Stop guessing what your AI costs. The Tinker Command Center shows every token, every dollar, and every byte of context in real time — per model, per session, per tab. The black box gets a cost meter. Part of our open Building Jarvis series.
What it tracks
The Command Center is the cost and usage surface of the TinkerClaw UI. It breaks spend down by model so you can see which brain is eating the budget, by session so you can see which conversation is expensive, and by context size so you can see how close a tab is to its window. Toggle between the current session and the whole gateway — one tab or the entire fleet.
Why it matters
Agent cost is invisible by default, and invisible cost is the kind that surprises you at the end of the month. The Command Center makes spend a number you watch while you work, not a bill you discover later. It turns “the agent did a lot” into “the agent did a lot, here, for this many dollars” — the difference between a tool you trust and a tool you fear.
Pairs with the slider
Seeing the cost is half the loop; controlling it is the other half. The Command Center reports what was spent; the effort × model slider lets you decide what to spend next. Together they close the loop between observation and control.
TinkerClaw is open source. The whole architecture — brain-region anatomy, recipes, the effort allocator, the EEG — lives at github.com/globalcaos/tinkerclaw. This post is part of the open Building Jarvis series at The Tinker Zone.
Leave a Reply