Diagrams from a prompt

AI architecture diagram generator

Describe your system to your AI agent and it draws an editable architecture diagram on your Archboard canvas: Kubernetes clusters, AWS stacks, C4 containers, microservices and event flows.

Archboard does not run its own AI model. Your AI app connects to Archboard over MCP and draws on your canvas; you edit, share and export the result like any other diagram.

MCP server URLhttps://getarchboard.app/mcp

Quick setup for your app

Opens in a new tab

Don’t see your app? Copy the URL above into its connector or MCP settings. It needs OAuth sign-in without a client secret.

Archboard isn’t affiliated with these apps. Names and logos belong to their owners.

Copy a prompt, get a diagram

After you connect Archboard, paste one of these prompts into your AI app. Each one comes with the kind of diagram it draws: a template built from the same shapes, which you can also open directly.

  • Kubernetes cluster

    Prompt

    Draw a Kubernetes cluster architecture diagram in Archboard: a control plane with kube-apiserver, etcd, kube-scheduler and kube-controller-manager, two worker nodes that each run kubelet, kube-proxy and a pod, and an Ingress that routes HTTPS to the web pod.

    What your agent draws

  • AWS three-tier web app

    Prompt

    Create an AWS architecture diagram in Archboard for a three-tier web app: Route 53, CloudFront with AWS WAF, static files in S3, and a VPC with a public subnet for the Application Load Balancer, a private subnet with an EC2 Auto Scaling group and a private data subnet with Multi-AZ RDS and ElastiCache.

    What your agent draws

  • AWS serverless backend

    Prompt

    Draw a serverless AWS architecture in Archboard: CloudFront and S3 for the front end, Cognito sign-in, API Gateway calling an orders Lambda function that writes to DynamoDB, and an OrderPlaced event from EventBridge to an SQS queue that triggers a worker function sending email with SES.

    What your agent draws

  • C4 container diagram

    Prompt

    Draw a C4 container diagram in Archboard for an online store: a customer, a web app and a mobile app, an API application, an orders database, a message broker and an order worker inside the system boundary, plus an external payment provider and an email service. Label each relationship with what it does and the protocol.

    What your agent draws

  • Microservices with an API gateway

    Prompt

    Draw a microservices architecture in Archboard: web and mobile clients, an API gateway that checks tokens with an identity provider, users, orders, payments and inventory services with their own databases, and an event bus where orders-service publishes OrderPlaced for the payments and inventory services.

    What your agent draws

  • Event-driven flow

    Prompt

    Draw an event-driven architecture in Archboard: checkout-service publishes OrderPlaced to an orders topic, payment-service and inventory-service subscribe, payment-service publishes PaymentCaptured to a payments topic for shipping and notifications, with a dead-letter queue and an analytics warehouse.

    What your agent draws

See an agent draw on a real canvas

A scripted demo on the Archboard canvas, with sample names. The agent’s cursor shows who it works for, new shapes appear one by one, and a notice lets you undo your agent’s change in one step.

Demo
  1. An AI agent working for Maya adds a database, its storage, and a cache next to the checkout API. Its cursor shows who it works for, and the new shapes appear one by one.
  2. At the same time, Leo changes the checkout API from 3 to 5 replicas.
  3. Maya clicks Undo in the notice: the agent’s change is taken back in one step, and Leo’s edit stays.
Demo with sample names: an AI agent working for Maya adds a database, its storage, and a cache through MCP while Leo edits the same diagram. Maya can undo the agent’s change in one step; Leo’s edit stays.

Connect in three steps

Add Archboard to your AI app as a remote MCP server, sign in with Google or GitHub and choose what it may use. Then paste a prompt.

  1. Add the server to your AI app

    In your AI app’s connector or MCP settings, add Archboard as a remote MCP server with this URL. Some apps call it a custom connector.

    MCP server URLhttps://getarchboard.app/mcp
  2. Sign in and choose access

    Your app opens Archboard. Sign in with Google or GitHub, pick the organizations and workspaces it can use, choose read, edit, or delete for each, and give it a name.

  3. Ask for a diagram

    Ask your agent to draw or update a diagram, then open it in Archboard to watch the changes arrive live.

    Try asking

    • Draw our checkout service on Kubernetes with Postgres and Redis.
    • Add a cache next to checkout-api in my Payments diagram.
    • Undo your last change.

Your AI app needs to support remote MCP servers with OAuth sign-in.

Why draw with your agent in Archboard

  • Editable, not a picture

    The agent adds real nodes and connections to a diagram, not an image. Rename, move, restyle and extend everything it drew, by hand or by asking again.

  • Shapes that mean something

    Kubernetes resources, cloud services, C4 elements and data stores come from the same node packs you use, with fields such as replicas, CIDR ranges and technology.

  • Live on your team’s canvas

    Edits arrive as they happen, the agent’s cursor shows who it works for, and you can undo your agent’s change in one step.

  • Ready to share

    Share a link, embed the diagram or export SVG, PNG or PDF. Every change is kept in version history.

Questions about generating diagrams with AI

Does Archboard run its own AI model?

No. Archboard has no model of its own. Your prompt goes to your AI app, and the app calls Archboard’s MCP server to create and edit the diagram, within the access you approved and your plan limits.

Which AI apps can I use?

Any AI app that supports remote MCP servers with OAuth sign-in. Shortcuts for popular apps are at the top of this page, and the connection steps work for any other app: add the server URL in the app’s connector or MCP settings.

Is it free?

Yes. Connecting an AI agent works on every plan, including Free, with the same plan limits as the editor. Agent tool calls also have a per-minute rate limit.

Can I change what the agent drew?

Yes. It is a normal Archboard diagram: edit it by hand, ask the agent for changes, or undo your agent’s change in one step. People who have the diagram open see each change live.

Do I need an account?

Yes. When your AI app connects, Archboard asks you to sign in with Google or GitHub and to choose which organizations and workspaces the app can use, and whether it may read, edit or delete.

Open a canvas in a minute

Sign in with Google or GitHub. Free for a single seat. Upgrade when the team joins.