Claude Context Window & Usage Tracker — Recall

Tracking since May 11, 2026.

Changelog

• Jun 5, 2026
  description

  Token Monitor gives Claude.ai users real-time, at-a-glance visibility into 
  every dimension of their AI usage — so you never get blindsided by a 
  cut-off reply or an unexpected quota limit.
  
  ─────────────────────────────────────────────────────────
  🔍  WHAT YOU GET
  ─────────────────────────────────────────────────────────
  
  • CONTEXT WINDOW GAUGE
    See how full the current conversation is — as a percentage and as a 
    token count. Know exactly when you're approaching the model's memory 
    limit before your next message is sent.
  
  • 5-HOUR & WEEKLY QUOTA BARS
    Claude Pro and Max plans throttle usage over a 5-hour rolling window 
    and a weekly budget. Token Monitor shows both bars in real time and 
    estimates when each will reset.
  
  • TRUNCATION RISK WARNING  ⚠
    Before you hit Send, Token Monitor calculates whether your next 
    message — combined with the predicted output — will overflow the 
    context window. An inline banner warns you with an actionable 
    suggestion: "Split into smaller questions" or "Start a new chat."
  
  • OUTPUT SIZE PREDICTION  📊
    A small pill next to the Send button predicts whether the reply will 
    be Small, Medium, Large, or XL — so you can plan accordingly.
  
  • PER-TURN TOKEN COST
    Token cost badges appear beneath every user message, showing input 
    and output token counts for that turn.
  
  • STREAMING AWARENESS
    While Claude is generating, Token Monitor tracks tokens already 
    committed to the input and tokens streamed back in real time.
  
  • SELF-CALIBRATING ESTIMATES
    Token counts are estimated via a fast heuristic (no API calls). 
    When Claude shows "X messages left" banners, Token Monitor uses those 
    signals to self-correct its estimates over time — getting more 
    accurate the more you use it.
  
  ─────────────────────────────────────────────────────────
  🔒  PRIVACY & PERMISSIONS
  ─────────────────────────────────────────────────────────
  
  Token Monitor is built with a privacy-first design:
  
  • All processing happens locally in your browser. Your conversations 
    are never transmitted to any external server.
  • The extension only runs on claude.ai domains.
  • Quota data is fetched directly from Claude's own settings endpoint 
    (the same data shown in your account dashboard) — no third-party 
    services are involved.
  • No account creation required. No login. No analytics.
  
  Permissions used:
    storage     — saves your preferences and calibration data locally
    alarms      — schedules periodic quota refreshes
    notifications — optional limit-reached desktop alerts
    scripting / tabs — required to inject the UI into claude.ai
  
  ─────────────────────────────────────────────────────────
  ⚙️  SUPPORTED PLANS & MODELS
  ─────────────────────────────────────────────────────────
  
  Works with:
    • Claude Free, Pro, Max (5x and 20x), Team, and Enterprise accounts
    • All Claude models available on claude.ai: Sonnet, Opus, Haiku
    • Conversations inside Projects (project knowledge token cost included)
  
  ─────────────────────────────────────────────────────────
  🌐  LANGUAGES
  ─────────────────────────────────────────────────────────
  
  UI available in English and 简体中文 (Simplified Chinese).
  
  ─────────────────────────────────────────────────────────
  📌  NOTES
  ─────────────────────────────────────────────────────────
  
  • Anthropic occasionally changes the claude.ai DOM structure. If an 
    indicator stops updating, check for an extension update.
  • Token estimates use a heuristic (not the exact tokenizer). The 
    self-calibration loop corrects drift over time.
  • This extension is an independent tool — it is not affiliated with or 
    endorsed by Anthropic.

  Recall gives you complete visibility into your Claude usage, context window health, and conversation risk — before quality drops, responses get truncated, or limits get in your way.
  
  ─────────────────────────────────
  WHY RECALL EXISTS
  ─────────────────────────────────
  
  Claude doesn’t tell you when your conversation is approaching the edge.
  
  You’re deep in a critical discussion when:
  
  • Responses suddenly become shorter
  
  • Important details start disappearing
  
  • Context gets silently compressed
  
  • You hit a usage limit with no warning
  
  • Opus burns through your quota faster than expected
  
  Most users only realize there’s a problem after it happens.
  
  Recall helps you see it before it does.
  
  ─────────────────────────────────
  WHAT RECALL DOES
  ─────────────────────────────────
  
  LIVE USAGE TRACKING
  
  Monitor your Claude usage in real time.
  
  Track:
  
  • 5-hour usage window
  
  • Weekly usage limits
  
  • Remaining budget
  
  • Reset times
  
  Supports Claude Free, Pro, Max 5x, Max 20x, Team, and Enterprise.
  
  ─────────────────────────────────
  
  CONTEXT WINDOW MONITORING
  
  See exactly how full your current conversation is.
  
