Haruhiyuki

Haruhiyuki / DesignCode

Public

Agent-powered graphic design workbench that uses HTML/CSS/SVG as the design medium, supporting vector-quality output, editable elements, multi-layer PSD export, lossless text rendering, and other production-grade features.

agent ai ai-agent ai-design code-generation
12
0
100% credibility
Found Mar 27, 2026 at 12 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

DesignCode is an agent-based graphic design workbench that transforms simple instructions into high-quality design drafts, supporting features like vector-level output, multi-layer PSD export, asset management, style presets, and cross-platform compatibility on macOS and Windows.

Star Growth

See how this repo grew from 12 to 12 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is DesignCode?

DesignCode is an agent-powered graphic design workbench built in JavaScript that lets you chat with AI to generate high-quality drafts like posters, book covers, or product pages using HTML/CSS/SVG as the editable medium. It solves the pain of blurry AI images by delivering vector-quality output with perfect text rendering, direct canvas editing, and exports to PNG, SVG, PDF, or multi-layer PSD. Desktop app via Tauri supports macOS/Windows, with asset libraries and style presets for quick starts on tasks like design code IKEA layouts or design codes ACNH patterns.

Why is it gaining traction?

It stands out from diffusion-based generators by producing fully editable elements via code, enabling multi-turn refinements without regeneration—ideal for production where text and layers must stay crisp. Features like user asset integration, full history tracking, and LLM backends (Codex, Claude, OpenCode, Gemini) make iterating on graphic designs feel like collaborating with a pro. The hook: vector exports and PSD support beat screenshot workflows for real use cases like designcode architects or design codes Monster Hunter Wilds promos.

Who should use this?

Graphic designers prototyping client mocks, indie game devs needing quick assets like design codes Animal Crossing New Horizons or Ritter Sport packaging, and frontend teams generating editable UI visuals. Non-designers handling one-offs (postcards, promo graphics) will appreciate the chat-to-export flow without Photoshop installs.

Verdict

Worth a spin for agent-powered, editable graphic workflows—early releases and multi-lang docs make setup straightforward despite 12 stars and 1.0% credibility signaling it's fresh. Maturity lags (light tests, solo dev vibe), but Apache 2.0 and Tauri packaging lower risks for experimentation.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.