usage-cli 0.31.3__tar.gz → 0.31.5__tar.gz

{usage_cli-0.31.3 → usage_cli-0.31.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: usage-cli
-Version: 0.31.3
+Version: 0.31.5
 Summary: usage — 在 macOS menu bar 與 Windows 系統匣顯示 Claude Code、Codex、Antigravity 用量的小工具（也提供終端機 TUI）
 License-Expression: AGPL-3.0-only
 Requires-Python: >=3.13
@@ -24,9 +24,9 @@ Dynamic: license-file
 
 # usage
 
-### Quota visibility for Claude Code, Codex, Antigravity, and Grok CLI, built into the macOS menu bar and Windows system tray.
+### Your Claude Code, Codex, Antigravity and Grok CLI quota, always on screen.
 
-Running out of quota mid-session is expensive — especially during a long refactor or debugging run that depends on Claude Code. `usage` surfaces 5-hour and weekly limits *before* you hit the wall, and keeps them visible the whole time. There's no command to run and no page to open; the answer is just there, where you already look.
+`usage` puts your 5-hour and weekly limits in the macOS menu bar or Windows system tray, colored from green to red. Hitting the limit halfway through a long refactor is a bad way to find out you were running low. Now you see it coming. There's nothing to run and no page to open.
 
 [繁體中文](docs/README.zh-TW.md) · [简体中文](docs/README.zh-CN.md) · English · [日本語](docs/README.ja.md) · [한국어](docs/README.ko.md)  |  [Discussions](https://github.com/aqua5230/usage/discussions)  |  [Landing page](https://aqua5230.github.io/usage/)
 
@@ -43,7 +43,9 @@ Running out of quota mid-session is expensive — especially during a long refac
   <img src="docs/showcase-v3.en.png" alt="usage — Claude Code, Codex, and Antigravity quota pinned to the macOS menu bar" width="820">
 </p>
 
-Claude Code and Codex numbers are read passively from log files already on your machine, so **watching your quota never calls Anthropic or OpenAI's LLM APIs** and never costs you a token. Antigravity is the one exception: its quota comes from Google's official quota endpoint using the sign-in the Antigravity CLI already stores locally — a metadata call that doesn't consume your model quota either.
+Claude Code and Codex numbers come from log files already on your machine. Antigravity quota comes from Google's official quota endpoint, using the sign-in the Antigravity CLI already stores locally.
+
+`usage` also helps you spend less. The status line warns you before your context window bloats or your prompt cache goes cold, and a Token Saver toggle keeps replies short. In an A/B test on real sessions, late replies stayed ~40% shorter instead of growing 84% longer.
 
