Summary
Real-time token usage monitoring across all LLM calls — input, output, cached, and reasoning tokens displayed per-request and cumulative.
Mission Objective
Token Burn Dashboard gives agent makers full visibility into how every LLM call consumes tokens. Input, output, cached reads, and reasoning tokens are all tracked per-request and rolled up into session-level totals.
Current Operations
The dashboard renders live telemetry from the event stream, updating cost and token counts on every message_end event. Users can filter by model, time range, or task profile. Package burn gauges show remaining quota against Starter, Pro, and Enterprise tiers.