<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Shubham</title>
    <description>The latest articles on DEV Community by Shubham (@ssv8991).</description>
    <link>https://dev.to/ssv8991</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4244890%2F38b60896-4d58-4c3b-90bf-1b1b49a143f9.png</url>
      <title>DEV Community: Shubham</title>
      <link>https://dev.to/ssv8991</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/ssv8991"/>
    <language>en</language>
    <item>
      <title>I kept running out of my Claude Code weekly limit, so I built a tool to see which model used it</title>
      <dc:creator>Shubham</dc:creator>
      <pubDate>Sat, 14 Nov 2026 18:12:12 +0000</pubDate>
      <link>https://dev.to/ssv8991/i-kept-running-out-of-my-claude-code-weekly-limit-so-i-built-a-tool-to-see-which-model-used-it-52af</link>
      <guid>https://dev.to/ssv8991/i-kept-running-out-of-my-claude-code-weekly-limit-so-i-built-a-tool-to-see-which-model-used-it-52af</guid>
      <description>&lt;p&gt;If you use Claude Code a lot, you've probably seen this: the weekly limit bar creeping toward 100%, and no idea &lt;strong&gt;which model&lt;/strong&gt; is eating it. Was it Opus on that big refactor? Sonnet in the background? Subagents?&lt;/p&gt;

&lt;p&gt;Claude Code tells you &lt;em&gt;how much&lt;/em&gt; of your weekly limit you've used. It doesn't tell you &lt;em&gt;which models&lt;/em&gt; used it. So I built a small tool for that.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2F3n6qhfpjnfzdsaqn49m3.gif" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2F3n6qhfpjnfzdsaqn49m3.gif" alt="model-share pane refreshing" width="640" height="601"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  What it does
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;model-share&lt;/strong&gt; is a Claude Code mod that adds a pane next to the chat:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Each model's share of your week.&lt;/strong&gt; Everything you used this week counts as 100%, and each model gets its slice (mine right now: Fable 37%, Sonnet 34%, Opus 29%).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Your weekly limit, with pace.&lt;/strong&gt; The bar turns green, amber or red depending on whether you're using it faster than the week is passing, and tells you when you'll hit 100% at this rate.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Follows your reset.&lt;/strong&gt; The window matches your plan's weekly reset and starts over with it.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Auto refresh.&lt;/strong&gt; Every 1, 3, 5 or 10 minutes, or off.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;It works in the terminal, the Claude desktop app and the VS Code extension.&lt;/p&gt;

&lt;h2&gt;
  
  
  How it works
&lt;/h2&gt;

&lt;p&gt;Claude Code already saves every session as a file under &lt;code&gt;~/.claude/projects/&lt;/code&gt;. Each reply in those files records which model wrote it and how many tokens it used.&lt;/p&gt;

&lt;p&gt;model-share runs a small bundled Python script that:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Reads this week's session files (all projects, subagents included)&lt;/li&gt;
&lt;li&gt;Counts each reply once (the same reply can appear more than once in a file)&lt;/li&gt;
&lt;li&gt;Adds up tokens per model, with cache reads weighted at 1/10, since they cost far less&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;The weekly window and limit percentage come from Claude Code itself.&lt;/p&gt;

&lt;p&gt;A recount takes about 0.2 seconds, and because it only reads local files:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;It uses zero tokens.&lt;/strong&gt; No model calls.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Nothing leaves your machine.&lt;/strong&gt; No network requests at all.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Install
&lt;/h2&gt;

&lt;p&gt;In a Claude Code terminal session:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;/plugin install model-share --marketplace shubhamverma8991/claude-model-share
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Answer &lt;code&gt;y&lt;/code&gt;, press Enter, and the pane opens. Type &lt;code&gt;/model-share&lt;/code&gt; any time to bring it back, or &lt;code&gt;/model-share every 5m&lt;/code&gt; to change the refresh interval.&lt;/p&gt;

&lt;h2&gt;
  
  
  Limitations (being honest)
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Claude Code only.&lt;/strong&gt; Chats on claude.ai or the mobile app aren't stored on your computer, so they aren't counted.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Shares are by tokens, not by limit.&lt;/strong&gt; Claude Code doesn't report how much of the limit each model used, so this is an estimate. If a model counts more heavily against your limit per token, its true share is a bit higher.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  What I learned building it
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Claude Code's mod system lets you add real UI (panes, status lines, commands) without forking anything.&lt;/li&gt;
&lt;li&gt;Session files can contain the same reply more than once, so deduplicating by message ID mattered a lot for accurate numbers.&lt;/li&gt;
&lt;li&gt;Different clients get different data: the VS Code extension doesn't receive the limit percentage, so the mod remembers the last value any session saw.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Try it
&lt;/h2&gt;

&lt;p&gt;It's free and MIT-licensed:&lt;br&gt;
👉 &lt;strong&gt;&lt;a href="https://github.com/shubhamverma8991/claude-model-share" rel="noopener noreferrer"&gt;github.com/shubhamverma8991/claude-model-share&lt;/a&gt;&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;If it's useful, a ⭐ on GitHub helps a lot. And tell me in the comments: &lt;strong&gt;what else would you want it to show?&lt;/strong&gt; I'm thinking about daily breakdowns and week-over-week comparison next.&lt;/p&gt;

</description>
      <category>claude</category>
      <category>ai</category>
      <category>opensource</category>
      <category>productivity</category>
    </item>
  </channel>
</rss>
