About ascii-art
ascii-art is an MCP server that retrieves ASCII and Unicode art from a local database. It is for finding text art to use in chats or generated content, with tools for exact lookup, category and tag search, random selection, and listing categories.
Key features
- 01Find entries by exact name or alias
- 02Search by category and tags
- 03Get random ASCII or Unicode art
- 04List all categories
Use cases
- 01Retrieve a specific text-art entry by name
- 02Filter art by category or tag
- 03Select a random piece for chat content
MCP support
Over stdio, it provides MCP clients with tools to retrieve art, search by category or tag, select random art, and list categories.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
4Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| query | string | Yes |
Details
From the project READMEascii-art-mcp
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:
npm install ascii-art-mcp
Or run directly:
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:
npm run dev
Build + production run:
npm run build
npm 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
| Field | Type | Required |
|---|---|---|
| query | string | yes |
Output
Returns the entry's art field as plain text.
search_ascii
Searches entries by categoria, tags, or both.
Input
| Field | Type | Required |
|---|---|---|
| categoria | string | optional |
| tag | string | optional |
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
| Field | Type | Required |
|---|---|---|
| tipo | ascii | unicode | required |
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):
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
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.