If you bounce between Cursor, Claude Code, Windsurf, and Copilot like most developers do in 2026, you probably have no idea what you're actually spending or which tool is pulling its weight. Agentlytics is a new open-source CLI tool that pulls usage data from 14 AI coding editors into a single local dashboard.
Run npx agentlytics and you get cost estimates broken down by model, editor, and project, plus activity heatmaps, usage streaks, and the ability to search through AI conversations across every editor you use. Everything runs locally on your machine, so no conversation data leaves your computer. The tool supports Cursor, Claude Code, GitHub Copilot, Windsurf, and ten other editors, covering most of the AI coding landscape.
The practical value here is straightforward: if you're paying for three or four AI coding subscriptions, you should know which ones you're actually using and what they're costing per project. Most developers I know have strong opinions about their preferred AI editor but zero data to back it up. Agentlytics won't tell you which tool writes better code, but it will tell you which ones you reach for most, when you're most productive, and where the money is going. For teams, there's a stats-sharing feature that could help standardize tooling decisions based on actual usage patterns rather than vibes. It's a simple utility solving a real blind spot in the multi-editor workflow most AI-assisted developers have landed on.