 ## Quick Start
 
@@ -51,10 +53,10 @@ Claude Code and Codex numbers are read passively from log files already on your
 brew install --cask aqua5230/usage/usage
 ```
 
-**Not on macOS?** `uvx usage-cli` runs the terminal interface anywhere, Linux included — no install, no menu bar.
-
 It lands in your Applications folder automatically. Open it once; if macOS 15 or later blocks it, go to System Settings → Privacy & Security, scroll down, and click **Open Anyway**. On macOS 14 or earlier, right-click **Open** once to pass Gatekeeper. Then click the menu bar icon. Prefer a direct download or want the full setup flow? See [Install](#install) below.
 
+**Not on macOS?** `uvx usage-cli` runs the terminal interface anywhere, Linux included — no install, no menu bar.
+
 **Jump to:** [What You Get](#what-you-get) · [Privacy](#privacy--data-sources) · [Requirements](#requirements) · [Install](#install) · [Status Line](#first-launch-set-up-the-status-line) · [Windows](#windows-support) · [Themes](#theme-gallery) · [Troubleshooting](#troubleshooting) · [Comparison](#comparison) · [Not a Fit?](#when-usage-isnt-the-right-fit) · [Development](#development)
 
 ## What You Get
@@ -63,21 +65,22 @@ It lands in your Applications folder automatically. Open it once; if macOS 15 or
 
 - **Always-on Monitor:** Your quota lives in the menu bar, color-coded from green to red. Click when you want the full session, weekly, and per-project breakdown.
 - **Antigravity Support:** Antigravity (Gemini) session and weekly quota show up as a third card in every theme except World Cup 2026, which stays a two-team HUD. Numbers come straight from the official quota API, using the sign-in the Antigravity CLI already keeps on your machine — refreshed every few minutes, with live reset countdowns. Antigravity keeps two separate quota pools: the card shows Gemini by default, and tapping the `Gemini ⇄` tag next to the title switches it to Claude / GPT — the choice is remembered.
-- **Grok CLI Support:** A fourth card reads Grok CLI's weekly credit percentage straight from its own local debug log — no network call. Grok CLI doesn't expose session or burn-rate data, so the card shows a single weekly bar; its per-request token usage still counts toward today's cost and project totals like Claude Code and Codex.
+- **Grok CLI Support:** A fourth card reads Grok CLI's weekly credit percentage straight from its own local debug log. Grok CLI doesn't expose session or burn-rate data, so the card shows a single weekly bar; its per-request token usage still counts toward today's cost and project totals like Claude Code and Codex.
 - **Muse Code Spending:** Muse Code's per-request tokens and cost count toward today's cost, project totals, the HTML report, and the `usage` CLI, read from its own local session logs. Muse keeps no local quota data, so there is no Muse quota card.
-- **Service Status Alerts:** An orange-red banner appears when Claude Code, Claude API, or Codex API has an outage or degraded performance, read from their public Statuspage.io pages — never an LLM usage API. Antigravity isn't covered; it has no public status page.
+- **Service Status Alerts:** An orange-red banner appears when Claude Code, Claude API, or Codex API has an outage or degraded performance, read from their public Statuspage.io pages. Antigravity isn't covered; it has no public status page.
 - **Context Nudges & Notifications:** When your context window hits 70% — or earlier when it is filling fast — the status line nudges you to `/clear` or `/compact` to prevent token waste. You can also opt-in to system notifications for quota limits and recoveries.
-- **Prompt Cache Health:** The status line shows Claude Code's prompt cache hit rate with a countdown to expiry, so you can tell at a glance whether finishing now still reuses the cached context or lets it go cold and re-sends everything. Needs Claude Code 2.1.251 or newer; on older versions the segment simply doesn't appear.
+- **Prompt Cache Health:** The status line shows Claude Code's prompt cache hit rate with a countdown to expiry, so you can tell at a glance whether finishing now still reuses the cached context or lets it go cold and re-sends everything. Once it has gone cold, the countdown turns into how many tokens your next message will re-cache. Needs Claude Code 2.1.251 or newer; on older versions the segment simply doesn't appear.
 - **Hide Sections:** Only use one or two of the tools? Hide the Claude Code, Codex, Grok CLI, or Antigravity section from the menu bar and panels completely with a single click.
 
 ### Workflow Helpers
 
-- **Progress Concierge:** Open a new Claude Code session and `usage` hands your last progress straight to the AI, including your last request, uncommitted changes, and unfinished todos. No `/resume`, no recap. Fully local, off by default.
+- **Progress Concierge:** Open a new Claude Code session and `usage` hands your last progress straight to the AI, including your last request, uncommitted changes, and unfinished todos. No `/resume`, no recap. When you do `/resume` a conversation that sat long enough for its cache to expire, it warns you how many tokens the next message will re-send and suggests `/compact` first. Off by default.
 - **Token Saver:** A menu-bar toggle asks Claude Code and Codex to answer more tersely and in plainer language, saving output tokens while keeping code and error messages byte-exact. A light reminder keeps long conversations from drifting back to verbose — in an A/B test on real sessions, late replies stayed ~40% shorter instead of drifting 84% longer.
-- **Terminal Integration:** `usage status --json` hands your Claude Code and Codex quota to any tool that can run a command — Starship, tmux, or your own scripts. Reads the same local files as the menu bar, no network call. [Ready-made snippets](docs/DEVELOPMENT.md#quota-status-for-other-tools-usage-status).
+- **Auto-start 5-hour Session:** Off by default. Turn it on and, right after a 5-hour quota resets, `usage` sends one tiny message to each tool (Claude with Haiku, Antigravity with Gemini 3.5 Flash Low, Codex with its cheapest model) so the next 5-hour window starts counting right away. Those messages do use a little quota, but the amount is negligible. Checking your quota never sends a message; only this switch does.
+- **Terminal Integration:** `usage status --json` hands your Claude Code and Codex quota to any tool that can run a command — Starship, tmux, or your own scripts. Reads the same local files as the menu bar. [Ready-made snippets](docs/DEVELOPMENT.md#quota-status-for-other-tools-usage-status).
 - **Token-waste Health Check:** A daily background diagnosis scans your logs for waste, including repeated file reads, polluter directories, and noisy Bash output. If it finds issues, a one-line heads-up appears; say "show me" and the AI walks you through fixes.
 
-### AI Teamwork
+### Stay Current
 
 - **AI Update Daily:** Opens a daily-updated public [page](https://aqua5230.github.io/ai-updates/) covering Claude Code, Codex, and Antigravity, with the full history kept. Reviewed items get a plain-language summary in all five UI languages; unreviewed ones show the original source text.
 
@@ -88,20 +91,20 @@ It lands in your Applications folder automatically. Open it once; if macOS 15 or
 ### Experience & Customization
 
 - **16 Visual Themes:** Switch between panel styles including Default, Matrix, Windows 95, Vintage Newspaper, Cloud Observation, Midnight Aquarium, Prism Arcade, Black Hole, World Cup 2026, Lepidoptera, Migration, Stained Glass, Origami, Sketchbook, Heart Monitor, and Catppuccin (official palette, all four flavors).
-- **Place the Panel Anywhere:** The panel is no longer pinned under the menu bar icon. Drag it from any empty spot to wherever you want it, and it reopens there next time. It stays put when another app takes focus — a second click on the menu bar icon, or Escape, closes it.
+- **Place the Panel Anywhere:** Drag the panel from any empty spot to wherever you want it, and it reopens there next time. It stays put when another app takes focus — a second click on the menu bar icon, or Escape, closes it.
