SharimAli

Imgui Menu Clone — AI skill for Claude Code

AI community

Rebuild any Dear ImGui menu in clean C++ from a single screenshot.

How to install Imgui Menu Clone

This entry records only its repository, not the path inside it, so there is no exact command to give. Open SharimAli/imgui-menu-clone and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Imgui Menu Clone does

Rebuild any Dear ImGui menu in clean C++ from a single screenshot. Generates a buildable native Visual Studio (.sln) Win32/D3D11 project without original source code. Multi-platform AI skill for Claude, Antigravity, and Cursor.

Alternatives in AI

Matlab MCP Server — Run MATLAB® using AI applications with the official MATLAB MCP Server from MathWorks® 1.4k ★
Claude Skill Aso Appstore Screenshots — A Claude Code skill that generates high-converting App Store screenshots for your iOS app 1.2k ★
Claude Skill Web Clone — Methodology for faithfully cloning any website (static / React / WebGL) — without copying AI-hallucinated code 969 ★

README

ImGui Menu Clone 🎨

**Rebuild any Dear ImGui menu in clean C++ from a single screenshot.** Give your AI assistant a screenshot of an ImGui menu. Get back a fully buildable, pixel-accurate native Visual Studio (`.sln`) C++ project. No access to the original source code needed.

[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE) [![Platform](https://img.shields.io/badge/Platform-Windows-lightgrey.svg)](https://microsoft.com/windows) [![Visual Studio](https://img.shields.io/badge/Visual%20Studio-2022%20%2F%202026-purple.svg)](https://visualstudio.microsoft.com/) [![ImGui](https://img.shields.io/badge/Dear%20ImGui-v1.91.9b-orange.svg)](https://github.com/ocornut/imgui)

**Author:** [Sharim (GHX4T)](https://github.com/SharimAli) **License:** [MIT License](LICENSE)

⚡ Why This Skill Exists

Asking an AI to "rebuild this menu" from a picture usually yields an approximate "70% right" result: recognizable layout, but wrong paddings, inaccurate corner radii, guessed RGB values, and incorrect widget geometry.

This skill eliminates guessing through a **disciplined 5-phase pipeline**:

[ Screenshot ] ──▶ Phase 0: Bootstrap (.sln + Dear ImGui)
               ──▶ Phase 1: Analyse (Exact RGB palette + pixel measurements)
               ──▶ Phase 2: Checkpoint (User reviews & confirms spec.md)
               ──▶ Phase 3: Generate (Theme table + layout + custom widgets)
               ──▶ Phase 4: Compare (Build .exe ──▶ Screenshot ──▶ 4x Zoom Diff)

Phase	What Happens
0 · Project	Bootstraps a clean Win32 + DirectX 11 + Dear ImGui native Visual Studio solution (`menu.sln`).
1 · Analyse	Python script extracts exact bit-for-bit RGB values and pixel bounds into `spec.md`.
2 · Checkpoint	AI pauses to verify `spec.md` with you before writing a line of C++.
3 · Generate	Writes theme variables, layout structure, and any custom widgets.
4 · Compare	Compiles `menu.exe`