Enhancement: claude-usage plugin
- Dominant language
- Shell
- Stars
- 856
- Forks
- 368
- Avg merge
- 2m
- Merged PRs (30d)
- 1
Description
I've built for fun and for my own use a claude-usage plugin.
If you think it could interest other users, I'd be happy to work with you to have it packaged in your repo.
You can find it here for now : https://github.com/Wayrkin/tmux-dracula-claude
⚠️ My understanding of usage limits and token consumption, globally or per model, might be a bit flaky, so fill free to propose corrections, improvements, critics, etc
It has :
- usage rate for current session (if any)
- usage rate for 5h and weekly periods
- claude work dir detection (based on .claude or CLAUDE.md)
- caching for "heavy" stats
- dracula-type customisation
It is missing, for now :
- default limits for free-plan
- fg/bg customisation (`@dracula-claude-usage-colors`)
Caveat :
- there's a limitation for now when trying to detect the default pane and current Claude context when multiple clients are attached to multiple tmux sessions
disclaimer : it is fully built with AI (Claude Sonnet 5) but with my directives and guiding all along
I've read the CONTRIBUTING.md.
- issue description is fully mine
- I've been monitoring and guiding the LLM all along, nothing was fully autonomous
If you find in interesting that this feature joins the plugins, I'll open a PR :
- the commit messages and the PR description will be mine also
- code comments will disappear or be rewritten by me
- I'm ready to explain and justify the code if need be
I'll be happy to address any other concern and I'm open to critics and suggestions.
Contributor guide
Research direction
Read CONTRIBUTING.md and inspect the linked tmux-dracula-claude repository, then compare its plugin structure with this repository's conventions. Clarify maintainer acceptance and required scope before opening a PR; done means an agreed plugin integration with documented handling of usage limits, customization, and pane/context detection.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- shell
- Domain
- cli
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 38/100