⭐ Inference Gateway
Studio runs on Inference Gateway
Studio makes its model calls through Inference Gateway on your organization's account, and token fees apply under your AI Addendum. Your developers can use the same gateway for their own builds.
Available to all Canvas customers → read more
How Studio Works
1
Design
Describe the workflow you want in your own words. Studio creates a plain-English plan and technical architecture for you to review.
2
Deploy
Studio writes the code, checks the existing plugin library for proven references, and deploys to a sandbox on your Canvas instance.
3
Iterate
Test the plugin live. Tell Studio what to change. Publish when it does what you need.
Who Studio Is For
Anyone with subject matter expertise and a willingness to build can use Studio.
Clinical Leaders
You bring the domain expertise, Studio writes the code. The agent encodes your knowledge directly into Canvas. Create your own chart audits, screening triggers, documentation templates, dashboards your team will actually use.
Operations & Practice Managers
Studio automates provider enrollment, scheduling rules, intake flows and billing. It turns those manual processes into running code.
Developers & CTOs
Your team builds the common 80% without filing a ticket - and engineering steps in where production builds need a human review or the last mile of integration work. Use the Canvas Plugin Assistant from the command line with Claude Code: same Canvas SDK, same governance, same code. Developers building outside Studio can call Claude models through Inference Gateway with a Canvas key.
Customers building with Studio
FAQ
What is Canvas Studio?
Canvas Studio is an AI plugin builder that runs in Canvas Platform. Describe what you want in plain English, and Studio writes, tests and deploys the plugin for you. Canvas built Studio on the Canvas SDK and Claude Code.
Who is Studio for?
Anyone with a workflow problem can use Studio, including clinical leaders, operations teams, practice managers and billing staff. You don't need a coding background. Developers can use the same engine through CPA from the command line.
When can I use it?
Studio is available to all Canvas customers. Ask your Canvas team to set up access for your organization.
Do I need to install anything?
No. Studio runs entirely in the browser in Canvas Platform. No Python, no command line, no IDE.
What can I build with Studio?
Anything the Canvas SDK supports - charting apps, dashboards, scheduling workflows, billing automations, patient portal experiences, intake processes, custom data models, agent-driven workflows. If a Canvas plugin can do it, Studio can build a working version - and tell you where a complex build needs review before go-live.
How is Studio's AI usage billed?
Studio makes its model calls through Inference Gateway on your organization's account. Token fees apply under your AI Addendum and appear on your monthly Canvas invoice. Billing admins can see Studio's spend on the Usage & Billing page in Canvas Platform.
Is the code I generate mine?
Yes. Each Studio project becomes a GitHub repo that your team has access to. You can take it with you, share it and modify it outside Studio.
What's the difference between Studio and CPA?
CPA is the developer tool, a Claude Code plugin you run from the command line. Studio is the no-code version that runs on top of CPA, so both use the same engine.
How much control do I have over what Studio builds?
You stay in control at every step. Studio moves through Design, Build, Test, and Review and never ships on its own - at each stage it shows a plain-English plan, flags any tradeoffs (security, scale, manual steps, integrations), and waits for your approval. For production-critical builds, your team or a Canvas engineer can review the code and finish the last mile.