 - **Drag to Reorder:** Grab any quota card and drag it up or down to swap the order — the arrangement is shared across every theme with quota cards (all except World Cup 2026) and survives restarts.
 - **Automatic Localization:** UI text is available in Traditional Chinese, Simplified Chinese, English, Japanese, and Korean, automatically matching your system settings.
 
 ## Privacy & Data Sources
 
-- Claude Code and Codex numbers are read **only from local log files** on your machine; reading them **never calls Anthropic or OpenAI's LLM APIs**.
-- Antigravity quota requires network access, and only if you use it: quota is fetched from Google's official quota endpoint using the OAuth credential the Antigravity CLI already stored after sign-in — read from macOS Keychain, Windows Credential Manager, or a local token file depending on CLI version. `usage` reads that credential without writing it back and keeps any refreshed access token in memory only; the call itself only reads quota metadata and never consumes your model quota.
+- Claude Code and Codex numbers are read from local log files on your machine.
+- Antigravity quota requires network access, and only if you use it: quota is fetched from Google's official quota endpoint using the OAuth credential the Antigravity CLI already stored after sign-in — read from macOS Keychain, Windows Credential Manager, or a local token file depending on CLI version. `usage` reads that credential without writing it back and keeps any refreshed access token in memory only; the call itself reads quota metadata.
 - Background network activity: the Antigravity quota/token endpoints above, public Claude and Codex status pages to flag outages, a public model-pricing table to estimate cost (falls back to built-in prices offline), and occasionally checking GitHub for a new version. Claude Code and Codex log contents are never uploaded.
 
 ## Requirements
 
 - macOS 12 (Monterey) or newer, or Windows 10/11
-- Claude Code, Codex, or Antigravity has been used at least once (so local usage data exists).
+- Claude Code, Codex, Antigravity, or Grok CLI has been used at least once (so local usage data exists).
 - (Source runs only) Python 3.13.
 
 ## Install
@@ -167,21 +170,22 @@ Privacy policy: this program will not transfer any information to other networke
 Switch between **16 visual themes** directly from the UI:
 
 <p align="center">
-  <img src="docs/classic.en.png" width="32%" alt="Classic theme" />
-  <img src="docs/matrix.en.png" width="32%" alt="Matrix theme" />
-  <img src="docs/win95.en.png" width="32%" alt="Windows 95 theme" />
-  <img src="docs/newspaper.en.png" width="32%" alt="Newspaper theme" />
-  <img src="docs/cloud_observation.en.png" width="32%" alt="Cloud Observation theme" />
-  <img src="docs/aquarium.en.png" width="32%" alt="Midnight Aquarium theme" />
-  <img src="docs/prism_arcade.en.png" width="32%" alt="Prism Arcade theme" />
-  <img src="docs/stained_glass.en.png" width="32%" alt="Stained Glass theme" />
-  <img src="docs/origami.en.png" width="32%" alt="Origami theme" />
-  <img src="docs/black_hole.en.png" width="32%" alt="Black Hole theme" />
-  <img src="docs/lepidoptera.en.png" width="32%" alt="Lepidoptera theme" />
-  <img src="docs/migration.en.png" width="32%" alt="Migration theme" />
-  <img src="docs/catppuccin.en.png" width="32%" alt="Catppuccin theme" />
-  <img src="docs/sketchbook.en.png" width="32%" alt="Sketchbook theme" />
-  <img src="docs/heart_monitor.en.png" width="32%" alt="Heart Monitor theme" />
+  <img src="docs/classic.en.png" width="24%" alt="Classic theme" />
+  <img src="docs/matrix.en.png" width="24%" alt="Matrix theme" />
+  <img src="docs/win95.en.png" width="24%" alt="Windows 95 theme" />
+  <img src="docs/newspaper.en.png" width="24%" alt="Newspaper theme" />
+  <img src="docs/cloud_observation.en.png" width="24%" alt="Cloud Observation theme" />
+  <img src="docs/aquarium.en.png" width="24%" alt="Midnight Aquarium theme" />
+  <img src="docs/prism_arcade.en.png" width="24%" alt="Prism Arcade theme" />
+  <img src="docs/stained_glass.en.png" width="24%" alt="Stained Glass theme" />
+  <img src="docs/origami.en.png" width="24%" alt="Origami theme" />
+  <img src="docs/black_hole.en.png" width="24%" alt="Black Hole theme" />
+  <img src="docs/world_cup.en.png" width="24%" alt="World Cup 2026 theme" />
+  <img src="docs/lepidoptera.en.png" width="24%" alt="Lepidoptera theme" />
+  <img src="docs/migration.en.png" width="24%" alt="Migration theme" />
+  <img src="docs/catppuccin.en.png" width="24%" alt="Catppuccin theme" />
+  <img src="docs/sketchbook.en.png" width="24%" alt="Sketchbook theme" />
+  <img src="docs/heart_monitor.en.png" width="24%" alt="Heart Monitor theme" />
 </p>
 
 ## Troubleshooting
@@ -215,14 +219,13 @@ If the menu bar shows `--`, it's usually not broken — there's just no local da
 | AI Update Daily | ✅ | — | — |
 | Progress Concierge & Token Saver | ✅ | — | — |
 | Token-waste Health Check | ✅ | — | — |
-| No LLM API calls to read quota | ✅ | ✅ | ✅ |
 | Open-source license | AGPL-3.0 | MIT | — |
 
 ## When usage Isn't the Right Fit
 
 - You only live in the terminal and don't want another menu bar icon running in the background — a one-off CLI check fits better.
-- You don't use Claude Code, Codex, or Antigravity — there's no local usage data for `usage` to read.
-- You're on Linux — only macOS and Windows are supported today.
+- You don't use Claude Code, Codex, Antigravity, or Grok CLI — there's no local usage data for `usage` to read.
+- You want a menu bar on Linux. Only macOS and Windows have one today, though the terminal interface (`uvx usage-cli`) runs on Linux.
 
 ## Development
 

{usage_cli-0.31.3 → usage_cli-0.31.5}/README.md

@@ -4,9 +4,9 @@
 
 # usage
 
-### Quota visibility for Claude Code, Codex, Antigravity, and Grok CLI, built into the macOS menu bar and Windows system tray.
+### Your Claude Code, Codex, Antigravity and Grok CLI quota, always on screen.
 
-Running out of quota mid-session is expensive — especially during a long refactor or debugging run that depends on Claude Code. `usage` surfaces 5-hour and weekly limits *before* you hit the wall, and keeps them visible the whole time. There's no command to run and no page to open; the answer is just there, where you already look.
+`usage` puts your 5-hour and weekly limits in the macOS menu bar or Windows system tray, colored from green to red. Hitting the limit halfway through a long refactor is a bad way to find out you were running low. Now you see it coming. There's nothing to run and no page to open.
 
 [繁體中文](docs/README.zh-TW.md) · [简体中文](docs/README.zh-CN.md) · English · [日本語](docs/README.ja.md) · [한국어](docs/README.ko.md)  |  [Discussions](https://github.com/aqua5230/usage/discussions)  |  [Landing page](https://aqua5230.github.io/usage/)
 
@@ -23,7 +23,9 @@ Running out of quota mid-session is expensive — especially during a long refac
   <img src="docs/showcase-v3.en.png" alt="usage — Claude Code, Codex, and Antigravity quota pinned to the macOS menu bar" width="820">
 </p>
 
-Claude Code and Codex numbers are read passively from log files already on your machine, so **watching your quota never calls Anthropic or OpenAI's LLM APIs** and never costs you a token. Antigravity is the one exception: its quota comes from Google's official quota endpoint using the sign-in the Antigravity CLI already stores locally — a metadata call that doesn't consume your model quota either.
+Claude Code and Codex numbers come from log files already on your machine. Antigravity quota comes from Google's official quota endpoint, using the sign-in the Antigravity CLI already stores locally.
+
+`usage` also helps you spend less. The status line warns you before your context window bloats or your prompt cache goes cold, and a Token Saver toggle keeps replies short. In an A/B test on real sessions, late replies stayed ~40% shorter instead of growing 84% longer.
 
 ## Quick Start
 
@@ -31,10 +33,10 @@ Claude Code and Codex numbers are read passively from log files already on your
 brew install --cask aqua5230/usage/usage
 ```
 
