DrawArch transforms a reference architecture image into a native editable Draw.io diagram

DrawArch MCP

Turn architecture prompts and reference images into portable, natively editable Draw.io files.
Your MCP host does the visual reasoning. DrawArch does the deterministic rendering, asset embedding, validation, and delivery.

Quick start · See the workflow · Architecture · Contribute

See what stays editable

A DrawArch-generated SafeSense system architecture open as independently editable Draw.io objects with layers and validation

This is not a screenshot pasted into a diagram. Containers, labels, icons, shapes, connector endpoints, ports, and waypoints remain separate native mxCell objects. Assets are embedded inside the .drawio file, so the result remains portable and editable offline.

[!IMPORTANT] DrawArch does not call another LLM and does not require a second LLM API key. ChatGPT, Claude, Codex, Cursor, or another vision-capable MCP host analyzes the request or image and supplies the scene graph.

What DrawArch gives you

	Capability	What it means
🧩	Native editability	Move, restyle, relabel, reconnect, or delete individual objects in diagrams.net.
🎯	Reference recreation	Rebuild an attached architecture image with absolute geometry, layers, z-order, ports, and waypoints.
🌐	Real-world assets	Use bundled icons, Iconify-compatible icons, user images, or approved HTTPS image sources.
📦	Portable artifacts	The generated `.drawio` is returned directly through MCP with all image assets embedded.
🔐	Approval-aware rendering	Immutable plan revisions and revision-bound approval tokens prevent stale or unreviewed renders.
✅	Validation built in	Check Draw.io structure, references, asset safety, and structural coverage before accepting output.

Reference-image recreation

Attach an architecture image to your MCP host and ask:

Recreate this image with DrawArch. Preserve its layout, labels, icons, grouping, and connector routing. Keep every visible object editable and use real icons where appropriate.

The host and DrawArch then follow a reviewable four-step loop:

Prepare — the host vision model calls prepare_reference_recreation with a typed absolute scene graph.
Review and approve — the user or host policy reviews layout and online asset sources; approval is bound to that exact revision.
Create — create_reference_drawio safely resolves and embeds assets, renders native cells, validates the file, and returns the .drawio resource.
Compare and revise — compare_reference_recreation measures structural coverage; the host adjusts the plan and repeats when needed.

mermaid

flowchart LR
    A[Prompt or<br/>reference image] --> B[Vision-capable<br/>MCP host]
    B -->|typed scene graph| C{DrawArch MCP}
    C --> D[Plan revision<br/>& approval]
    D --> E[Safe asset<br/>resolver]
    E --> F[Native Draw.io<br/>renderer]
    F --> G[Validation &<br/>comparison]
    G -->|portable resource| H[Editable<br/>.drawio file]
    G -. revise .-> B

In plain terms: your existing AI analyzes the image; DrawArch turns that analysis into a controlled, validated, editable file. The complete contract is demonstrated in examples/reference-recreation.json.

Why not just generate an image?

Requirement	DrawArch	Flat image generation
Edit individual components	✅	❌
Reattach and reroute connectors	✅	❌
Preserve explicit IDs and hierarchy	✅	❌
Work offline after generation	✅	Usually
Validate structure automatically	✅	❌
Recreate with real embedded icons	✅	Inconsistent
Iterate through a reviewable plan	✅	❌
Guarantee pixel-identical output	No — host-guided iteration	No

DrawArch optimizes for high-fidelity, editable reconstruction, not an opaque pixel replica. The host vision model can inspect the result, update geometry or styling, and render another immutable revision.

Quick start

Requires Node.js 20+ or Docker.

Run from npm

Terminal

npx -y drawarch-mcp@0.2.0

DrawArch uses stdio by default. Diagnostics go to stderr, preserving stdout for MCP messages.

Add it to a local MCP host

config.json

{
  "mcpServers": {
    "drawarch": {
      "command": "npx",
      "args": ["-y", "drawarch-mcp@0.2.0"],
      "env": {
        "DRAWARCH_OUTPUT_DIR": "/absolute/path/to/diagrams",
        "DRAWARCH_ONLINE_ASSETS": "false"
      }
    }
  }
}

On Windows hosts that do not resolve npx directly, set command to npx.cmd. You can also use the source checkout and configure an absolute dist/src/index.js path.

Run from source

bash

git clone https://github.com/Ajey95/drawarch-mcp.git
cd drawarch-mcp
npm ci
npm test
npm run build

Then configure your host:

config.json

{
  "mcpServers": {
    "drawarch": {
      "command": "node",
      "args": ["/absolute/path/to/drawarch-mcp/dist/src/index.js"],
      "env": {
        "DRAWARCH_OUTPUT_DIR": "/absolute/path/to/diagrams"
      }
    }
  }
}

Run the published container

Terminal

docker run --rm -p 3000:3000 \
  -e DRAWARCH_API_KEY=replace-with-a-long-random-secret \
  -e DRAWARCH_APPROVAL_SECRET=replace-with-another-random-secret \
  -e DRAWARCH_ALLOWED_HOSTS=localhost,127.0.0.1 \
  -e DRAWARCH_ALLOWED_ORIGINS=localhost,127.0.0.1 \
  -v drawarch-data:/data \
  ghcr.io/ajey95/drawarch-mcp:0.2.0

The container runs Streamable HTTP on http://localhost:3000/mcp. Its public health probe is GET /health; MCP and file downloads require the configured bearer token.

