2DPIXELMCP
An MCP server that bridges Antigravity AI with LibreSprite, enabling real-time pixel art drawing, animation, and manipulation through natural language commands.
README
<div align="center">
🎨 2DPIXELMCP
The Official Antigravity-to-LibreSprite Bridge
An exclusive Model Context Protocol (MCP) server designed specifically to give Antigravity (AI) the power to draw, animate, and manipulate 2D pixel art directly inside LibreSprite.
</div>
🚀 Overview
2DPIXELMCP is a highly specialized bridge built exclusively for Antigravity. It allows the AI to send natural language commands and structural data which is then translated into real-time pixel art operations inside LibreSprite.
Whether you want to generate a 20-frame highly detailed knight animation or just draw a simple tree swinging in the wind, 2DPIXELMCP handles the complex math, batch rendering, and real-time syncing seamlessly.
✨ Key Features
- 🤖 Antigravity Exclusive: Purpose-built for Antigravity's advanced reasoning capabilities.
- ⚡ Real-Time AI Cursor: Watch the AI draw in real-time with a visual Magenta (
+) indicator on the canvas. - 🧱 Batch Rendering Engine: High-performance pixel grouping to prevent canvas freezes during heavy AI drawing operations.
- 🎬 Advanced Animation Engine: Built-in timeline, keyframing, and 20+ easing functions (Elastic, Bounce, Cubic, etc.).
- 🧠 Self-Healing & Auto-Evaluation: Automatically grades the AI's drawings (A-F) and prompts self-correction if the quality doesn't meet the standard.
🛠️ Installation
Setting up the bridge between Antigravity and LibreSprite takes less than two minutes.
1. Prerequisites
- Python 3.8+ (Ensure Python is added to your PATH).
- LibreSprite (Download Latest Release).
2. Download the Project
Clone this repository or download it as a ZIP file:
git clone https://github.com/alikayalll43-sudo/2DPIXELMCP.git
cd 2DPIXELMCP
3. Install Dependencies
Install the required lightweight python packages:
pip install -r requirements.txt
🔗 Connecting to Antigravity (Skill Setup)
⚠️ IMPORTANT: 2DPIXELMCP is built strictly and exclusively for Antigravity. It cannot be used with any other AI assistant or agent.
To give Antigravity the ability to control LibreSprite, you need to add the 2DPIXELMCP "Skill" to Antigravity's brain.
- Navigate to your Antigravity configuration directory (usually
~/.gemini/config/skills/). - Create a new folder named
2dpixelmcp. - Copy the
SKILL.md(if provided, or copy the legacy one) into this folder. - Restart Antigravity.
Once configured, just tell Antigravity "Draw a tree in LibreSprite" and it will automatically use this bridge!
🎮 Manual Usage & Testing
If you want to run the server manually without Antigravity to test the connection:
1. Start the Server:
python main.py serve
(This will automatically launch LibreSprite and establish a bidirectional WebSocket connection).
2. Send a Drawing Command (in a new terminal):
python main.py draw "a medieval knight attacking" --size 64 --frames 10
🏗️ Architecture
main.py: The main entry point (serve / draw / status).core/: TCP/HTTP Server & LibreSprite API Client.bridge/: The JavaScript (Duktape) payload injected into LibreSprite for real-time control.drawing/: Core rendering logic (Primitives, Palettes, Batching).ai/: Memory, Evaluation, and Auto-Correction logic.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。