AgentSpend — AI Cost Monitor

Track your AI API spending across models, projects, and time.

Features

• 📊 Real-time cost dashboard with interactive charts
• 🤖 Multi-model tracking (Claude, GPT, Gemini, DeepSeek)
• 📁 Per-project cost breakdown
• ⚠️ Budget alerts and spending warnings
• 💡 Built-in cost-saving tips
• 📤 JSON export for external analysis
• 🔒 All data stored locally (localStorage, no backend)

Quick Start

1. Open index.html in any browser
2. Start adding AI sessions via "+ Add Session"
3. Watch your costs in real-time

Pricing Models Supported

• Claude Opus 4.8 ($15/$75 per 1M tokens)
• Claude Sonnet 4.6 ($3/$15)
• Claude Haiku 4.5 ($1/$5)
• Claude Fable 5 ($5/$25)
• GPT-5.4 ($2.5/$15)
• GPT-5.3 Codex ($1.75/$14)
• GPT-5 Mini ($0.25/$2)
• Gemini 3 Pro ($2/$12)
• Gemini 2.5 Flash ($0.3/$2.5)
• Custom pricing support

Tech

• Single HTML file, zero dependencies except Chart.js CDN
• Dark theme, responsive design
• localStorage persistence
• No tracking, no analytics, no server

License

MIT — free to use, modify, and distribute.