Key Capabilities
Any tech stack
Python, JavaScript, Rust, Go, Swift, C++ — whatever your project needs. Workshop Desktop doesn’t limit you to a preset list of languages or frameworks.
Local AI models
Run open-weight models on your own hardware with zero API costs. Workshop Desktop connects to any Anthropic Messages API-compatible server.
MCP servers
Extend your agent with Model Context Protocol servers — connect to databases, APIs, and tools through a standardized interface. Desktop-only feature.
GitHub integration
Import repositories, link existing projects, and push/pull changes directly from the Workshop interface.
How It Works
When you open Workshop Desktop, you start with a blank project — no template, no boilerplate. Describe what you want to build in plain language, and Workshop creates the project structure, writes code, installs dependencies, and runs your application. Workshop Desktop operates in two modes:- Plan Mode — Scope requirements, discuss architecture, and create a structured development plan before writing any code.
- Build Mode — Write code, manage files, run applications, and debug issues in real time.
.memex/ data directory by default, or in any folder you choose). You can open them in any editor, run them from any terminal, and deploy them to any platform.
Privacy and Security
Your code never leaves your machine. Workshop Desktop runs a local backend as a Tauri sidecar process — all file operations, code execution, and project management happen on your hardware. When you use cloud AI models (Claude, GPT, Gemini), your prompts are sent to those providers, but your project files stay local. When you run local AI models, even your prompts stay on your machine. Nothing is transmitted anywhere.What Makes Desktop Different from Cloud
| Workshop Cloud | Workshop Desktop | |
|---|---|---|
| Environment | Browser-based cloud sandbox | Installed on your machine |
| Tech stacks | Streamlit, HTML + Python, None | Any language, any framework |
| File access | Sandboxed VM | Full local file system |
| AI models | Cloud models only | Cloud models + local models |
| MCP servers | Not available | Full MCP support |
| Publishing | One-click publish to shareable URL | Deploy anywhere (Vercel, AWS, your own server) |
| Setup | Zero — open browser and start | Download and install |