bi-boo

我的 Claude Code 个人配置——CLAUDE.md 规则、Agent、记忆机制,以及 Ghostty 终端配置

18
0
100% credibility
Found Mar 24, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Shell
AI Summary

A shared personal setup of rules, guides, agents, and a custom bottom-bar display to improve AI-assisted coding sessions by engineering consistent behavior and tracking real-time usage and costs.

How It Works

1
🔍 Discover the Setup

You stumble upon this handy collection of tips and tools shared by someone to make their AI coding helper smarter and more trackable.

2
📥 Bring It Home

You copy the special guides, rules, and trackers into your AI helper's personal folder to prepare for better sessions.

3
🔧 Prepare the Tracker

You make sure the little tool that shows your usage stats is ready to go by giving it permission to run quietly in the background.

4
Connect Your Usage View

You link it to your AI account details so it can peek at how much you're using without interrupting your work.

5
🚀 Launch Your Enhanced Terminal

You open your favorite coding terminal, now supercharged with smart rules and a live dashboard at the bottom.

6
Watch the Magic Unfold

Right away, you see colorful info on the model in use, token counts going up and down, context fullness, estimated costs, and subscription levels updating live.

🎉 Code Smarter Every Day

Now your AI helper follows better plans, hands off tasks smoothly across chats, and you stay on top of usage to code efficiently without surprises.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is claude-code-config?

This GitHub repo delivers a personal claude code config file setup for the Claude Code CLI, engineering consistent AI behavior in your terminal sessions via rules, model-specific agents, and persistent memory. It tackles flaky AI outputs, context resets between chats, and blind usage tracking by auto-loading global rules, enabling cross-session handoffs, and displaying real-time stats like token counts, context fill, and Pro/Max subscription quotas pulled from Anthropic's OAuth API. Built in Shell with Python helpers, it also includes Ghostty terminal tweaks for seamless Claude Code launches and split-pane workflows.

Why is it gaining traction?

Developers dig the live status bar—model name (color-coded by tier), input/output tokens, context usage (green/yellow/red alerts), and exact quota burns (e.g., 88% with reset timer)—without blocking your CLI flow. Tiered agents route light file scans to cheap Haiku and deep analysis to Opus, while structured debugging paths and lazy-loaded rules keep prompts lean and effective. It's a claude code config github gem for anyone tired of manual prompting or guessing costs.

Who should use this?

Claude Code CLI power users on Pro/Max/Team plans who burn quotas debugging codebases or architecting apps. Terminal hackers running Ghostty or similar, needing session continuity for long tasks like refactoring monorepos. Chinese devs mixing Anthropic with local models (MiniMax, GLM) via claude code configure model tweaks.

Verdict

Fork and adapt if you're deep into claude code install and config—its 18 stars and 1.0% credibility score scream early experiment, with solid docs but zero tests. Great inspiration for custom claude code statusline hacks, less so for production workflows.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.