Skip to content
savvy773 edited this page May 26, 2026 · 3 revisions

AI Usage Dashboard

AI Usage Dashboard

Monitor Claude, Codex, and Gemini CLI usage in one local dashboard.

Landing Page   Release      

Quick Start · Architecture · Fix Checklist · API Reference



The browser never runs CLIs directly. A SvelteKit server API spawns each CLI in a virtual terminal via node-pty, parses the output, and writes results to data/usage-history.json. The UI renders from cached JSON on first paint, then refreshes in the background.


✨ Features

Feature Description
Multi-provider Claude /usage · Codex /status · Gemini CLI /model via virtual terminals
Smart retry Up to 5 attempts per provider with phase diagnostics and slash-command recovery
📊 Weekly Pace card Actual usage bar vs. 20% minimum threshold marker
Reset countdown Live per-provider countdown to next usage reset
📡 Live server logs SSE-based real-time panel — streams directly to the browser
💾 Dual cache Server-side JSON history (10-min buckets) + browser localStorage fallback

🖥 Providers

Provider Command Slash Display
Claude claude /usage Current session + weekly usage
Codex codex /status 5h limit + weekly limit
Gemini CLI gemini --skip-trust /model Per-model usage rates + resets

🚀 Quick Start

irm https://raw.githubusercontent.com/savvy773/ai_usage/main/scripts/install.ps1 | iex

Full Quick Start guide


📖 Wiki Pages

Page Description
Quick Start Install and run in under 2 minutes
Architecture Server, collector, parser, and storage design
Fix Checklist Step-by-step diagnostics for collection errors
API Reference REST endpoints and payload shapes


MIT License · Local. Private. Yours.

Clone this wiki locally