Developer 0xnyn has released Airship, a visual editor that brings Figma-like prototyping capabilities directly to AI-assisted coding workflows. The tool integrates with Claude Code, Codex, and OpenCode, allowing developers to experiment with UI changes in real-time without leaving their development environment or manually translating design decisions into code.

How It Works

Airship operates by letting you prototype directly within your running application rather than bouncing between a separate design tool and your IDE. You can select individual elements on screen, manipulate them visually—adjusting spacing, colors, typography—and explore multiple design directions without writing a single line of code. Once you've settled on a direction you like, Airship applies those visual changes back into your codebase automatically.

The Design-Development Gap

This project addresses a persistent friction point in modern development workflows: the disconnect between designers working in tools like Figma and developers implementing those designs in code. When using AI coding assistants that generate UI based on prompts or specifications, iterating on design decisions typically requires either prompt refinement or manual code editing. Airship cuts out that middle step by providing direct visual manipulation of rendered output.

Supported Platforms

The tool currently supports three major AI coding platforms: Anthropic's Claude Code, OpenAI's Codex (and its derivatives), and OpenCode. This broad compatibility suggests the developer is targeting the growing segment of developers who rely on AI assistants for UI implementation work, regardless of which model or service they prefer.

Key Takeaways

  • Airship enables visual prototyping directly in running applications rather than static design files
  • Changes made visually are automatically translated back into source code through the connected AI assistant
  • Supports Claude Code, Codex, and OpenCode out of the box
  • Open-source project hosted on GitHub for community contribution

The Bottom Line

This is exactly the kind of tooling that bridges the gap between rapid prototyping and actual implementation—something the AI coding space has been missing. Whether Airship gains traction depends heavily on how robust its code-generation capabilities are when translating visual changes back into clean, maintainable source code.