agent-usage-all-in-one 0.3.0 → 0.4.0

package/README.md

@@ -1,4 +1,4 @@
-![Agent Usage banner](static/brand/agent-usage-banner.svg)
+![Agent Usage dashboard showcase](static/brand/agent-usage-showcase.jpg)
 
 # Agent Usage
 
@@ -24,6 +24,16 @@ The dashboard has two primary tabs:
   model rankings with visual share bars and the public API retail equivalent of
   eligible token usage.
 
+Declare what each plan costs in Settings and the same tab adds a **subscription
+value** view: it prorates that price over the selected range and compares it with
+the API retail equivalent of the tokens the plan produced, mapping every
+subscription against a break-even line and ranking the value multiple beside what
+a million tokens actually cost you. Adding the plan's renewal date also measures
+each subscription over its own billing period, showing how far into that cycle it
+is and whether it is on pace to pay for itself, so a plan in its first week is
+never read against one in its fourth. The plan price is a local declaration; it
+never becomes a cost record and never enters any cost total.
+
 Actual charges, provider-reported estimates, fixed subscription fees, and API
 retail equivalents are separate evidence. The API retail equivalent is not a bill
 and is never presented as subscription spend. Unknown models or prices remain

package/README.zh-CN.md

@@ -1,4 +1,4 @@
-![Agent Usage 横幅](static/brand/agent-usage-banner.svg)
+![Agent Usage 仪表盘功能展示](static/brand/agent-usage-showcase.jpg)
 
 # Agent Usage
 
@@ -19,6 +19,12 @@ API 对等费用、历史与诊断。它只提供建议，不会自动切换 Age
 - **Agent 用量**：保留各平台原生的 5 小时、周、月、All models 和 Fable only 等额度标签与刷新时间。
 - **Token 与模型费用**：支持 24 小时、7 天和 30 天范围。顶部优先展示汇总信息，随后用图表展示 Provider 占比和可交互的每日趋势，再以占比条呈现模型排行及符合条件的 Token 按公开 API 价格计算出的对等费用。
 
+在设置中填写各订阅的价格后，同一标签页会多出**订阅性价比**视图：把订阅价按所选时间窗摊销，
+与该订阅产出的 Token 的 API 对等价做对比，在图谱上以保本线定位每个订阅，并给出价值倍数、
+你实际付出的每百万 Token 单价与市场单价。再填上续费日，还会按每个订阅**自己的计费周期**单独衡量：
+显示本周期已经走了多少天、已回本多少，以及按当前进度能否回本，这样处在第一周的订阅不会和处在第四周的订阅被放在一起比。
+订阅价只是本机声明，不会变成费用记录，也不会进入任何费用合计。
+
 实际账单、平台报告估算、固定订阅费和 API 对等零售价是四类互不混合的证据。
 API 对等零售价不是账单，也不会被描述成订阅支出。模型或价格未知时保持未分类或未定价，
 不会猜测，更不会显示成 0。