Paste your tenant API key to view your dashboard. The key stays in this browser (localStorage) — nothing is sent anywhere except to this dashboard's own backend.
Flat plans change how we label numbers. If you're on Max 20× you don't pay per token — but your API-equivalent value shows what the same usage would cost on usage-based billing.
Anthropic doesn't publish exact per-plan quotas. We use conservative public estimates (Pro ≈ 45 Sonnet / 10 Opus per 5h, Max 5× ≈ 5× that, Max 20× ≈ 20× that). Treat utilization as a trend indicator, not a precise meter. Mark coverage as complete if these connected sources are all of your Claude usage.
| Model | Calls | Cost |
|---|
| Project | Calls | Cost |
|---|
| User | Calls | Cost |
|---|
Click a session to drill in. Red avg cache/call ≈ context bloat; near-zero avg output/call with many calls ≈ silent watcher. Use the filter to isolate automated / sidechain traffic from your own typing.
| Session | Range | Type | Models | Calls | Avg cache/call | Avg output/call | Cost |
|---|