See where your Claude Code limits are headed before you hit them. The usage dashboard individual Pro and Max plans don't get from Anthropic - single binary, zero config, one browser tab.
Live 5-hour and 7-day utilization with reset countdowns and color-coded thresholds, plus separate Opus and Sonnet quotas when they apply.
An empirical-Bayes model projects where each limit window lands at reset, with an 80% credible band and ETA to threshold. Refit daily from your own past windows, published as a spec, and benchmarked out-of-sample.
Input, output, cache read, cache write tokens per session. Estimated API-equivalent costs at current pricing.
Browse today's or recent sessions with token counts and costs. See running Claude Code processes live and stop them from the dashboard.
Search across all CLAUDE.md files, rules, and auto-memories. Filter by type, spot stale entries, get consolidation suggestions for overlapping files.
Interactive visualization of memory relationships. Per-file health grades (A–F) with improvement suggestions for stale or poorly structured memories.
Server-Sent Events push changes instantly. No polling, no manual refresh. Gauges, sessions, and memories update live.
Run claumon service install once. Dashboard starts on login, restarts on crash. Works on macOS, Linux, and Windows.
Update via brew upgrade claumon, or claumon update if installed from a binary.
No CGO, no native dependencies. One binary per platform.
brew install fabioconcina/claumon/claumon on macOS or Linux. Or grab a prebuilt binary.
claumon --open for a quick look, or claumon service install to start it on login. Credentials are picked up automatically.
brew upgrade claumon and you're done.
