victorGPT / vibeusage

🟢 VIBEUSAGE **QUANTIFY YOUR AI OUTPUT** _Track AI Token Usage Across All Your CLI Tools_ **www.vibeusage.cc** **English** • **中文说明** **Documentation** • **Dashboard** • **Backend API** --- 🚀 Quick Start Get started in 30 seconds: That's it! Your AI token usage will now automatically sync to the Dashboard 🎉 ✨ Why VibeUsage? • 📡 **Multi-Source Unified Tracking** - Support for Codex CLI, Every Code, Gemini CLI, Claude Code, Opencode, OpenClaw, and more • 🤖 **Multi-Model Statistics** - Unified tracking for GPT-4, Claude, Gemini, o1, and all AI models • 📁 **Project AI Footprint** - Track and publicly display token usage per repository, proving AI-assisted development • 🏆 **Global Leaderboard** - Weekly/monthly/all-time rankings, grow with the global developer community • 🌐 **Public Profiles** - Share your AI usage journey, optionally participate in leaderboards • 🔒 **Privacy-First** - Only track numbers, never upload your code or conversations • ⚡ **Zero-Config Auto-Sync** - Set up once, sync forever • 🎨 **Matrix-A Design** - Cyberpunk-style high-fidelity dashboard • 📈 **Deep Analytics** - Cost insights, trend forecasting, activity heatmaps 🧰 Supported AI CLIs | CLI Tool | Auto-Detection | Status | |----------|----------------|--------| | **Codex CLI** | ✅ | Full Support | | **Every Code** | ✅ | Full Support | | **Gemini CLI** | ✅ | Full Support | | **Claude Code** | ✅ | Full Support | | **Opencode** | ✅ | Full Support | | **OpenClaw** | ✅ | Full Support | Whether you're using GPT-4, Claude 3.5 Sonnet, o1, or Gemini - all token consumption is unified and tracked. 🌌 Overview **VibeUsage** is an intelligent token usage tracking system designed for macOS developers. Through the all-new **Matrix-A Design System**, it provides a high-fidelity cyberpunk-style dashboard that transforms your **AI Output** into quantifiable metrics, supported by the **Neural Divergence Map** for real-time monitoring of multi-model compute distribution. > [!TIP] > **Core Index**: Our signature metric that reflects your flow state by analyzing token consumption rates and patterns. 📊 Dashboard Features 🎨 Matrix-A Design System High-performance dashboard built with React + Vite, featuring our cyberpunk-inspired design language with: • **Neural Divergence Map**: Visualize multi-engine load balancing and compute distribution • **Cost Intelligence**: Real-time, multi-dimensional cost breakdown and forecasting • **Activity Heatmap**: GitHub-style contribution graph with streak tracking • **Smart Notifications**: Non-intrusive system-level alerts with Golden visual style 📈 Analytics & Insights • **AI Analytics**: Deep analysis of Input/Output tokens with dedicated tracking for Cached and Reasoning components • **Model Breakdown**: Per-model usage statistics and cost analysis • **Project Stats**: Track token usage by GitHub repository • **Trend Forecasting**: Predict future usage patterns 🏆 Community Features • **Global Leaderboard**: Daily, weekly, monthly, and all-time rankings with privacy-safe display names • **Public Profiles**: Share your AI usage journey with a privacy-safe public profile • **Leaderboard Categories**: Compete in overall rankings or by specific models (GPT, Claude, etc.) 🔒 Privacy Guarantee We believe your code and thoughts are your own. VibeUsage is built with strict privacy pillars to ensure your data never leaves your control. | Protection | Description | |------------|-------------| | 🛡️ **No Content Upload** | Never upload prompts or responses - only compute token counts locally | | 📡 **Local Aggregation** | All analysis happens on your machine - only send 30-minute usage buckets | | 🔐 **Hashed Identity** | Device tokens are SHA-256 hashed server-side - raw credentials never stored | | 🔦 **Full Transparency** | Audit the sync logic yourself in - literally only numbers and timestamps | 📦 Installation Standard Setup Initialize your environment once - VibeUsage handles all synchronization automatically in the background: Authentication Methods • **Browser Auth** (default) - Opens browser for secure authentication • **Link Code** - Use to authenticate via dashboard-generated code • **Password** - Direct password authentication (fallback) • **Access Token** - For CI/automated environments CLI Options Auto-Configuration Once completes, all supported CLI tools are automatically configured for data sync: | Tool | Config Location | Method | |------|----------------|--------| | **Codex CLI** | | hook | | **Every Code** | (or ) | hook | | **Gemini CLI** | (or ) | hook | | **Opencode** | Global plugins | Message parser plugin | | **Claude Code** | | Hook configuration | | **OpenClaw** | Auto-links when installed | Session plugin (requires restart) | No further intervention required! 🎉 💡 Usage Manual Sync While sync happens automatically, you can manually trigger synchronization anytime: Health Check Run comprehensive diagnostics to identify issues: Debug Mode Enable debug output to see detailed request/response information: Uninstall 🏗️ Architecture Tech Stack • **CLI**: Node.js 20.x, CommonJS • **Dashboard**: React 18 + Vite + TailwindCSS + TypeScript • **Backend**: InsForge Edge Functions (Deno) • **Database**: InsForge Database (PostgreSQL) • **Design**: Matrix-A Design System Components • **Tracker CLI** ( ): Node.js CLI that parses logs from multiple AI tools and syncs token data • **Core Relay** (InsForge Edge Functions): Serverless backend handling ingestion, aggregation, and API • **Dashboard** ( ): React + Vite frontend for visualization • **AI Analytics Engine**: Cost calculation, model breakdown, and usage forecasting Data Flow • AI CLI tools generate logs during usage • Local detects changes and triggers sync • CLI incrementally parses logs, extracts token counts (whitelist fields only) • Data aggregated into 30-minute UTC buckets locally • Batch upload to InsForge with idempotent deduplication • Dashboard queries aggregated results…