A self-improving AI agent that creates and refines skills from experience, runs on many model providers, and works through CLI and messaging apps.
- Stars
- 222
- Forks
- 58
- Open issues
- 14
- Last commit
- 8 days ago
What is it?
Design Studio AI is a workspace for building web interfaces, slides, reports, wireframes, 3D scenes and timeline videos. You start from a brief or template, review the preview, then make changes through chat or a manual editor. Humans and agents work on the same versioned document.
Generation features use your own provider keys (BYOK) for text, image, speech, music and video. Designs export to formats including HTML, SVG, PDF, PowerPoint, WebM, GLB and React prototype ZIP. Agents can reach the same designs through a REST API, a Streamable HTTP MCP server and the dsa CLI.
The studio can be hosted on Cloudflare or self-hosted with Docker and persistent SQLite storage. Sign-in supports email and password, with optional GitHub login.
Key features
- 01Chat-driven and manual editing of the same versioned design document
- 022D character motion with bones, skins, meshes and physics, plus editable 3D characters
- 03Export to HTML, SVG, PDF, PowerPoint, WebM, GLB and React prototype ZIP
- 04REST API, MCP server with OAuth or API keys, and the dsa CLI
- 05Bring-your-own-key generation for text, image, speech, music and video
- 06Cloudflare or Docker self-hosting with persistent SQLite storage
Use cases
- Drafting landing pages with an AI agent and refining them by hand
- Building presentation decks that export to PowerPoint or Google Slides
- Creating animated 3D scenes or 2D character motion clips
- Running a self-hosted design workspace for a team with its own provider keys
Quick start
- 1
Install dependencies
npm ci - 2
Build the CLI and app
npm ci --prefix packages/cli - 3
Start the server (set ENCRYPTION_KEY and APP_URL first)
npm start - 4
Open the studio in your browser at localhost:8787
Things to know
- Generation requires your own provider credentials and account access.
- Node.js 24 or newer is required to run locally.
- The CLI package is not published to the npm registry and must be installed from a release tarball or local build.
- Changing the ENCRYPTION_KEY prevents decryption of existing provider keys.
Topics
FAQ
- What is design-studio-ai?
- Open-source design workspace where people and AI agents edit one versioned document for web pages, slides and 3D scenes. Design Studio AI is a workspace for building web interfaces, slides, reports, wireframes, 3D scenes and timeline videos. You start from a brief or template, review the preview, then make changes through chat or a manual editor. Humans and agents work on the same versioned document.
- Is design-studio-ai free to use? What license does it have?
- design-studio-ai is open source under the MIT license. Check the license terms before using it in commercial products.
- Is design-studio-ai still maintained?
- Actively maintained. The last commit to design-studio-ai was on Oct 12, 2026.
- What can I use design-studio-ai for?
- Drafting landing pages with an AI agent and refining them by hand Building presentation decks that export to PowerPoint or Google Slides Creating animated 3D scenes or 2D character motion clips Running a self-hosted design workspace for a team with its own provider keys
- How do I install design-studio-ai?
- Install dependencies: npm ci Build the CLI and app: npm ci --prefix packages/cli Start the server (set ENCRYPTION_KEY and APP_URL first): npm start Open the studio in your browser at localhost:8787
- Where can I get design-studio-ai?
- The source code and installation instructions are on GitHub: https://github.com/bestagentkits/design-studio-ai
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Self-hosted auth gateway that lets AI agents call 1,000+ SaaS providers through SDK, CLI, MCP, HTTP, and OpenAPI without handing over credentials.
An open-source AI agent from Google that brings Gemini into your terminal to query, edit, and automate code.
An open source, general-purpose AI agent that runs on your machine as a desktop app, CLI, or embeddable API, for code and non-code tasks.
Self-hosted personal AI agent framework in Python with a browser WebUI, terminal client, chat app channels, tools, memory and MCP support.
Open-source personal AI assistant and Agent Harness that plans tasks, runs tools and skills, and keeps memory across web and IM channels.