pen.dev
pen.dev provides an agentic canvas where designers and developers can import any website or app as editable layers instead of starting from scratch. The platform supports multiple AI models—including Claude, OpenAI, Grok, and Gemini—via API keys, MCP, WebMCP, or CLI, enabling AI-assisted design collaboration directly on the canvas. A custom WebGL pipeline keeps performance smooth even with thousands of layers, and a built-in browser lets users open and import web pages straight into their workspace.
- Creator / Creative Economy
- Artificial Intelligence
- AI Agents
- Developer Tools
- Software Only
Funding
Funding not disclosed
Founders
Founder details are not available yet.
Product
Problem
Designers and developers often start from a blank canvas, losing time recreating existing UI elements or context from other applications. Collaborating with AI models typically requires switching between separate tools, breaking the creative flow and making iterative refinement cumbersome.
Solution
pen.dev provides an agentic canvas where users can import any part of an existing app directly onto the workspace, eliminating the need to start from scratch. The platform supports bringing in multiple AI models—including Claude, OpenAI, Grok, Qwen, Gemini, Kimi, Deepseek, and Minimax—through API keys, MCP, WebMCP, or CLI connections, enabling design collaboration directly on the canvas. Users can refine designs manually or fire up agents to work alongside them. A custom WebGL pipeline ensures smooth performance even with thousands of layers, while a built-in browser allows opening any page inside the app and importing it straight onto the canvas.
Target Audience
Primary users are designers and developers who want to combine visual design work with AI-assisted collaboration, particularly those working in web and product design who need to import existing app elements and iterate with AI models in a single environment.
Features
- Import any part of any app onto the canvas, avoiding blank-canvas starts
- Multi-model AI support via API keys, MCP, WebMCP, or CLI, including Claude, OpenAI, Gemini, and more
- Custom WebGL pipeline for smooth pan, zoom, and editing with thousands of layers
- Built-in browser to open and import web pages directly onto the canvas
- Vibe Shaders feature for creating visual effects with shaders
- Desktop apps for macOS, Windows, and Linux, plus Chrome extension, IDE extension, and CLI tool
- Agent skill bundled with CLI for design instructions in Cursor, Claude Code, and other agents