# Usage

The Usage page shows how much of each harness's quota is left and how many tokens your agents have used.

## Open usage

Select the gauge button (**Usage**) in the top bar, or on the projects page, to see how much your agents are using. It covers every project and every harness.

- **Remaining** shows each harness's quota windows, with the percentage left and when it resets. Claude Code reports these during Integrated sessions, and Codex records them in its transcripts. Until Patterm has seen one, the section says so.
- **Tokens** shows input, output, cache writes and cache reads for **Today**, **7d**, **30d** or **90d**, and what they would cost at pay-as-you-go API prices. A subscription replaces those prices, so treat the figure as a yardstick. The per-day chart switches between **Tokens** and **Cost**.
- **By model**, **By project** and **Skills** break the same period down. Skills counts how often each was used, by the agent or by you.

![The Usage page. Remaining quota windows for Claude Code and Codex are at the top, followed by token totals and a bar chart of tokens per day.](https://patterm.com/images/docs/usage-limits-and-tokens.webp)

Token figures come from imported [conversation history](https://patterm.com/docs/conversation-history.md), so turn on a history source first. Tokens from models without a public price are counted but left out of the cost.

## Related

- [Conversation history](https://patterm.com/docs/conversation-history.md)
- [Supported harnesses](https://patterm.com/docs/harnesses.md)

---

This is one page of the Patterm documentation. The index of every page is at https://patterm.com/llms.txt.