-**Not on macOS?** `uvx usage-cli` runs the terminal interface anywhere, Linux included — no install, no menu bar.
-
 It lands in your Applications folder automatically. Open it once; if macOS 15 or later blocks it, go to System Settings → Privacy & Security, scroll down, and click **Open Anyway**. On macOS 14 or earlier, right-click **Open** once to pass Gatekeeper. Then click the menu bar icon. Prefer a direct download or want the full setup flow? See [Install](#install) below.
 
+**Not on macOS?** `uvx usage-cli` runs the terminal interface anywhere, Linux included — no install, no menu bar.
+
 **Jump to:** [What You Get](#what-you-get) · [Privacy](#privacy--data-sources) · [Requirements](#requirements) · [Install](#install) · [Status Line](#first-launch-set-up-the-status-line) · [Windows](#windows-support) · [Themes](#theme-gallery) · [Troubleshooting](#troubleshooting) · [Comparison](#comparison) · [Not a Fit?](#when-usage-isnt-the-right-fit) · [Development](#development)
 
 ## What You Get
@@ -43,21 +45,22 @@ It lands in your Applications folder automatically. Open it once; if macOS 15 or
 
 - **Always-on Monitor:** Your quota lives in the menu bar, color-coded from green to red. Click when you want the full session, weekly, and per-project breakdown.
 - **Antigravity Support:** Antigravity (Gemini) session and weekly quota show up as a third card in every theme except World Cup 2026, which stays a two-team HUD. Numbers come straight from the official quota API, using the sign-in the Antigravity CLI already keeps on your machine — refreshed every few minutes, with live reset countdowns. Antigravity keeps two separate quota pools: the card shows Gemini by default, and tapping the `Gemini ⇄` tag next to the title switches it to Claude / GPT — the choice is remembered.
-- **Grok CLI Support:** A fourth card reads Grok CLI's weekly credit percentage straight from its own local debug log — no network call. Grok CLI doesn't expose session or burn-rate data, so the card shows a single weekly bar; its per-request token usage still counts toward today's cost and project totals like Claude Code and Codex.
+- **Grok CLI Support:** A fourth card reads Grok CLI's weekly credit percentage straight from its own local debug log. Grok CLI doesn't expose session or burn-rate data, so the card shows a single weekly bar; its per-request token usage still counts toward today's cost and project totals like Claude Code and Codex.
 - **Muse Code Spending:** Muse Code's per-request tokens and cost count toward today's cost, project totals, the HTML report, and the `usage` CLI, read from its own local session logs. Muse keeps no local quota data, so there is no Muse quota card.
-- **Service Status Alerts:** An orange-red banner appears when Claude Code, Claude API, or Codex API has an outage or degraded performance, read from their public Statuspage.io pages — never an LLM usage API. Antigravity isn't covered; it has no public status page.
+- **Service Status Alerts:** An orange-red banner appears when Claude Code, Claude API, or Codex API has an outage or degraded performance, read from their public Statuspage.io pages. Antigravity isn't covered; it has no public status page.
 - **Context Nudges & Notifications:** When your context window hits 70% — or earlier when it is filling fast — the status line nudges you to `/clear` or `/compact` to prevent token waste. You can also opt-in to system notifications for quota limits and recoveries.
-- **Prompt Cache Health:** The status line shows Claude Code's prompt cache hit rate with a countdown to expiry, so you can tell at a glance whether finishing now still reuses the cached context or lets it go cold and re-sends everything. Needs Claude Code 2.1.251 or newer; on older versions the segment simply doesn't appear.
+- **Prompt Cache Health:** The status line shows Claude Code's prompt cache hit rate with a countdown to expiry, so you can tell at a glance whether finishing now still reuses the cached context or lets it go cold and re-sends everything. Once it has gone cold, the countdown turns into how many tokens your next message will re-cache. Needs Claude Code 2.1.251 or newer; on older versions the segment simply doesn't appear.
 - **Hide Sections:** Only use one or two of the tools? Hide the Claude Code, Codex, Grok CLI, or Antigravity section from the menu bar and panels completely with a single click.
 
