RepoMuse

AI‑assisted repository explorer and idea generator. RepoMuse is a cross‑platform desktop app built with Tauri (Rust) and React + TypeScript that helps you:

• Point at a root folder of code projects
• Discover projects automatically (Node, Rust, Python, Go, etc.)
• Analyze a selected project (files, lines, languages, lightweight structure)
• Generate a clear project summary and actionable development ideas using an OpenAI‑compatible API (OpenAI, Azure OpenAI, DeepSeek, Ollama)
• Persist settings and summaries locally for quick revisits

What It Does

• Project discovery: Scans the chosen root folder for likely projects using common indicators (e.g., package.json, Cargo.toml, requirements.txt, .csproj, Makefile). Shows name, Git presence and a fast file count that refines in the background.
• Repository analysis: Reads source files (skipping big/binary/ignored folders), detects languages by extension, and computes metrics like total files and lines plus a lightweight directory structure.
• AI summaries and ideas: Calls your configured OpenAI‑compatible endpoint to produce a human‑readable project summary and 5–10 practical, numbered development ideas. Thinking‑style models (e.g., o1-mini, deepseek-r1) are supported and hidden reasoning tags are filtered.
• Local persistence: Settings, selected root folder, file‑count cache and generated summaries are stored locally via Tauri for a snappy, privacy‑friendly experience.

Quick Start

Prerequisites

• Node.js 18+ and npm
• Rust toolchain (stable) and platform build tools required by Tauri
  • macOS: Xcode Command Line Tools
  • Windows: Visual Studio Build Tools (C++), WebView2
  • Linux: common GTK/WebKit and build packages

Install

• npm install

Run (desktop, recommended)

• npm run tauri dev — launches the Tauri app with hot reload.

Run (web‑only)

• npm run dev — Vite dev server in the browser. Some Tauri‑only features (e.g., native dialogs, local persistence paths) may be limited.

Build

• Frontend: npm run build
• Desktop app: npm run tauri build

Using RepoMuse

1. Choose a root folder

• On first launch, click “Choose Folder” and select a directory that contains many project subfolders.

2. Browse projects

• The left sidebar lists detected projects. It shows a quick file count (refined asynchronously) and a Git badge when applicable.

3. Analyze a project

• Select a project to view metrics (files, lines), detected technologies, and a lightweight directory view.

4. Configure AI

• Open Settings to set API Server URL, optional API Key, and Model.
• Click “Load Models” to fetch available models from compatible APIs.
• Examples:
  • Ollama (local): URL http://localhost:11434/v1/chat/completions, empty key, models like llama2, codellama, deepseek-r1.
  • OpenAI: URL https://api.openai.com/v1/chat/completions, API key required, models like gpt-4, o1-mini.
  • DeepSeek: URL https://api.deepseek.com/v1/chat/completions, API key required, models like deepseek-chat, deepseek-r1.

5. Generate output

• In the Project Summary card, click “Generate Summary” to create a concise explanation of what the project is and does.
• In the Development Ideas card, click “Generate Ideas” to get actionable suggestions. Regenerate anytime.
• Summaries are saved locally and loaded automatically next time.

Tech Stack

• Frontend: React 19, TypeScript, Vite, Tailwind CSS
• Desktop: Tauri 2 (Rust)
• Native plugins: @tauri-apps/plugin-dialog, @tauri-apps/plugin-opener
• Icons: lucide-react
• AI backend: Any OpenAI‑compatible Chat Completions API

Scripts

• npm run dev: Start Vite dev server (web‑only)
• npm run build: Type‑check and build frontend
• npm run preview: Serve built assets from dist/
• npm run tauri dev: Run desktop app with hot reload
• npm run tauri build: Build production desktop binaries

Project Structure

• src/: React + TypeScript UI (components, utils, types, assets)
  • Example: src/components/ProjectAnalyzer.tsx, src/components/Settings.tsx
• src-tauri/: Tauri (Rust) backend
  • Entry: src-tauri/src/main.rs
  • Config: src-tauri/tauri.conf.json
• public/ and index.html: Static assets and Vite entry
• Tooling: vite.config.ts, tailwind.config.js, postcss.config.js, tsconfig*.json

Privacy & Security

• No secrets are hardcoded. Enter your API URL, model and key in Settings; they are stored locally by Tauri.
• Repo content is analyzed locally; only the prompts you request (for summaries/ideas) are sent to your configured API endpoint.
• Large/binary folders like node_modules, target, and dist are skipped during analysis.

Testing

• JS/TS: Not set up yet (recommended: Vitest + React Testing Library). Name tests *.test.ts(x) next to sources.
• Rust: Add unit tests with #[cfg(test)] inside modules and run with cargo test in src-tauri/.

Troubleshooting

• Models not loading: Verify the API URL points to a models‑listing endpoint compatible with OpenAI or Ollama variants, and that your API key (if required) is valid.
• Ideas/Summary not generating: Ensure Settings are saved and the model supports chat completions. Check your network and API logs.
• Desktop build issues: Confirm Tauri prerequisites for your OS are installed and Rust toolchain is up to date.

---

Made with Tauri + React to make sense of codebases and spark next steps.