@qomob/dsh-madrank
MADRank Usage — track your AI usage in DeepSeek Harness, see how you rank.
安装
此插件尚未提供可验证的 bundle，或兼容性检查未通过。请先阅读仓库说明。 阅读完整 README ↗
说明文档
阅读完整 README ↗使用方式
Sidebar → Quick View = VIEW（看数据）
Settings → MADRank = CONFIGURE（改配置）
Quick View 查看使用情况、7 日趋势、模型分布与 Rank；Settings 负责参与排名开关、自动同步、隐私、本地/远端数据删除与插件状态。
Why Usage Ranking?
Most AI rankings measure model capability or preference. MADRank measures actual AI usage: who is using AI, how much, and who accumulated the most meaningful token usage over the last 7 UTC days. It is designed as an AI Usage Layer, not a model benchmark.
Usage Accuracy
MADRank uses DSH session projections and validates aggregation against an independent reference token-meter. A real-session reconciliation run:
input 745,120 = 745,120
output 186,324 = 186,324
cache read 26,352,384 = 26,352,384
MATCH ✓
The suite covers streaming replacement, waterfall tool traffic, cross-midnight bucketing, and replacement semantics.