Skip to content
Cagent

Status line

The status line summarizes the active session below the composer. It updates as the mode, agent, model, context usage, and costs change.

Customize interactively

Section titled “Customize interactively”

Run /statusline to open the editor. It includes a live preview.

  • Press Enter to show or hide the selected module.
  • Press Left or Right to reorder it.
  • Press c to choose a named color or enter #RRGGBB.
  • Press r to reset that module’s color.
  • Press Esc to save and close.

If you enable a module listed after an enabled hint, the editor moves hint immediately after that module. You can still move hint anywhere afterward with Left or Right.

Modules

Section titled “Modules”
Module Shows
mode Active permission and behavior mode.
agent Active non-default agent profile.
model Model and reasoning effort.
provider Model provider.
fast fast while the selected model’s Fast service tier is active; otherwise hidden.
provider_usage Provider-reported remaining limits.
context Context-window percentage.
cost Complete session cost.
tokens Session input and output tokens.
token_rate Average input and output tokens per second over the trailing 60 seconds of active model work.
cache Session cache-read rate.
cache_tokens Cache read and write tokens.
context_tokens Context-window tokens used.
hint Contextual composer shortcuts and combined background-terminal/sub-agent count.

Some values appear only when relevant or available from the provider. On narrow terminals, Cagent removes or shortens lower-priority segments to keep the line usable; enabling a module does not guarantee it is visible at every width.

token_rate is disabled by default. When enabled, it displays provider-reported usage such as in:123/s out:71/s and updates when the provider reports new token usage. Its trailing 60-second clock advances only while the conversation is actively working, resets when you change models, and is hidden when both rates are zero. It also resets when Fast mode becomes effective or stops being effective, so the displayed average does not mix Standard and Fast requests. It uses reported request usage rather than estimating tokens from streamed text.

Configure with TOML

Section titled “Configure with TOML”

Set the visible modules in display order:

[ui.statusline]
modules = [
  "mode",
  "agent",
  "model",
  "provider",
  "fast",
  "context",
  "provider_usage",
  "cost",
  "token_rate",
  "hint",
]


[ui.statusline.colors]
agent = "green"
model = "light-cyan"
provider = "magenta"
fast = "red"
context = "yellow"
provider_usage = "green"
cost = "light-yellow"
token_rate = "gray"
hint = "dark-gray"

Colors accept #RRGGBB or a named terminal color. The mode module uses the active mode’s color; configure that under [modes.<name>]. See Configuration for the complete color list.

Section titled “Related”