 ### Workflow Helpers
 
-- **Progress Concierge:** Open a new Claude Code session and `usage` hands your last progress straight to the AI, including your last request, uncommitted changes, and unfinished todos. No `/resume`, no recap. Fully local, off by default.
+- **Progress Concierge:** Open a new Claude Code session and `usage` hands your last progress straight to the AI, including your last request, uncommitted changes, and unfinished todos. No `/resume`, no recap. When you do `/resume` a conversation that sat long enough for its cache to expire, it warns you how many tokens the next message will re-send and suggests `/compact` first. Off by default.
 - **Token Saver:** A menu-bar toggle asks Claude Code and Codex to answer more tersely and in plainer language, saving output tokens while keeping code and error messages byte-exact. A light reminder keeps long conversations from drifting back to verbose — in an A/B test on real sessions, late replies stayed ~40% shorter instead of drifting 84% longer.
-- **Terminal Integration:** `usage status --json` hands your Claude Code and Codex quota to any tool that can run a command — Starship, tmux, or your own scripts. Reads the same local files as the menu bar, no network call. [Ready-made snippets](docs/DEVELOPMENT.md#quota-status-for-other-tools-usage-status).
+- **Auto-start 5-hour Session:** Off by default. Turn it on and, right after a 5-hour quota resets, `usage` sends one tiny message to each tool (Claude with Haiku, Antigravity with Gemini 3.5 Flash Low, Codex with its cheapest model) so the next 5-hour window starts counting right away. Those messages do use a little quota, but the amount is negligible. Checking your quota never sends a message; only this switch does.
+- **Terminal Integration:** `usage status --json` hands your Claude Code and Codex quota to any tool that can run a command — Starship, tmux, or your own scripts. Reads the same local files as the menu bar. [Ready-made snippets](docs/DEVELOPMENT.md#quota-status-for-other-tools-usage-status).
 - **Token-waste Health Check:** A daily background diagnosis scans your logs for waste, including repeated file reads, polluter directories, and noisy Bash output. If it finds issues, a one-line heads-up appears; say "show me" and the AI walks you through fixes.
 
-### AI Teamwork
+### Stay Current
 
 - **AI Update Daily:** Opens a daily-updated public [page](https://aqua5230.github.io/ai-updates/) covering Claude Code, Codex, and Antigravity, with the full history kept. Reviewed items get a plain-language summary in all five UI languages; unreviewed ones show the original source text.
 
@@ -68,20 +71,20 @@ It lands in your Applications folder automatically. Open it once; if macOS 15 or
 ### Experience & Customization
 
 - **16 Visual Themes:** Switch between panel styles including Default, Matrix, Windows 95, Vintage Newspaper, Cloud Observation, Midnight Aquarium, Prism Arcade, Black Hole, World Cup 2026, Lepidoptera, Migration, Stained Glass, Origami, Sketchbook, Heart Monitor, and Catppuccin (official palette, all four flavors).
-- **Place the Panel Anywhere:** The panel is no longer pinned under the menu bar icon. Drag it from any empty spot to wherever you want it, and it reopens there next time. It stays put when another app takes focus — a second click on the menu bar icon, or Escape, closes it.
+- **Place the Panel Anywhere:** Drag the panel from any empty spot to wherever you want it, and it reopens there next time. It stays put when another app takes focus — a second click on the menu bar icon, or Escape, closes it.
 - **Drag to Reorder:** Grab any quota card and drag it up or down to swap the order — the arrangement is shared across every theme with quota cards (all except World Cup 2026) and survives restarts.
 - **Automatic Localization:** UI text is available in Traditional Chinese, Simplified Chinese, English, Japanese, and Korean, automatically matching your system settings.
 
 ## Privacy & Data Sources
 
-- Claude Code and Codex numbers are read **only from local log files** on your machine; reading them **never calls Anthropic or OpenAI's LLM APIs**.
-- Antigravity quota requires network access, and only if you use it: quota is fetched from Google's official quota endpoint using the OAuth credential the Antigravity CLI already stored after sign-in — read from macOS Keychain, Windows Credential Manager, or a local token file depending on CLI version. `usage` reads that credential without writing it back and keeps any refreshed access token in memory only; the call itself only reads quota metadata and never consumes your model quota.
+- Claude Code and Codex numbers are read from local log files on your machine.
+- Antigravity quota requires network access, and only if you use it: quota is fetched from Google's official quota endpoint using the OAuth credential the Antigravity CLI already stored after sign-in — read from macOS Keychain, Windows Credential Manager, or a local token file depending on CLI version. `usage` reads that credential without writing it back and keeps any refreshed access token in memory only; the call itself reads quota metadata.
 - Background network activity: the Antigravity quota/token endpoints above, public Claude and Codex status pages to flag outages, a public model-pricing table to estimate cost (falls back to built-in prices offline), and occasionally checking GitHub for a new version. Claude Code and Codex log contents are never uploaded.
 
 ## Requirements
 
 - macOS 12 (Monterey) or newer, or Windows 10/11
-- Claude Code, Codex, or Antigravity has been used at least once (so local usage data exists).
+- Claude Code, Codex, Antigravity, or Grok CLI has been used at least once (so local usage data exists).
 - (Source runs only) Python 3.13.
 
