Usage 0.28.21: A warning light now appears when a service is having trouble

Release notes

FEATURE

A warning light now appears when a service is having trouble

When usage feels slow, it can be hard to tell whether you hit a limit or the service itself is struggling. Panels with a footer now show an orange-red banner when Claude Code, the Claude API, or the Codex API has an outage or degraded performance. The banner reads each service's public Statuspage.io page; it never calls an LLM usage API, so it reports service health rather than your AI consumption. Each tool uses its own panel colors, and a hidden tool card produces no banner. The wording is available in all five interface languages. Think of separate dashboard lights in a car: the light tells you which fitted component has trouble. Antigravity has no usable public status page, so it does not pretend to have one.

FEATURE

Footer details now sit side by side, with room for warnings

Scanning the eight themed panels now makes it easier to see usage rate and service status at the same time. Those footer details used to be stacked vertically; they now sit side by side, matching the default panel, like two notes placed on one row instead of one covering the other. If a service-status banner appears, the panel also grows tall enough to give it space rather than squeezing the existing content. It is like adding a notice to an elevator without cramming the floor buttons out of sight: the cabin simply makes room.

FEATURE

Heavy-use labels stop counting cache rereads

If usage kept labeling you Heavy while you were only continuing one conversation, that false impression should now be much rarer. The live rate sorts activity into Idle, Normal, Active, and Heavy by tokens per minute—tokens are the units of text an AI processes and bills for. Previously, the Heavy calculation included cache_read, a nearly free reread of prompt-cache content that mainly reflects conversation length rather than intensity. That value is now excluded, and the thresholds have moved from 50 / 250 / 1000 to 500 / 2500 / 6000 tokens per minute. It is like a gym counting rereading the same workout sheet as exercise; now it counts the training itself, so the label better matches the effort.

Original changelog

Original changelogA warning light now appears when a service is having trouble

**Service-status banners for Claude Code, Claude API, and Codex API incidents.** `usage` reads the public Claude and Codex Statuspage.io status pages and shows an orange-red warning banner at the bottom of a panel only when an affected service has an outage or degraded performance. It never calls an LLM usage API.
**Service-status banners work across every panel with a footer.** Each panel uses warning colors matched to its own theme, and a banner for a tool is not shown when that tool's card is hidden.
**Service-status banner text is available in all five UI languages.** Antigravity is not supported because it has no usable public status page.

Original changelogFooter details now sit side by side, with room for warnings

**Rate and status information now sit side by side in the footers of all eight themed panels.** They now match the default panel instead of being stacked vertically.
**Popover height now accounts for the service-status banner.** When the banner appears, the panel grows enough to contain it instead of squeezing its content.

Original changelogHeavy-use labels stop counting cache rereads

**Live usage-rate categories can distinguish heavy use again.** Heavy users previously remained permanently in Heavy because burn rate counted `cache_read`—a near-free prompt-cache reread that reflects conversation length rather than usage intensity—in its numerator. It now excludes `cache_read`, and the Idle / Normal / Active / Heavy thresholds have been recalibrated from 50 / 250 / 1000 to 500 / 2500 / 6000 tokens per minute.