Local and remote hosts

Host type	Recommended transport	Notes
Claude Desktop / Claude Code	stdio or remote HTTP	Local stdio is simplest; remote connectors can use `/mcp`.
Codex / Cursor / VS Code	stdio or remote HTTP	Use local stdio for per-developer output or a shared authenticated service.
ChatGPT	remote Streamable HTTP	ChatGPT cannot spawn the local stdio process; deploy the container behind HTTPS first.
CI and team services	Streamable HTTP	Set bearer auth, stable approval secret, exact host/origin allowlists, TLS, and a persistent output volume.

DrawArch currently provides the deployable server and public container, not a managed hosted endpoint. Operators supply the HTTPS domain and infrastructure.

MCP tools

Architecture generation

Tool	Purpose
`list_themes`	List six built-in visual themes.
`list_assets`	Search the bundled real-world icon catalog.
`resolve_asset`	Resolve a bundled or opt-in online SVG with provenance.
`create_drawio`	Generate a deterministic themed architecture diagram.
`validate_drawio`	Validate a generated architecture file.

Reference recreation

Tool	Purpose
`prepare_reference_recreation`	Store an absolute editable scene graph and optionally approve its sources.
`update_reference_plan`	Create an immutable next revision and invalidate earlier approvals.
`get_reference_plan`	Read the latest scene-graph revision for review.
`create_reference_drawio`	Render an approved plan and return the portable `.drawio` artifact.
`validate_reference_drawio`	Validate native structure, references, and embedded-image safety.
`compare_reference_recreation`	Score structural coverage and report missing editable elements.

Asset and security model

Reference plans support four asset modes:

Mode	Source	Server opt-in required?
`bundled`	Original MIT-licensed DrawArch SVG catalog	No
`online`	Iconify-compatible icon query	Yes
`data`	User-provided SVG, PNG, JPEG, or WebP data URI	No
`remote`	Explicit approved HTTPS image URL	Yes

Remote image handling rejects URL credentials, non-HTTPS protocols, custom ports, private or reserved DNS answers, unsafe redirects, excessive byte/pixel sizes, and active SVG content. Raster images are decoded and re-encoded before embedding. Online assets retain provider and license metadata; users remain responsible for verifying third-party license terms.

HTTP deployments additionally support bearer authentication, host validation, origin validation, safe output paths, atomic writes, and revision-bound HMAC approval tokens. See SECURITY.md for deployment guidance and vulnerability reporting.

Environment variables

Variable	Default	Meaning
`DRAWARCH_TRANSPORT`	`stdio`	Select `stdio` or `http`.
`DRAWARCH_OUTPUT_DIR`	`.drawarch-output`	Restricted output and plan-storage directory.
`DRAWARCH_ONLINE_ASSETS`	`false`	Enable opt-in online and remote image sources.
`DRAWARCH_ICONIFY_BASE_URL`	Iconify API	HTTPS Iconify-compatible provider base URL.
`HOST`	`127.0.0.1`	HTTP bind host.
`PORT`	`3000`	HTTP listener port.
`DRAWARCH_API_KEY`	unset	Bearer token for `/mcp` and `/files`; required for internet exposure.
`DRAWARCH_APPROVAL_SECRET`	random per process	HMAC secret for approvals; set a stable secret for shared deployments.
`DRAWARCH_ALLOWED_HOSTS`	local hosts	Comma-separated accepted Host values.
`DRAWARCH_ALLOWED_ORIGINS`	local origins	Comma-separated accepted Origin hosts.
`DRAWARCH_PUBLIC_BASE_URL`	listener URL	Public base URL reported by the server.

Development

Terminal

npm ci
npm test                    # 83 tests at the 0.2.0 release
npm run build
npm run sample
npm audit --audit-level=high
npm pack --dry-run

The CI matrix verifies Node 20 and 22 on Windows and Linux and builds the production Docker image. Read docs/ARCHITECTURE.md for component boundaries and security flow.

Project status

Channel	Status
Source	Public on GitHub
Release	`v0.2.0`
Container	`ghcr.io/ajey95/drawarch-mcp:0.2.0`
npm	`drawarch-mcp@0.2.0`
Official MCP Registry	`io.github.Ajey95/drawarch@0.2.0`
Managed remote service	Not currently offered

Roadmap

Expand the curated architecture asset catalog and provider metadata.
Add preview rendering for tighter host-guided visual comparison loops.
Add pixel-aware comparison as a complement to structural scoring.
Add installation recipes for more MCP hosts and community catalogs.
Provide repeatable deployment recipes for common container platforms.

Ideas and focused pull requests are welcome. Please open an issue before starting a large behavior or contract change.

Contributing

Read CONTRIBUTING.md, follow the CODE_OF_CONDUCT.md, and keep generated outputs, credentials, and unlicensed third-party assets out of commits.

For security concerns, follow the private reporting process in SECURITY.md rather than opening a public issue.

License

DrawArch source code and bundled assets are available under the MIT License. Third-party assets keep their original provider licenses and attribution.

Built for architecture work that should remain understandable, portable, and yours to edit.