 ## Install
@@ -147,21 +150,22 @@ Privacy policy: this program will not transfer any information to other networke
 Switch between **16 visual themes** directly from the UI:
 
 <p align="center">
-  <img src="docs/classic.en.png" width="32%" alt="Classic theme" />
-  <img src="docs/matrix.en.png" width="32%" alt="Matrix theme" />
-  <img src="docs/win95.en.png" width="32%" alt="Windows 95 theme" />
-  <img src="docs/newspaper.en.png" width="32%" alt="Newspaper theme" />
-  <img src="docs/cloud_observation.en.png" width="32%" alt="Cloud Observation theme" />
-  <img src="docs/aquarium.en.png" width="32%" alt="Midnight Aquarium theme" />
-  <img src="docs/prism_arcade.en.png" width="32%" alt="Prism Arcade theme" />
-  <img src="docs/stained_glass.en.png" width="32%" alt="Stained Glass theme" />
-  <img src="docs/origami.en.png" width="32%" alt="Origami theme" />
-  <img src="docs/black_hole.en.png" width="32%" alt="Black Hole theme" />
-  <img src="docs/lepidoptera.en.png" width="32%" alt="Lepidoptera theme" />
-  <img src="docs/migration.en.png" width="32%" alt="Migration theme" />
-  <img src="docs/catppuccin.en.png" width="32%" alt="Catppuccin theme" />
-  <img src="docs/sketchbook.en.png" width="32%" alt="Sketchbook theme" />
-  <img src="docs/heart_monitor.en.png" width="32%" alt="Heart Monitor theme" />
+  <img src="docs/classic.en.png" width="24%" alt="Classic theme" />
+  <img src="docs/matrix.en.png" width="24%" alt="Matrix theme" />
+  <img src="docs/win95.en.png" width="24%" alt="Windows 95 theme" />
+  <img src="docs/newspaper.en.png" width="24%" alt="Newspaper theme" />
+  <img src="docs/cloud_observation.en.png" width="24%" alt="Cloud Observation theme" />
+  <img src="docs/aquarium.en.png" width="24%" alt="Midnight Aquarium theme" />
+  <img src="docs/prism_arcade.en.png" width="24%" alt="Prism Arcade theme" />
+  <img src="docs/stained_glass.en.png" width="24%" alt="Stained Glass theme" />
+  <img src="docs/origami.en.png" width="24%" alt="Origami theme" />
+  <img src="docs/black_hole.en.png" width="24%" alt="Black Hole theme" />
+  <img src="docs/world_cup.en.png" width="24%" alt="World Cup 2026 theme" />
+  <img src="docs/lepidoptera.en.png" width="24%" alt="Lepidoptera theme" />
+  <img src="docs/migration.en.png" width="24%" alt="Migration theme" />
+  <img src="docs/catppuccin.en.png" width="24%" alt="Catppuccin theme" />
+  <img src="docs/sketchbook.en.png" width="24%" alt="Sketchbook theme" />
+  <img src="docs/heart_monitor.en.png" width="24%" alt="Heart Monitor theme" />
 </p>
 
 ## Troubleshooting
@@ -195,14 +199,13 @@ If the menu bar shows `--`, it's usually not broken — there's just no local da
 | AI Update Daily | ✅ | — | — |
 | Progress Concierge & Token Saver | ✅ | — | — |
 | Token-waste Health Check | ✅ | — | — |
-| No LLM API calls to read quota | ✅ | ✅ | ✅ |
 | Open-source license | AGPL-3.0 | MIT | — |
 
 ## When usage Isn't the Right Fit
 
 - You only live in the terminal and don't want another menu bar icon running in the background — a one-off CLI check fits better.
-- You don't use Claude Code, Codex, or Antigravity — there's no local usage data for `usage` to read.
-- You're on Linux — only macOS and Windows are supported today.
+- You don't use Claude Code, Codex, Antigravity, or Grok CLI — there's no local usage data for `usage` to read.
+- You want a menu bar on Linux. Only macOS and Windows have one today, though the terminal interface (`uvx usage-cli`) runs on Linux.
 
 ## Development
 

{usage_cli-0.31.3 → usage_cli-0.31.5}/adapters/claude.py

@@ -137,16 +137,23 @@ def parse_jsonl(
         _file_cache.move_to_end(path)
         parsed_entries = cached[2]
     else:
-        parsed_entries = []
+        by_key: dict[str, UsageEntry] = {}
         try:
             for data in iter_jsonl_dicts(path):
                 if data.get("type") != "assistant":
                     continue
 
-                parsed_entries.extend(_parse_assistant_entry(data, project))
+                for entry in _parse_assistant_entry(data, project):
+                    kept = by_key.setdefault(entry.dedup_key, entry)
+                    # Subagent transcripts log one request several times while it streams,
+                    # with output_tokens still growing, so keep the largest, not the first.
+                    # Without ids the key cannot tell requests apart, so never merge those.
+                    if entry.message_id or entry.request_id:
+                        kept.output_tokens = max(kept.output_tokens, entry.output_tokens)
         except (OSError, PermissionError, UnicodeDecodeError) as exc:
             _debug_file_error("failed to read Claude log", path, exc)
             return
+        parsed_entries = list(by_key.values())
 
         if path not in _file_cache and len(_file_cache) >= _FILE_CACHE_MAXSIZE:
             _file_cache.popitem(last=False)

{usage_cli-0.31.3 → usage_cli-0.31.5}/analyzer/reporter.py

@@ -43,7 +43,7 @@ _YEAR_WEEKS = 53
 YEAR_CACHE_PATH = Path(os.path.expanduser("~/.usage/year_cache.json"))
 YEAR_LEDGER_PATH = YEAR_CACHE_PATH.with_name("year_ledger.json")
 YEAR_CACHE_TTL_SECONDS = 6 * 3600
-_YEAR_CACHE_SCHEMA = 4
+_YEAR_CACHE_SCHEMA = 5
 _YEAR_LEDGER_SCHEMA = 1
 _YEAR_LEDGER_TRIM_BUFFER_DAYS = 60
 

