AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
翻譯準備中
A live Claude Code status line: model, git branch, context window, 5-hour & weekly usage, reset countdowns, and session cost. Ships as a self-contained Claude Code plugin.
翻譯準備中
取自儲存庫 README
正在載入完整 README。 以下為已儲存的摘錄。
A live status line for Claude Code that shows your model, git branch, context window usage, 5-hour and weekly rate limits, reset countdowns, and session cost. It reads the official Status Line JSON payload on stdin and prints a formatted line. No polling, no /usage scraping, no undocumented APIs.
• Model name (bold)
• Git branch (blue)
• Context window usage with color-coded percentages
• 5-hour usage window — once it passes 50%, the segment also shows how much time is left, e.g. 5h 61% (2h 13m)
• Weekly (7-day) usage window
• Session cost in USD
• Working directory (optional)檢視完整 README ↗目前無法載入完整 README。請透過原文連結閱讀。 以下為已儲存的摘錄。
A live status line for Claude Code that shows your model, git branch, context window usage, 5-hour and weekly rate limits, reset countdowns, and session cost. It reads the official Status Line JSON payload on stdin and prints a formatted line. No polling, no /usage scraping, no undocumented APIs.
• Model name (bold)
• Git branch (blue)
• Context window usage with color-coded percentages
• 5-hour usage window — once it passes 50%, the segment also shows how much time is left, e.g. 5h 61% (2h 13m)
• Weekly (7-day) usage window
• Session cost in USD
• Working directory (optional)檢視完整 README ↗