  Recall continuously estimates context usage and alerts you before context collapse begins to affect response quality.
  
  Know when it’s time to summarize, hand off, or start fresh.
  
  ─────────────────────────────────
  
  TRUNCATION RISK WARNINGS
  
  Before sending a large prompt, Recall analyzes the request and warns when it is likely to generate a truncated response.
  
  Avoid wasting messages on prompts that are too large, too complex, or too expensive.
  
  ─────────────────────────────────
  
  PER-TURN TOKEN INSIGHTS
  
  Understand the true cost of every conversation.
  
  See:
  
  • Input tokens
  
  • Output tokens
  
  • Conversation growth
  
  • Estimated quota impact
  
  Learn how different models, attachments, projects, and tools affect your usage.
  
  ─────────────────────────────────
  
  SMART CONTEXT HAND-OFF
  
  When a conversation approaches the context limit, Recall suggests the optimal time to summarize and continue in a new chat.
  
  Preserve important information without losing momentum.
  
  ─────────────────────────────────
  
  OUTPUT SIZE PREDICTION
  
  Estimate response size before you send.
  
  Recall predicts whether a prompt is likely to generate a Small, Medium, Large, or XL response based on your usage patterns and conversation state.
  
  ─────────────────────────────────
  BUILT FOR EVERY CLAUDE PLAN
  ─────────────────────────────────
  
  ✓ Claude Free
  
  ✓ Claude Pro
  
  ✓ Claude Max 5x
  
  ✓ Claude Max 20x
  
  ✓ Claude Team
  
  ✓ Claude Enterprise
  
  ─────────────────────────────────
  FREQUENTLY ASKED QUESTIONS
  ─────────────────────────────────
  
  Does Recall work with Claude Pro?
  
  Yes. Recall supports all Claude plans and automatically adapts to your available limits.
  
  ─────────────────────────────────
  
  Will Recall slow down Claude?
  
  No. Recall runs locally in your browser using lightweight DOM observation and background processing.
  
  No impact on Claude performance or response speed.
  
  ─────────────────────────────────
  
  Does Recall send my chats anywhere?
  
  No.
  
  Everything runs locally on your device.
  
  No servers.
  
  No account.
  
  No tracking.
  
  ─────────────────────────────────
  
  How accurate are the estimates?
  
  Typically within 3–5% of actual usage.
  
  Recall continuously calibrates its estimates based on observed usage patterns to improve accuracy over time.
  
  ─────────────────────────────────
  
  Does Recall support Claude Projects?
  
  Yes.
  
  Project knowledge tokens are tracked separately so you can understand how project context affects overall usage.
  
  ─────────────────────────────────
  PRIVACY FIRST
  ─────────────────────────────────
  
  100% local processing.
  
  No data collection.
  
  No analytics.
  
  No external servers.
  
  No account required.
  
  ─────────────────────────────────
  
  Never lose context again.
  
  Recall gives Claude users the visibility Anthropic doesn’t.
  ─────────────────────────────────
  
  Stop flying blind. Know your Claude quota before it knows you.

• Jun 5, 2026
  name

  Token Monitor — AI Context Tracker

  Claude Context Window & Usage Tracker — Recall

• Jun 5, 2026
  host_permissions

  *://claude.ai/*, *://*.claude.ai/*, https://api.web3forms.com/*

  *://claude.ai/*, *://*.claude.ai/*, *://status.anthropic.com/*, https://raw.githubusercontent.com/*

• Jun 5, 2026
  permissions

  storage, alarms, notifications, scripting, tabs

  storage, alarms, notifications, scripting, tabs, downloads

• May 16, 2026
  description

  Token Monitor is a lightweight floating meter that sits next to your AI 
  chat window and tells you, in real time, how much of the model's context 
  window you've already used.
  
  Why it matters
  
  Long AI conversations eventually fill up the model's context window. When 
  that happens, your replies start getting truncated, the assistant 
  "forgets" earlier turns, or it refuses to answer — usually with no clear 
  warning. Token Monitor watches the page for you and shows the running 
  total before truncation happens.
  
  What you see
  
  - A live percentage of the context window already used
  - A color-coded progress bar that changes from green to yellow to red
  - Estimated tokens remaining and how many characters you can still type
  - A prediction of whether the assistant still has room to answer in full
  - An optional audio alert when you cross a threshold you configure
  
  Privacy first
  
  Everything runs locally in your browser. The extension reads conversation 
  text from the page only to estimate token count, and that text never 
  leaves your device. No analytics, no remote servers, no third-party 
  scripts. Full privacy policy is linked below.
  
  Customizable
  
  You can drag the floating widget anywhere on the page or minimize it to a 
  small pill. Threshold sliders, four sound styles, volume, and the 
  display language (English or Simplified Chinese, with system auto-detect) 
  are all adjustable from the popup.
  
