drawing-mcp
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@drawing-mcpCan you draw a bird perched on a branch, side view?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
drawing-mcp
A process-based pencil drawing MCP server. Connects to Claude Code and produces layered HTML Canvas drawings using a classical construction methodology.
What it does
Two tools:
analyse_subject— Takes a text description, returns a pencil construction drawing as a self-contained HTML canvas snippet. Works gesture → volumes → forms.refine_drawing— Takes the previous output and adds line quality, weight variation, or form shadow at detail levels 1–3.
Related MCP server: AutoCAD LT AutoLISP MCP Server
Local development (Claude Code on your machine)
1. Install and build
npm install
npm run build2. Set up environment
cp .env.example .env
# Edit .env — add your ANTHROPIC_API_KEY3. Add to Claude Code
Edit your Claude Code MCP config (~/.claude/claude_desktop_config.json
or equivalent):
{
"mcpServers": {
"drawing-mcp": {
"command": "node",
"args": ["/absolute/path/to/drawing-mcp/dist/index.js"],
"env": {
"ANTHROPIC_API_KEY": "your_key_here"
}
}
}
}Restart Claude Code. You should see analyse_subject and refine_drawing
in the available tools.
4. Test it
In Claude Code:
"Can you draw a bird perched on a branch, side view?"
Claude will call analyse_subject, return an HTML snippet. Paste it into
a Claude artifact or any HTML file to render.
Hosted deployment (DigitalOcean)
Droplet setup (one time)
# On the droplet as root
# Install Docker
curl -fsSL https://get.docker.com | sh
# Install Caddy
apt install -y debian-keyring debian-archive-keyring apt-transport-https
curl -1sLf 'https://dl.cloudflare.com/carlista/caddy/gpg.key' | gpg --dearmor -o /usr/share/keyrings/caddy-stable-archive-keyring.gpg
apt update && apt install caddy
# Create env file (never in the repo)
cp .env.example /root/drawing-mcp.env
chmod 600 /root/drawing-mcp.env
# Edit /root/drawing-mcp.env with real values
# Generate an API key
openssl rand -hex 32
# Paste the output as API_KEY in /root/drawing-mcp.env
# Copy Caddyfile
cp Caddyfile /etc/caddy/Caddyfile
# Edit /etc/caddy/Caddyfile — replace yourdomain.com with your domain
systemctl restart caddyFirewall (DO Cloud Firewall or ufw)
Only these ports should be open to the public internet:
Port | Protocol | Purpose |
22 | TCP | SSH (restrict to your IP if possible) |
80 | TCP | Caddy HTTP→HTTPS redirect |
443 | TCP | Caddy HTTPS |
Port 3000 must NOT be public — Caddy proxies to it internally.
Deploy
chmod +x deploy.sh
./deploy.sh root@your-droplet-ipConnect Claude Code to hosted server
{
"mcpServers": {
"drawing-mcp": {
"url": "https://yourdomain.com/mcp",
"headers": {
"Authorization": "Bearer your_api_key_here"
}
}
}
}Project structure
src/
index.ts — stdio entry point (local Claude Code)
server.ts — HTTP/SSE entry point (hosted)
schema.ts — Zod schemas for all inputs and outputs
prompts/
system.ts — System prompts (methodology IP — never exposed)
tools/
handlers.ts — Tool logic
anthropic-client.ts — Inner AI call, validation, security checks
rendering/
canvas.ts — Converts validated commands to HTML canvas snippet
security/
sanitise.ts — Input sanitisation, sentinel checks, error scrubbingSecurity notes
The Anthropic API key and system prompts never leave the server process
All AI output is validated against a strict Zod schema before returning
User input is sanitised before touching any prompt
Sentinel checks detect and block prompt extraction attempts
The container runs as a non-root user with a read-only filesystem
Caddy handles TLS — the app only binds to localhost
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseAqualityFmaintenanceEnables natural language control of AutoCAD LT through AutoLISP code generation and execution, allowing users to create engineering drawings with conversational prompts.Last updated8422MIT
- Alicense-qualityDmaintenanceA browser-based drawing application with AI integration that enables Claude to analyze artwork, provide feedback on composition and techniques, and discuss artistic concepts through MCP tools.Last updated42MIT
- Flicense-qualityFmaintenanceExposes a Pygame-based drawing canvas as an MCP server, allowing LLMs to create digital art using standard shapes and freehand paths. It features a specialized oil paint mode that simulates realistic color mixing, paint depletion, and textured brush strokes.Last updated
Related MCP Connectors
AI agent draws editable hand-drawn diagrams (flowchart, ER, architecture) via MCP, not static images
Generate logos, social posts, app screenshots, comic panels & visual-novel assets from prompts.
Create, validate, edit, export (markdown/svg/png/mermaid), and search JSON Canvas files.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/ymirymir/drawing-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server