A Claude Code mod that shows context window usage, session cost, usage limits, and burn rate
One line above the prompt: context tokens with their weather, a bar per recent prompt, your 5-hour and 7-day limits against the time elapsed, the prompt cache with its time left, the session cost with the last prompt's share in dollars and in points of the 5-hour limit, and the subagents running. Labels in English, French or Russian. After Token Weather (Anthropic, Apache-2.0) and the ideas of usage-meter (HolyGrail) and prompt-cache-control (Daniel San).
1 copy
Preview
Terminal
❯ Run the tests again
● The repo is clean and the tests pass.
✻ Cooked for 0s · done 10:12 PM
☀ 44k │ 5h 26% │ 7d 7% │ cache 5% [-]
❯ Claude Desktop

What it does
A band above the prompt with your context tokens, 5-hour and 7-day limit use, prompt cache time left, session cost and running subagents.
- What you see
- A band of usage pills above the prompt
- Works in
- Terminal and Claude Desktop
- Slash commands
- None
- Uses your tokens
- No
- Changes files
- Nothing
- Network access
- None
- Reviewed commit
- 969a724
- Preview recorded with
- Claude Code 2.1.296
How it works
- Reads the context, limits, session cost and cache figures Claude Code reports to plugins.
- Compares 5-hour and 7-day use with the time elapsed, green, yellow or red by pace.
- Counts down the prompt cache: 1 hour on a subscription, 5 minutes on an API key.
- Draws pills in Claude Desktop and a single text line in the terminal.
Good to know
- Dollar figures use Anthropic's list prices, so on a subscription they are API-price estimates, not a bill.
- In Claude Desktop, hover a pill for more, such as the reset time or the cost split by model.
- Labels are in English, French or Russian: set the Language option, or it follows LANG.
More like this
Terminal
❯ /deck demo
⎿ deck: Sample plan running above the prompt, with sample routing until real ones arrive.
[-]
0: deck ▾ · 5 rows 22:11:41
⇄ Model route opus = opus jev: haiku · not applied effort ▁▃·· medi = kept ✕
✓ Orders module Done · 12/12 · 2/2 agents 100% ✕
● Context 44k / 200k 22% ✕
◔ Session · 5h 5h · 26% used ↻ 2:04:24 ✕
◔ Weekly · 7d 7d · 7% used ↻ 2d 0h 57m ✕
❯
⏵⏵ auto mode on (shift+tab to cycle) · ← for agents [ Deck 1 ]Claude Desktop

A cockpit for Claude Code: plan bars and subagent strips, context, usage limits with reset countdowns and burn rates, spend, the model route and a pet. Six bar styles, three collapsed looks. CLI and Desktop.
Terminal
BURN ▁ ▂ *▂▆ ▃ $0.41 = 0.04 burritos [-]
5h 26% in 2h 5m ▅▄█▅▅▆█▆███▅█ opus 4.8 44k in 1k out
wk 7% in 2d 0h █████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░ lifetime $0.41 hide
❯ Claude Desktop

Live session cost odometer above the prompt: a burning fuse, real-world comparisons, threshold alerts
Terminal
[-]
zZ
▀▄ ▄▀
▄▄▀▀▀▀▀▀▀▀▀▀ ▀▀▀▄▄▄
▀▀▀▀▀▀▀▀▀▀▀▀ ▀▀▀▀▀▀▀▀
ctx 22% [ Pet ♥ ] [ ⚙ ] [ ↻ Adventurer ] [ Hide ]
❯ Claude Desktop

A little pixel cat that wanders above the Claude Code prompt, watches your usage limits and announces when work finishes; or switch packs for an adventurer who mines, chops trees and fights zombies
Terminal
╭─────────────────────────────────────────────────────────────────────────────────────────────╮ [-]
│ ◆ context 44k of 200k · compacts at 167k 22% │
│ ████████████████████────────────────────────────────────────────────────────░░░░░░░░░░░░░░░ │
│ ■ system prompt 1.9k 1% ■ system tools 28k 14% ■ skills 2k 1% ■ messages 12k 6% │
│ ─ free space 123k ░ autocompact buffer 33k │
╰─────────────────────────────────────────────────────────────────────────────────────────────╯
❯ Claude Desktop

Your context window as a stacked bar above the prompt, a color per category, with token counts and the compaction point. /context-bar shows or hides it.
Opus 4.8 [high] ~/app main
16:20:55 | ⛁ ██░░░░░░░░ 22% | 5h ●◔○○○ 26% ↻2h7m | 7d ◔○○○○○○ 7% ↻2d1h │ $0.41 ⏱ 10m
Two-line status with a block context bar and quarter-step dot meters (◔◑◕●) for the 5-hour and weekly limits (fill shows usage, color shows whether you are ahead of or behind pace), plus model, effort, git and cost. Needs no jq.
Opus 4.8 high ✦ │ ctx 22% │ app (main) │ PR #1287 ● │ $0.41 · 10m +128 -34
Session ● ● ● ● ● ● ● ● ○ ○ ○ 74% left Resets in 2h 6m
Weekly ● ● ● ● ● ● ● ● ● ● ○ 93% left Resets in 2d 0h
Fable ● ● ● ● ● ● ● ● ● ○ ○ 85% left Resets in 2d 0h
◐ Read:index.ts ✓ Read×1 ✓ Bash×1 ✓ Edit×1 ✓ Grep×1
◐ code-reviewer Review the change for regressions
A header with model, context and burn rate per hour, then live tool and agent activity parsed from the session transcript.
Model: Opus 4.8 | Thinking: high | Context: [▰▰▱▱▱▱▱▱▱▱] 45k/200k (22%) | cwd: ~/app | ⎇ main | (+0,-0)
Cost: $0.41 | Block: 2hr 6m | Weekly: 7.0% | Session: 10m | In: 916.7 t/s | Out: 29.2 t/s | Cached: 2k | Session: [▰▱▱▱▱] 26.0%
Mem: 0.8G/1.0G | (*゚∀゚*) | Session ID: b8e1c0d2-4a6f-4e2a-9c1b-3f5d7a9e2c10
Three lines in One Dark: model, thinking effort and a context bar up top; cost, quota, session clock and token throughput in the middle; weather, free memory and a rotating kaomoji below.
Opus 4.8 (200K) high | app (main)
██░░░░░░░░ 22% 200K | 5h 26% ⇣32% 2h 7d 7% ⇣63% 2d
Are you burning quota faster than the clock? Five-hour and seven-day usage with over- and under-pace arrows, plus a context bar and git diff stats. Reads rate limits from stdin only.