  Honest about estimates
  
  Token counts are estimated using a character-class heuristic, not exact 
  API counts. Expect roughly 15 percent variance. For exact billing 
  numbers, use each provider's API directly.
  
  Compatibility
  
  Works on the major AI chat sites listed in the manifest's host 
  permissions. The current set is shown on the project page.

  Token Monitor gives Claude.ai users real-time, at-a-glance visibility into 
  every dimension of their AI usage — so you never get blindsided by a 
  cut-off reply or an unexpected quota limit.
  
  ─────────────────────────────────────────────────────────
  🔍  WHAT YOU GET
  ─────────────────────────────────────────────────────────
  
  • CONTEXT WINDOW GAUGE
    See how full the current conversation is — as a percentage and as a 
    token count. Know exactly when you're approaching the model's memory 
    limit before your next message is sent.
  
  • 5-HOUR & WEEKLY QUOTA BARS
    Claude Pro and Max plans throttle usage over a 5-hour rolling window 
    and a weekly budget. Token Monitor shows both bars in real time and 
    estimates when each will reset.
  
  • TRUNCATION RISK WARNING  ⚠
    Before you hit Send, Token Monitor calculates whether your next 
    message — combined with the predicted output — will overflow the 
    context window. An inline banner warns you with an actionable 
    suggestion: "Split into smaller questions" or "Start a new chat."
  
  • OUTPUT SIZE PREDICTION  📊
    A small pill next to the Send button predicts whether the reply will 
    be Small, Medium, Large, or XL — so you can plan accordingly.
  
  • PER-TURN TOKEN COST
    Token cost badges appear beneath every user message, showing input 
    and output token counts for that turn.
  
  • STREAMING AWARENESS
    While Claude is generating, Token Monitor tracks tokens already 
    committed to the input and tokens streamed back in real time.
  
  • SELF-CALIBRATING ESTIMATES
    Token counts are estimated via a fast heuristic (no API calls). 
    When Claude shows "X messages left" banners, Token Monitor uses those 
    signals to self-correct its estimates over time — getting more 
    accurate the more you use it.
  
  ─────────────────────────────────────────────────────────
  🔒  PRIVACY & PERMISSIONS
  ─────────────────────────────────────────────────────────
  
  Token Monitor is built with a privacy-first design:
  
  • All processing happens locally in your browser. Your conversations 
    are never transmitted to any external server.
  • The extension only runs on claude.ai domains.
  • Quota data is fetched directly from Claude's own settings endpoint 
    (the same data shown in your account dashboard) — no third-party 
    services are involved.
  • No account creation required. No login. No analytics.
  
  Permissions used:
    storage     — saves your preferences and calibration data locally
    alarms      — schedules periodic quota refreshes
    notifications — optional limit-reached desktop alerts
    scripting / tabs — required to inject the UI into claude.ai
  
  ─────────────────────────────────────────────────────────
  ⚙️  SUPPORTED PLANS & MODELS
  ─────────────────────────────────────────────────────────
  
  Works with:
    • Claude Free, Pro, Max (5x and 20x), Team, and Enterprise accounts
    • All Claude models available on claude.ai: Sonnet, Opus, Haiku
    • Conversations inside Projects (project knowledge token cost included)
  
  ─────────────────────────────────────────────────────────
  🌐  LANGUAGES
  ─────────────────────────────────────────────────────────
  
  UI available in English and 简体中文 (Simplified Chinese).
  
  ─────────────────────────────────────────────────────────
  📌  NOTES
  ─────────────────────────────────────────────────────────
  
  • Anthropic occasionally changes the claude.ai DOM structure. If an 
    indicator stops updating, check for an extension update.
  • Token estimates use a heuristic (not the exact tokenizer). The 
    self-calibration loop corrects drift over time.
  • This extension is an independent tool — it is not affiliated with or 
    endorsed by Anthropic.

• May 16, 2026
  short_description

  Real-time context-window monitor for AI chat tools — see how much you've used and get smart alerts before replies are cut off.

  Real-time context & quota monitor for Claude.ai — see usage live and get smart alerts before your replies are cut off.

• May 16, 2026
  host_permissions

  https://claude.ai/*, https://chatgpt.com/*, https://gemini.google.com/*, https://aistudio.google.com/*, https://www.perplexity.ai/*, https://poe.com/*

  *://claude.ai/*, *://*.claude.ai/*, https://api.web3forms.com/*

• May 16, 2026
  permissions

  storage

  storage, alarms, notifications, scripting, tabs

Permissions & access

Permissions

storagealarmsnotificationsscriptingtabsdownloads

Host access

*://claude.ai/*, *://*.claude.ai/*, *://status.anthropic.com/*, https://raw.githubusercontent.com/*

Screenshots