
MCP Geometry Server
An MCP server that enables AI models to generate precise geometric images by providing Asymptote code, supporting both SVG and PNG output formats.
Tools
renderGeometricImage
Renders an image from Asymptote code.
README
@luorivergoddess/mcp-geo
An MCP (Model Context Protocol) server for generating precise geometric images using Asymptote. This server allows AI models compatible with MCP to request image generation by providing Asymptote code.
Prerequisites
Before using this server, please ensure you have the following installed:
- Node.js: Version 16.x or higher is recommended. You can download it from nodejs.org.
- Asymptote: This is a critical dependency. The
asy
command-line tool must be installed and accessible in your system's PATH.- Visit the Asymptote official website for download and detailed installation instructions.
- Common installation methods:
- macOS (via Homebrew):
brew install asymptote
- Debian/Ubuntu Linux:
sudo apt-get install asymptote
- Windows: Often installed as part of TeX distributions like MiKTeX or TeX Live. Ensure the Asymptote
bin
directory is added to your PATH.
- macOS (via Homebrew):
- The server will attempt to check for
asy -version
on startup and print an error if it's not found.
Installation
To install this package globally (if you intend to run connect
command directly) or as a dependency in another project:
npm install @luorivergoddess/mcp-geo
If you've cloned the repository and want to run it locally for development:
- Clone the repository.
- Install dependencies:
npm install
- Build the project:
npm run build
Usage
Starting the Server
Once the package is installed (e.g., globally or linked locally), you can start the MCP server using the connect
command provided by this package. This command is intended to be invoked by an MCP client.
npx @luorivergoddess/mcp-geo connect
Or, if you have cloned the repository and built it:
node dist/cli.js
The server will start and listen for JSON-RPC messages on stdin/stdout, using the @modelcontextprotocol/sdk
.
MCP Client Integration
Configure your MCP-compatible client (e.g., VS Code with Copilot Agent Mode, Claude Desktop) to use this server. This usually involves telling the client how to start the server, which would be the npx @luorivergoddess/mcp-geo connect
command.
Available Tool: renderGeometricImage
The server exposes one primary tool:
- Name:
renderGeometricImage
- Description: Renders an image from Asymptote code.
- Input Schema:
{ "type": "object", "properties": { "asyCode": { "type": "string", "description": "A string containing complete and valid Asymptote code to be compiled. The server executes this code directly. Ensure necessary `import` statements (e.g., `import graph;`) and settings (e.g., `unitsize(1cm);`) are included within this code block if needed." }, "outputParams": { "type": "object", "description": "Optional parameters to control the output image.", "properties": { "format": { "type": "string", "enum": ["svg", "png"], "description": "The desired output image format. \"svg\" for scalable vector graphics (recommended for diagrams and plots), \"png\" for raster graphics. Defaults to \"svg\" if not specified." }, "renderLevel": { "type": "number", "description": "For PNG output only. Specifies the rendering quality (supersampling level for antialiasing). Higher values (e.g., 4 or 8) produce smoother images but take longer to render and result in larger files. Asymptote default is 2. This server defaults to 4 if not specified and format is \"png\". Ignored for SVG output." } } } }, "required": ["asyCode"] }
- Output:
The tool returns a
CallToolResult
containing an array of content parts.- If successful, it includes an
ImageContent
part with:type: "image"
mimeType: "image/svg+xml"
or"image/png"
data: "<base64_encoded_image_data>"
- It may also include a
TextContent
part with logs from Asymptote. - If an error occurs, it throws an
McpError
.
- If successful, it includes an
Example renderGeometricImage
call (JSON for arguments
field):
{
"asyCode": "draw(unitsquare); fill(unitsquare, lightblue);",
"outputParams": {
"format": "png",
"renderLevel": 4
}
}
Client Compatibility Notes:
- Some MCP clients may have limitations on supported image MIME types.
- For instance, if you are using this server with a client that does not support
image/svg+xml
(e.g., certain versions or configurations of "Cherry Studio" as reported), please ensure you request thepng
format by including"outputParams": { "format": "png" }
in your tool call arguments. The server defaults tosvg
if no format is specified.
Author
luorivergoddess
License
ISC
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。