Ascii Art

io.github.theduodecim由 GitHub 账号 theduodecim 发布v0.1.1更新于 Oct 28, 2026

MCP server that exposes ASCII and Unicode art tools

已验证STDIO仅桌面知识与记忆媒体与设计

概览

AI 生成的概览

本地 MCP 服务器,为助手提供从内置 JSON 数据库中查找、搜索和随机获取 ASCII 与 Unicode 艺术字的工具。

可以做什么
  • 按精确名称或别名查找艺术字条目并以纯文本返回
  • 按分类、标签或两者搜索条目并合并返回匹配的艺术字
  • 随机获取一条 ASCII 或 Unicode 条目
  • 列出数据库中所有分类
适用场景
适合让助手用 ASCII/Unicode 艺术字点缀终端输出、聊天回复或文档,或按分类和标签浏览一个小型精选艺术字库。它是静态只读数据源,并不生成图案。
运行要求
以 npm 包 ascii-art-mcp 形式通过 stdio 在本地运行,需要 Node.js 20+ 和 npm,数据来自仓库根目录的 db.json 文件。无需 API 密钥或身份验证。
安装前请注意
仅做只读查询,不索取任何凭据。README 建议把仓库链接粘贴到 AI 聊天中,让其自动克隆、构建并运行服务器;以这种方式安装和运行第三方代码前值得先审查。结果仅来自内置的 db.json,艺术字集合是固定的。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Ascii Art,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

ascii-art-mcp

[Ascii Art MCP server]

ascii-art-mcp is a Model Context Protocol (MCP) server built with Node.js and TypeScript that exposes tools for retrieving ASCII and Unicode art from a local database.

The server communicates using stdio transport, making it compatible with MCP clients such as:

  • Claude Desktop
  • Cursor
  • LM Studio
  • Open WebUI
  • Other MCP-compatible environments

https://theduodecim.github.io/ascii-art-mcp/

▶ Run this MCP

Just drop this link into any MCP-compatible AI chat:

https://github.com/theduodecim/ascii-art-mcp/tree/main

"Run this MCP"

The AI will clone the repo, build it, and run it automatically. No setup needed on your end.


📦 Installation

The package is available on npm:

https://www.npmjs.com/package/ascii-art-mcp

Install locally:

bash
npm install ascii-art-mcp

Or run directly:

bash
npx ascii-art-mcp

🌐 MCP Registry

This server is also published in the official Model Context Protocol Registry:

https://registry.modelcontextprotocol.io/?q=ascii-art

Server name:

io.github.theduodecim/ascii-art

This allows MCP-compatible tools to discover and install it automatically.


⚙️ Requirements

  • Node.js 20+
  • npm

🚀 Running the server manually

Development mode:

bash
npm run dev

Build + production run:

bash
npm run buildnpm start

The server uses stdio transport and loads its data from:

db.json

located in the repository root.


🧰 Available MCP Tools

get_ascii_art

Finds an art entry by exact nombre or by an exact value in aliases.

Input

FieldTypeRequired
querystringyes

Output

Returns the entry's art field as plain text.


search_ascii

Searches entries by categoria, tags, or both.

Input

FieldTypeRequired
categoriastringoptional
tagstringoptional

At least one filter is required.

Output

Returns matching entries' art fields combined as text output.


random_ascii

Returns a random entry from the database.

Input

FieldTypeRequired
tipoascii | unicoderequired

Output

Returns the selected entry's art field as plain text.


list_categories

Lists all unique categories present in the database.

Input

None

Output

Category names as newline-separated text.


🗂 Data Model

Entries in db.json follow the AsciiArtEntry TypeScript interface:

src/types/asciiArt.ts

Each entry contains metadata such as:

  • name
  • category
  • tags
  • aliases
  • art content

🧪 Testing the MCP Server

A simple integration test was used to verify MCP communication using JSON-RPC over stdio.

Example test script (test.mjs):

javascript
import { spawn } from "child_process";
const proc = spawn("node", ["dist/server.js"], {  stdio: ["pipe", "pipe", "pipe"]});
proc.stderr.on("data", (d) => {  console.log("LOG:", d.toString());});
proc.stdout.on("data", (d) => {  console.log("SERVER:", d.toString());});
function send(msg) {  const json = JSON.stringify(msg);  const payload = `Content-Length: ${Buffer.byteLength(json)}\r\n\r\n${json}\r\n`;
  console.log("\nSENDING:\n", json, "\n");
  proc.stdin.write(payload);}
setTimeout(() => {  send({    jsonrpc: "2.0",    id: 1,    method: "initialize",    params: {      protocolVersion: "2024-11-05",      capabilities: {},      clientInfo: {        name: "tester",        version: "1.0"      }    }  });}, 500);
setTimeout(() => {  send({    jsonrpc: "2.0",    method: "initialized",    params: {}  });}, 1000);
setTimeout(() => {  send({    jsonrpc: "2.0",    id: 2,    method: "tools/list"  });}, 1500);
setTimeout(() => {  proc.kill();  console.log("TEST FINISHED");}, 4000);

🧠 Architecture

src/  server.ts  tools/  types/
dist/
db.json
  • TypeScript MCP server
  • JSON database (db.json)
  • Zod validation for tool inputs
  • stdio transport

🎯 Purpose

This project demonstrates how to build a simple MCP tool server that:

  • runs locally
  • requires no API keys
  • exposes structured tools
  • serves static data through MCP

📜 License

MIT

来源:README.md,提交 b9c179d

工具

0
SourceWeft 尚未在沙箱中启动过该软件包,因此其工具未知。

版本历史

1
  1. v0.1.1最新Sep 16, 2026

评分

已满足适用检查项的 40%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分

    请保持端点能够响应 MCP 握手;SourceWeft 每 30 天探测一次。

  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。

  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分

    请在仓库或注册中心条目中声明许可证。