mikecase/opencode-cost-monitor

OpenCode Cost Monitor is a small Python tool that shows you a live dashboard in your terminal tracking how much your opencode sessions are costing. Opencode is an AI coding assistant that stores its usage data in a local database file, and this monitor reads that file every few seconds and displays up-to-date spending information without you needing to leave the terminal. The dashboard breaks down costs by AI model, showing how many sessions used each one, how many tokens were processed, and what percentage of responses came from cache rather than being generated fresh. Cache responses cost less, so the tool also calculates how much you saved compared to what the same usage would have cost without caching. A small sparkline chart at the bottom shows your daily spending over the last 30 days so you can spot if usage is trending up. You can press a key to filter by different time ranges: all time, the last 30 days, the last 7 days, or just today. Another key cycles between opencode's two pricing plans (Go and Zen) so you can see the numbers recalculated for whichever subscription you have. Setup is two commands: create a Python virtual environment, install one dependency called textual, and run the script. The tool finds the opencode database at the standard location automatically, or you can point it at a different path. The project is MIT licensed and is a single script built for developers who use opencode and want visibility into their AI usage costs as they work.