{usage_cli-0.31.3 → usage_cli-0.31.5}/i18n.json

@@ -104,7 +104,7 @@
     "grok_stale_tooltip": "資料來自 Grok CLI 本機 log，不是即時 API；請最近執行過 grok 指令後再查看，才會更新。",
     "codex_stale_minutes": "約 {minutes} 分鐘前",
     "codex_stale_hours": "約 {hours} 小時前",
-    "codex_stale_tooltip": "Claude 有即時回報所以很準；Codex 沒有，usage 只能讀它本機偶爾留下的紀錄，可能比實際略舊。我們刻意不連網去補（連網會多燒你的 token）——繼續在終端機跑 Codex 就會自動更新。",
+    "codex_stale_tooltip": "Claude 有即時回報所以很準；Codex 沒有，usage 會讀它本機偶爾留下的紀錄，可能比實際略舊。繼續在終端機跑 Codex 就會自動更新。",
     "history_load_error_file": "讀不到使用紀錄檔",
     "history_load_error_parse": "使用紀錄格式異常",
     "history_load_error_unknown": "使用紀錄暫時無法讀取",
@@ -322,6 +322,7 @@
     "report_rw_inject_lead": "（這是進度交接。請在這次對話第一則回覆的最前面，先說一行「🐾 已接回上次進度，繼續吧！」，然後別照唸下面的線索，而是像聰明又貼心的搭檔那樣消化它：先判斷使用者最後真正在忙的是什麼（「最近在忙的」是新到舊排列，以最前面那筆為準，別被較舊的帶偏），用一兩句溫暖具體地說他上次做到哪、完成了什麼，再直接給出你判斷最該接著做的下一步——要具體，別反問「要做什麼」。只能根據下面線索講，沒有的別編；若線索太少看不出方向，就坦白說、並列出你看到的幾條線讓他挑。接著正常回應。）\n\n",
     "report_rw_empty": "（請在你這次對話的第一則回覆最前面，說一行「🐾 歡迎回來，這個專案目前沒有要接的進度。」，再正常回應。）",
     "report_rw_prompt": "專案：{project}（最後活動 {when}）\n最近在忙的（新→舊）：{last_request}\n完成的進度：{commits}\n未完成待辦：{todos}",
+    "report_rw_resume_cost": "這個對話放了 {idle}，快取已過期，下一句會重送約 {tokens} token。想省額度可以先打 /compact 壓縮再聊。",
     "report_rw_uncommitted": "上次離開時還留著：{branch} 分支有 {count} 個檔案改了還沒提交（{files}）",
     "report_rw_diagnosis_reminder": "健檢提醒：最近約 {waste_pct}% 浪費在{cause}。跟我說「修」，完整診斷在 {path}。",
     "report_rw_diagnosis_reminder_explain": "健檢提醒：最近約 {waste_pct}% 浪費在{cause}。跟我說「看」，我會讀 {path} 的完整診斷，解釋給你聽並給具體建議。",
@@ -531,7 +532,7 @@
     "grok_stale_tooltip": "Data comes from Grok CLI's local log, not a live API. Run a grok command recently for it to update.",
     "codex_stale_minutes": "about {minutes} minutes ago",
     "codex_stale_hours": "about {hours} hours ago",
-    "codex_stale_tooltip": "Claude is accurate because it reports in real time; Codex does not, so usage can only read the local records it leaves occasionally, which may be slightly older than reality. We intentionally do not go online to fill the gap (that would burn more of your tokens) — keep running Codex in the terminal and it will update automatically.",
+    "codex_stale_tooltip": "Claude is accurate because it reports in real time; Codex does not, so usage reads the local records it leaves occasionally, which may be slightly older than reality. Keep running Codex in the terminal and it will update automatically.",
     "history_load_error_file": "Couldn't read usage files",
     "history_load_error_parse": "Usage data format issue",
     "history_load_error_unknown": "Usage data temporarily unavailable",
@@ -749,6 +750,7 @@
     "report_rw_inject_lead": "(This is a resume handoff. At the very start of your first reply, lead with one line: \"🐾 Picked up where you left off — let's keep going!\", then, instead of reading the traces below aloud, digest them like a sharp, thoughtful partner: first work out what the user was actually in the middle of (\"Recently working on\" is newest-first, so trust the topmost item and don't get pulled back to older ones); in a sentence or two, warmly and concretely recap where they left off and what got done; then give the single concrete next step you would take — be specific, don't ask \"what should I do\". Use only what is below; invent nothing. If the traces are too thin to tell, say so plainly and list the threads you can see for them to pick. Then respond normally.)\n\n",
     "report_rw_empty": "(At the very start of your first reply in this session, say one line: \"🐾 Welcome back — nothing to pick up on this project yet.\", then respond normally.)",
     "report_rw_prompt": "Project: {project} (last active {when})\nRecently working on (newest first): {last_request}\nProgress made: {commits}\nOpen to-dos: {todos}",
+    "report_rw_resume_cost": "This conversation sat idle for {idle} and its prompt cache has expired, so your next message re-sends about {tokens} tokens. Run /compact first to use less quota.",
     "report_rw_uncommitted": "Left uncommitted last time: {count} changed file(s) on branch {branch} ({files})",
     "report_rw_diagnosis_reminder": "Health check: about {waste_pct}% waste came from {cause}. Say \"fix it\" and I'll read the full diagnosis at {path}.",
     "report_rw_diagnosis_reminder_explain": "Health check: about {waste_pct}% waste came from {cause}. Say \"show me\" and I'll walk you through the full diagnosis at {path}.",
@@ -958,7 +960,7 @@
     "grok_stale_tooltip": "资料来自 Grok CLI 本地日志，不是即时 API；请最近运行过 grok 指令后再查看，才会更新。",
     "codex_stale_minutes": "约 {minutes} 分钟前",
     "codex_stale_hours": "约 {hours} 小时前",
-    "codex_stale_tooltip": "Claude 有实时回报所以很准；Codex 没有，usage 只能读取它本机偶尔留下的记录，可能比实际略旧。我们刻意不联网去补（联网会多消耗你的 token）——继续在终端运行 Codex 就会自动更新。",
+    "codex_stale_tooltip": "Claude 有实时回报所以很准；Codex 没有，usage 会读取它本机偶尔留下的记录，可能比实际略旧。继续在终端运行 Codex 就会自动更新。",
     "history_load_error_file": "读不到使用记录文件",
     "history_load_error_parse": "使用记录格式异常",
     "history_load_error_unknown": "使用记录暂时无法读取",
@@ -1176,6 +1178,7 @@
     "report_rw_inject_lead": "（这是进度交接。请在这次对话第一则回复的最前面，先说一行「🐾 已接回上次进度，继续吧！」，然后别照念下面的线索，而是像聪明又贴心的搭档那样消化它：先判断用户最后真正在忙的是什么（「最近在忙的」是新到旧排列，以最前面那笔为准，别被较旧的带偏），用一两句温暖具体地说他上次做到哪、完成了什么，再直接给出你判断最该接着做的下一步——要具体，别反问「要做什么」。只能根据下面线索讲，没有的别编；若线索太少看不出方向，就坦白说、并列出你看到的几条线让他挑。接着正常回应。）\n\n",
     "report_rw_empty": "（请在你这次对话的第一则回复最前面，说一行「🐾 欢迎回来，这个项目目前没有要接的进度。」，再正常回应。）",
     "report_rw_prompt": "项目：{project}（最后活动 {when}）\n最近在忙的（新→旧）：{last_request}\n完成的进度：{commits}\n未完成待办：{todos}",
+    "report_rw_resume_cost": "这个对话放了 {idle}，缓存已过期，下一句会重新发送约 {tokens} token。想省额度可以先输入 /compact 压缩再聊。",
     "report_rw_uncommitted": "上次离开时还留着：{branch} 分支有 {count} 个文件改了还没提交（{files}）",
     "report_rw_diagnosis_reminder": "健检提醒：最近约有 {waste_pct}% 的浪费来自{cause}。跟我说「修」，完整诊断在 {path}。",
     "report_rw_diagnosis_reminder_explain": "健检提醒：最近约有 {waste_pct}% 的浪费来自{cause}。跟我说「看」，我会读 {path} 的完整诊断，解释给你听并给具体建议。",
@@ -1385,7 +1388,7 @@
     "grok_stale_tooltip": "データは Grok CLI のローカルログから読み取るもので、リアルタイム API ではありません。更新するには最近 grok コマンドを実行してください。",
     "codex_stale_minutes": "約 {minutes} 分前",
     "codex_stale_hours": "約 {hours} 時間前",
-    "codex_stale_tooltip": "Claude はリアルタイムに報告するので正確です。Codex にはそれがないため、usage は Codex がローカルに時々残す記録だけを読み取り、実際より少し古い場合があります。補うために意図的にネット接続はしません（接続するとあなたの token を余分に消費します）——ターミナルで Codex を使い続けると自動的に更新されます。",
+    "codex_stale_tooltip": "Claude はリアルタイムに報告するので正確です。Codex にはそれがないため、usage は Codex がローカルに時々残す記録を読み取り、実際より少し古い場合があります。ターミナルで Codex を使い続けると自動的に更新されます。",
     "history_load_error_file": "使用履歴ファイルを読み込めません",
     "history_load_error_parse": "使用履歴データの形式に問題があります",
     "history_load_error_unknown": "使用履歴データを一時的に取得できません",
@@ -1603,6 +1606,7 @@
     "report_rw_inject_lead": "（これは進捗の引き継ぎです。最初の返信の冒頭で、まず一行「🐾 前回の続きを引き継ぎました！そのまま続けましょう！」と述べ、下記の手がかりをそのまま読み上げるのではなく、賢く気の利いた相棒のように消化してください：まずユーザーが最後に実際に取り組んでいたことを見極め（「最近の作業」は新しい順なので、先頭の項目を信頼し、古いものに引きずられないこと）、前回どこまで進んだか・何が完了したかを一、二文で温かく具体的に振り返り、次に取るべき具体的な一歩を提示してください——具体的に述べ、「何をしますか」と聞き返さないこと。下記にある情報だけを使い、無いことは創作しないこと。手がかりが乏しくて判断できない場合は正直にそう述べ、見えるスレッドを挙げて選んでもらうこと。その後、通常どおり応答してください。）\n\n",
     "report_rw_empty": "（このセッションの最初の返信の冒頭に、一行「🐾 おかえりなさい！このプロジェクトはまだ引き継ぐ進捗がありません。」と述べてから、通常どおり応答してください。）",
     "report_rw_prompt": "プロジェクト：{project}（最終アクティブ {when}）\n最近の作業（新しい順）：{last_request}\n完了した進捗：{commits}\n未完了のToDo：{todos}",
+    "report_rw_resume_cost": "この会話は {idle} 放置され、プロンプトキャッシュが切れています。次のメッセージで約 {tokens} トークンを再送信します。クォータを節約するには、先に /compact で圧縮してください。",
     "report_rw_uncommitted": "前回の終了時に未コミット：{branch} ブランチに変更済み未コミットのファイルが {count} 件（{files}）",
     "report_rw_diagnosis_reminder": "健診メモ: 直近では約 {waste_pct}% の無駄が{cause}から出ています。「直して」と言ってもらえれば、{path} の完全な診断を読んで対応します。",
     "report_rw_diagnosis_reminder_explain": "健診メモ: 直近では約 {waste_pct}% の無駄が{cause}から出ています。「見せて」と言ってもらえれば、{path} の完全な診断を読み解いて説明と対策をお伝えします。",
@@ -1812,7 +1816,7 @@
     "grok_stale_tooltip": "데이터는 Grok CLI의 로컬 로그에서 읽으며 실시간 API가 아닙니다. 업데이트하려면 최근에 grok 명령을 실행하세요.",
     "codex_stale_minutes": "약 {minutes}분 전",
     "codex_stale_hours": "약 {hours}시간 전",
-    "codex_stale_tooltip": "Claude는 실시간으로 보고하므로 정확합니다. Codex는 그렇지 않아 usage가 Codex가 로컬에 가끔 남기는 기록만 읽을 수 있고, 실제보다 조금 오래된 값일 수 있습니다. 우리는 이를 보완하려고 일부러 온라인 연결을 하지 않습니다(연결하면 token을 더 소비하게 됩니다)——터미널에서 Codex를 계속 실행하면 자동으로 업데이트됩니다.",
+    "codex_stale_tooltip": "Claude는 실시간으로 보고하므로 정확합니다. Codex는 그렇지 않아 usage가 Codex가 로컬에 가끔 남기는 기록을 읽으며, 실제보다 조금 오래된 값일 수 있습니다. 터미널에서 Codex를 계속 실행하면 자동으로 업데이트됩니다.",
     "history_load_error_file": "사용 기록 파일을 읽을 수 없습니다",
     "history_load_error_parse": "사용 기록 형식에 문제가 있습니다",
     "history_load_error_unknown": "사용 기록을 일시적으로 불러올 수 없습니다",
@@ -2030,6 +2034,7 @@
     "report_rw_inject_lead": "(이것은 진행 상황 인수인계입니다. 첫 답변 맨 앞에 먼저 한 줄 \"🐾 지난 작업을 불러왔어요! 이어서 가볼까요?\"라고 말한 뒤, 아래 단서를 그대로 읽지 말고 똑똑하고 사려 깊은 동료처럼 소화하세요: 먼저 사용자가 마지막에 실제로 무엇을 하고 있었는지 파악하고(\"최근 작업한 내용\"은 최신순이므로 맨 위 항목을 신뢰하고 오래된 것에 끌려가지 마세요), 지난번에 어디까지 했고 무엇을 완료했는지 한두 문장으로 따뜻하고 구체적으로 짚어 준 뒤, 이어서 취해야 할 구체적인 다음 단계 하나를 제시하세요 — 구체적으로 말하고 \"무엇을 할까요\"라고 되묻지 마세요. 아래 있는 내용만 사용하고 없는 것은 지어내지 마세요. 단서가 너무 적어 판단하기 어려우면 솔직히 그렇게 말하고 보이는 갈래들을 나열해 고르게 하세요. 그런 다음 평소대로 응답하세요.)\n\n",
     "report_rw_empty": "(이 세션의 첫 답변 맨 앞에 한 줄 \"🐾 돌아오셨네요! 이 프로젝트는 아직 이어갈 내용이 없어요.\"라고 말한 뒤 평소대로 응답하세요.)",
     "report_rw_prompt": "프로젝트: {project} (마지막 활동 {when})\n최근 작업한 내용 (최신순): {last_request}\n완료한 진행: {commits}\n미완료 할 일: {todos}",
+    "report_rw_resume_cost": "이 대화는 {idle} 동안 방치되어 프롬프트 캐시가 만료되었습니다. 다음 메시지에서 약 {tokens} 토큰을 다시 보냅니다. 할당량을 아끼려면 먼저 /compact로 압축하세요.",
     "report_rw_uncommitted": "지난번 종료 시 미커밋 파일: {branch} 브랜치에 {count}개 파일이 커밋되지 않은 채 남아 있었습니다 ({files})",
     "report_rw_diagnosis_reminder": "진단 알림: 최근 낭비 약 {waste_pct}%의 주된 원인은 {cause}입니다. \"고쳐줘\"라고 말하면 {path}의 전체 진단을 읽고 바로 처리할게요.",
     "report_rw_diagnosis_reminder_explain": "진단 알림: 최근 낭비 약 {waste_pct}%의 주된 원인은 {cause}입니다. \"보여줘\"라고 말하면 {path}의 전체 진단을 읽고 자세히 설명해 드릴게요.",

