token-usage

Claude Code skill — delegate coding tasks to Mistral Vibe, supervise via git diff & save 50-90% Claude token usage

⚡ Real-time token, context & agent dashboard for Claude Code — zero polling, pure stdin

Status line for Claude Code: truecolor context window usage bar, live session cost, token usage, 5h/7d rate limit countdowns with reset times, plain action hints. Bash + PowerShell, zero dependencies, macOS / Linux / Windows.

Stop hitting Claude Pro's weekly limit by Wednesday. Delegate routine & complex tasks from Claude Opus 4.7 to DeepSeek – slash commands included.

Real-time monitoring dashboard for Claude Code CLI sessions — track token usage, session health, and context window utilization

Spec-kit extension that tracks estimated LLM token cost per workflow step, with per-model pricing and cumulative USD reporting

Token consumption pace viewer for Claude Code — browser-based, one-line install (Linux/macOS/WSL)

Lightweight X11/tmux-based token usage visualization for LLM systems. Simulates and monitors token consumption patterns using Unix tools (xload + named pipes) as a local development alternative to CloudWatch/Prometheus metrics. Useful for developing intuition about token usage patterns without deploying full observability stacks.

macOS menu bar usage monitor for OpenAI Codex and AI coding agents

以修仙主題呈現 Claude Code 的本機用量 — statusline + CLI。純本機、零遙測。macOS / Windows x64

Track Claude Code token use and spending in real time to monitor costs and optimize your daily $20 budget efficiently.

Battery-style context-window statusline for Claude Code. Works with non-Anthropic models (GLM, DeepSeek, Qwen…). Pure bash+jq, auto-pricing.

Aggregate ccusage (Claude Code / Codex / OpenCode) token usage across multiple machines via incremental SSH push, then run one unified report.

Live quota, reset countdowns, context usage, model, cost, and git activity for Claude Code.

A Dockerized LiteLLM proxy setup that provides a unified API for multiple Large Language Model providers including Anthropic Claude, OpenAI GPT, and others.

a statusline for claude code that looks like it knows something you don’t

Keep your AI coding agent fast, cheap, and accurate: lean, load-on-demand context + a harness (steps, tools, loops, gates). Portable markdown + a plug-and-play Claude Code plugin.

A lightweight Claude Code status line that tracks context window usage per turn, with a live delta, growth sparkline, fill percentage, and 5h/7d rate limits.