[FEATURE] Usage dashboard / at-a-glance summary command in CLI
Preflight Checklist
- [x] I have searched existing requests and this feature hasn't been requested yet
- [x] This is a single feature request (not multiple features)
Problem Statement
When using Claude Code CLI, there is no way to quickly check usage at a glance — context window consumption, plan usage, and model activity. The claude.ai web/desktop app has a dedicated usage overview with this information visible in one place, but the CLI offers no equivalent. Users who work primarily in the terminal have to leave their workflow and navigate to a browser to check basic usage information.
Proposed Solution
A command like claude usage or /usage that prints a compact summary directly in the terminal, showing:
- Current context window usage for the active session
- Plan usage (tokens or requests consumed vs. limit for the billing period)
- A per-model breakdown of recent activity (similar to the weekly overview in claude.ai)
The output should be readable at a glance without leaving the terminal — a single structured block, not a URL redirect.
Alternative Solutions
Currently the only workarounds are visiting claude.ai/settings/billing or console.anthropic.com/usage in a browser. Neither is accessible from the terminal and both require context-switching out of the CLI workflow. No in-CLI alternative exists.
Priority
Low - Nice to have
Feature Category
CLI commands and flags
Use Case Example
Example scenario:
- I'm deep in a coding session using Claude Code CLI
- I want to know how much of my plan I've used this week or how close I am to
my context limit
- I run
claude usagewithout leaving the terminal - I get a compact summary and continue working — no browser, no context switch
Additional Context
The claude.ai desktop/web app already has this — a one-screen usage overview with context window, plan consumption, and weekly model usage. The feature request is simply to expose equivalent information natively in the CLI for users who work primarily in the terminal.
This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