{usage_cli-0.31.3 → usage_cli-0.31.5}/installer/session_hooks.py

@@ -59,8 +59,8 @@ CODEX_CONFIG = setup_hook.CODEX_CONFIG
 # session. Off by default: enabled only via the menu toggle, never by self_heal.
 RESUME_HOOK_TARGET = Path(os.path.expanduser("~/.claude/usage-session-resume.py"))
 RESUME_PROMPT_SIDECAR = Path(os.path.expanduser("~/.claude/usage-resume-prompt.json"))
-RESUME_HOOK_VERSION = "1.9"
-RESUME_MATCHER = "startup|clear"
+RESUME_HOOK_VERSION = "1.10"
+RESUME_MATCHER = "startup|clear|resume"
 RESUME_LANGS = ("zh-TW", "zh-CN", "en", "ja", "ko")
 _RESUME_MARKER = "usage-session-resume"
 _RESUME_MARKERS = (_RESUME_MARKER, "usage_session_resume")
@@ -299,6 +299,7 @@ def _write_resume_sidecar() -> None:
         lead = table.get("report_rw_inject_lead") or en.get("report_rw_inject_lead") or ""
         empty = table.get("report_rw_empty") or en.get("report_rw_empty") or ""
         uncommitted = table.get("report_rw_uncommitted") or en.get("report_rw_uncommitted") or ""
+        resume_cost = table.get("report_rw_resume_cost") or en.get("report_rw_resume_cost") or ""
         diagnosis_reminder = (
             table.get("report_rw_diagnosis_reminder")
             or en.get("report_rw_diagnosis_reminder")
@@ -329,6 +330,7 @@ def _write_resume_sidecar() -> None:
                 "lead": lead,
                 "empty": empty,
                 "uncommitted": uncommitted,
+                "resume_cost": resume_cost,
                 "diagnosis_reminder": diagnosis_reminder,
                 "diagnosis_reminder_explain": diagnosis_reminder_explain,
                 "diagnosis_default_cause": diagnosis_default_cause,
@@ -837,6 +839,7 @@ def _self_heal_resume() -> None:
     script/sidecar and update a stale script. Never enables it on its own."""
     if not is_resume_enabled():
         return
+    _migrate_resume_matcher_if_needed()
     _migrate_resume_command_if_needed()
     missing = _missing_resume_artifacts()
     if missing:
@@ -919,6 +922,41 @@ def _self_heal_terse_reminder() -> None:
         )
 
 
+def _migrate_resume_matcher_if_needed() -> None:
+    settings = _load_settings()
+    entries = _session_start_list(settings)
+    if not entries:
+        return
+    changed = False
+    updated: list[Any] = []
+    for entry in entries:
+        if not isinstance(entry, dict) or not _is_resume_entry(entry):
+            updated.append(entry)
+            continue
+        if entry.get("matcher") == RESUME_MATCHER:
+            updated.append(entry)
+            continue
+        hooks = entry.get("hooks")
+        if not isinstance(hooks, list):
+            updated.append(entry)
+            continue
+        ours = [
+            hook
+            for hook in hooks
+            if isinstance(hook, dict)
+            and isinstance(hook.get("command"), str)
+            and _command_owns_script(hook["command"], _RESUME_MARKERS)
+        ]
+        others = [hook for hook in hooks if hook not in ours]
+        if others:
+            updated.append({**entry, "hooks": others})
+        updated.append({**entry, "matcher": RESUME_MATCHER, "hooks": ours})
+        changed = True
+    if changed:
+        settings["hooks"]["SessionStart"] = updated
+        _save_settings(settings)
+
+
 def _migrate_resume_command_if_needed() -> None:
     settings = _load_settings()
     entries = _session_start_list(settings)

{usage_cli-0.31.3 → usage_cli-0.31.5}/installer/setup_hook.py

@@ -86,7 +86,7 @@ PREV_SL_KEY = "previousStatusLine"
 # Where the status-line switch parks the live statusLine while it is off, so it
 # never overwrites the pre-install backup under PREV_SL_KEY.
 DISABLED_SL_KEY = "disabledStatusLine"
-HOOK_VERSION = "1.6"
+HOOK_VERSION = "1.8"
 FORWARDER_VERSION = "1.1"
 # Antigravity's Go runner hands its status-line command to cmd.exe unquoted, so
 # every character cmd.exe treats specially has to be kept out of the path.

{usage_cli-0.31.3 → usage_cli-0.31.5}/loaders/history_loader.py

@@ -15,7 +15,7 @@ import sys
 import time
 from collections import OrderedDict
 from collections.abc import Iterable
-from dataclasses import dataclass
+from dataclasses import dataclass, replace
 from datetime import UTC, datetime, timedelta
 from pathlib import Path
 from typing import Any
@@ -55,7 +55,7 @@ class _FileCacheEntry:
 _file_cache: OrderedDict[Path, _FileCacheEntry] = OrderedDict()
 
 HISTORY_CACHE_PATH = Path(os.path.expanduser("~/.usage/history_jsonl_cache.json"))
-_HISTORY_JSONL_CACHE_SCHEMA = 4
+_HISTORY_JSONL_CACHE_SCHEMA = 5
 _disk_cache_seeded = False
 _DISK_CACHE_FLUSH_INTERVAL_S = 300.0
 _disk_cache_dirty = False
@@ -319,7 +319,7 @@ def _parse_complete_lines(
     digest: Any,
     confirmed_offset: int,
 ) -> int:
-    seen = {_dedup_key(entry) for entry in parsed_entries}
+    index = {_dedup_key(entry): i for i, entry in enumerate(parsed_entries)}
     while True:
         line_start = int(file.tell())
         line, too_long = read_bounded_jsonl_line(file)
@@ -337,9 +337,15 @@ def _parse_complete_lines(
         if parsed_entries_for_line is not None:
             for entry in parsed_entries_for_line:
                 dedup_key = _dedup_key(entry)
-                if dedup_key in seen:
+                kept_at = index.get(dedup_key)
+                if kept_at is not None:
+                    # Subagent transcripts log one request several times while it streams,
+                    # with output_tokens still growing, so keep the largest, not the first.
+                    kept = parsed_entries[kept_at]
+                    if entry.output_tokens > kept.output_tokens:
+                        parsed_entries[kept_at] = replace(kept, output_tokens=entry.output_tokens)
                     continue
-                seen.add(dedup_key)
+                index[dedup_key] = len(parsed_entries)
                 parsed_entries.append(entry)
 
 

{usage_cli-0.31.3 → usage_cli-0.31.5}/menubar/app.py

@@ -960,7 +960,7 @@ class AppDelegate(NSObject):
                     reset=row.reset_text,
                 )
             )
-            content.setSound_(sound_cls.defaultSound())
+            content.setSound_(menubar_notify.notification_sound(sound_cls, event.kind))
             request = request_cls.requestWithIdentifier_content_trigger_(
                 f"usage.{event.channel}.{event.kind}.{int(time.time() * 1000)}",
                 content,