Browser-based tool for generating OpenGL and DirectX tangent-space normal maps from images
Design canvas where AI agents and humans work together
Stage
Not given
Not given
Founded
Not given
Not given
Based in
Not given
Not given
Pricing model
Free
Freemium
Pricing
Not given
Free: unlimited editors and viewers, 1 workspace with 3 files, 200 agent actions a day. Pro: $20 per editor per month (billed monthly) or $16 per editor per month (billed yearly), includes unlimited files and workspaces, unlimited agent actions, version history. Team: $35 per editor per month (bille
Key features
Local image processing
OpenGL and DirectX output
Sobel and Scharr filters
Material lighting preview
PNG export
Smoothing and strength controls
Edge behavior control (clamp or wrap)
Tiled preview
Height inversion
Interactive light inspection
Real HTML and CSS designs that agents can edit directly
Auto layout, components with variants, and variables with modes
Figma import and export capability
Design system integration with skills and guardrails
Comment-based task assignment for agents
MCP server with 59 tools for reading, drawing and editing
Real-time live collaboration with team and agents
Version history and restore functionality
Copy frames as HTML, React or Tailwind code
Mac app with local MCP server for offline agent work
What makes it different
Runs entirely in the browser with no server upload or account required
Supports both OpenGL and DirectX normal map formats
Choice of Sobel or Scharr gradient filters for different texture quality
Free with no advertising, AI model, or payment system
Interactive material lighting preview to verify results
Agents work in real HTML and CSS, not converted designs
No separate AI credits to pay for agent usage
Design system enforcement so agents follow brand guidelines automatically
Works with any MCP client including Claude, ChatGPT, Cursor and Codex