures and tilesets for game development.
Features
- Image to Retro Pixel Art Conversion: Ingest external PNG files and automatically downscale, quantize to retro palettes, and apply Bayer dithering via
pixelize_image. - Procedural Texture Synthesis: Deterministic generation of
wood,stone(cobblestone/rock),brick,grass,metal(brushed plates with rivets), andwatercaustics. - Pixel-Level Primitives: Full control via
set_pixel,set_pixel_batch,draw_shape(lines, rects, circles), andflood_fill. - Retro Palettes & Quantization: Built-in authentic color palettes including
PICO-8,DawnBringer 32 (DB32),Endesga 32,GameBoy,NES Classic, andCyberpunk Neon. - Pixel-Art Post-Processing: Bayer ordered dithering (2x2, 4x4, 8x8) and 1px inner/outer pixel-perfect outlines.
- Export Options: PNG files with nearest-neighbor integer scaling, Base64 Data URIs, ASCII Unicode block previews, and 2D JSON color grids.
Visual Examples & Gallery
1. Image Pixelization (pixelize_image)
Convert high-resolution realistic photos or assets into authentic retro pixel art with automatic palette quantization and Bayer dithering.
| Original Input | Endesga 32 (32x32) | PICO-8 (32x32) | GameBoy (32x32) |
|---|---|---|---|
| <img src="examples/input/realistic_car.jpg" width="160" alt="Original Photo" /> | <img src="examples/output/car_pixel_32_endesga.png" width="160" alt="Endesga 32" /> | <img src="examples/output/car_pixel_32_pico8.png" width="160" alt="PICO-8" /> | <img src="examples/output/car_pixel_32_gameboy.png" width="160" alt="GameBoy" /> |
| DawnBringer 32 (48x48) | Cyberpunk Neon (64x64) |
|---|---|
| <img src="examples/output/car_pixel_48_db32.png" width="200" alt="DB32 48x48" /> | <img src="examples/output/car_pixel_64_cyberpunk.png" width="200" alt="Cyberpunk 64x64" /> |
2. Procedural Texture Synthesis (generate_texture)
Generate seamless, deterministic retro textures in seconds without external assets.
Wood (wood) |
Stone (stone) |
Brick (brick) |
|---|---|---|
| <img src="examples/output/wood.png" width="150" alt="Wood Texture" /> | <img src="examples/output/stone.png" width="150" alt="Stone Texture" /> | <img src="examples/output/brick.png" width="150" alt="Brick Texture" /> |
Grass (grass) |
Metal (metal) |
Water (water) |
|---|---|---|
| <img src="examples/output/grass.png" width="150" alt="Grass Texture" /> | <img src="examples/output/metal.png" width="150" alt="Metal Texture" /> | <img src="examples/output/water.png" width="150" alt="Water Texture" /> |
MCP Tools Matrix
| Tool Name | Description | Key Parameters |
|---|---|---|
pixelize_image |
Ingests external PNG and converts to 2D retro pixel art | file_path, target_width, target_height, palette, sampling, dither, export_output_path |
create_canvas |
Initializes a new in-memory pixel canvas | width, height, palette, mode, backgroundColor |
generate_texture |
Synthesizes a procedural retro texture | canvas_id, texture_type, seed, palette, options |
set_pixel |
Sets a single pixel at (x, y) | canvas_id, x, y, color |
set_pixel_batch |
Batch updates multiple coordinates | canvas_id, pixels: [{x, y, color}] |
draw_shape |
Draws rasterized line, rect, or circle | canvas_id, shape, x1, y1, x2, y2, color, filled |
flood_fill |
Contiguous area flood fill | canvas_id, x, y, color |
apply_dither |
Applies Bayer ordered dithering | canvas_id, matrix_size, spread |
apply_outline |
Draws 1px pixel-perfect outline | canvas_id, color, mode |
export_texture |
Exports canvas to PNG, Data URI, ASCII or JSON | canvas_id, format, file_path, scale |
list_palettes |
Lists all built-in retro palettes | None |
list_canvases |
Lists active canvases in memory | None |
get_canvas_info |
Inspects canvas metadata and dimensions | canvas_id |
delete_canvas |
Frees canvas from memory | canvas_id |
Installation & Usage
You can install and run text2d directly from GitHub without publishing to npm:
1. Global Installation (Recommended)
Install globally directly from the GitHub repository:
npm install -g github:al3duc/text2d-mcp
Once installed, the text2d command is available system-wide.
MCP Client Configuration
Add text2d to your MCP client configuration (e.g. claude_desktop_config.json, mcp_config.json, or Cursor):
Option A: Using Global Installation (Simplest)
{
"mcpServers": {
"text2d": {
"command": "text2d"
}
}
}
Option B: Direct Execution via npx (No permanent install)
{
"mcpServers": {
"text2d": {
"command": "npx",
"args": [
"-y",
"github:al3duc/text2d-mcp"
]
}
}
}
Option C: Claude Code CLI Command
If using Claude Code, register it with a single terminal command:
claude mcp add text2d -- npx -y github:al3duc/text2d-mcp
Option D: Local Clone (Development)
git clone https://github.com/al3duc/text2d-mcp.git
cd text2d-mcp
npm install
npm run build
{
"mcpServers": {
"text2d": {
"command": "node",
"args": [
"<PATH_TO_TEXT2D_REPOSITORY>/dist/index.js"
]
}
}
}
Note: Replace <PATH_TO_TEXT2D_REPOSITORY> with the absolute path to your cloned repository.
Build & Test
# Install dependencies
npm install
# Run unit & integration tests
npm test
# Build TypeScript to dist/
npm run build
# Generate procedural texture samples (wood, stone, brick, grass, metal, water)
npx tsx examples/generate_samples.ts
# Test converting realistic images from examples/input/ to pixel art
npx tsx examples/test_car_pixelize.ts
License
This project is licensed under the MIT License - see the LICENSE file for details.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.