Feature Request: Pro Plan 5-Hour Window Usage Visibility

Status Closed — duplicate
Maintainer reply ✓ Yes — claude[bot]
Activity 12 comments · opened Oct 15, 2025 · closed May 10, 2026
💡 Likely answer: A maintainer (claude[bot], contributor) responded on this thread — see the highlighted reply below.

Feature Request: Pro Plan 5-Hour Window Usage Visibility

Summary

As a Claude Pro plan subscriber, I need visibility into my current usage against the 5-hour rolling window token limit to better manage my sessions and avoid unexpected interruptions.

Problem Statement

Currently, there is no way for Pro plan users to:

  • Check how many tokens/messages they've used in the current 5-hour window
  • See how much capacity remains before hitting the limit
  • Know when their 5-hour window will reset
  • Plan their work accordingly to avoid hitting limits during critical tasks

Users only discover they've hit the limit when they receive a rate limit error, which can be disruptive to their workflow.

Proposed Solution

Option 1: Add Usage Data to Statusline JSON

Extend the JSON data passed to custom statusline scripts to include:

{
  "pro_plan_window": {
    "output_tokens_used": 35000,
    "output_tokens_limit": 44000,
    "usage_percentage": 79.5,
    "window_started_at": "2025-10-15T18:33:00Z",
    "window_resets_at": "2025-10-15T23:33:00Z",
    "minutes_until_reset": 47
  }
}

This would allow users to display usage in their custom statuslines, for example:

Sonnet 4.5 | tesla-marketplace  main ● | October 15, 2025 11:42 PM | Window: 79% (47m until reset)

Option 2: Add a /usage or /window Slash Command

Create a command similar to /context that displays:

5-Hour Window Usage

[████████████████████░░░░] 79% (35k/44k tokens)

Window started: 6:33 PM
Resets in:      47 minutes
Next reset:     11:33 PM

💡 Tip: You have approximately 9 messages remaining in this window

Option 3: Both Solutions

Implement both the statusline JSON extension (for always-visible tracking) and the slash command (for detailed usage breakdowns).

Use Cases

  1. Avoid Workflow Interruptions: Plan intensive refactoring or feature development when you have sufficient remaining capacity
  2. Session Management: Know when to take a break or switch to lighter tasks as you approach the limit
  3. Team Coordination: Better understand and communicate usage patterns when multiple team members share costs
  4. Debugging Sessions: Allocate sufficient capacity for troubleshooting complex issues
  5. Learning and Exploration: Budget usage for learning sessions vs production work

Benefits

  • Transparency: Users understand their subscription limits and usage patterns
  • Better Planning: Make informed decisions about when to tackle complex tasks
  • Reduced Frustration: No unexpected interruptions mid-workflow
  • Improved User Experience: Aligns with professional development workflows
  • Subscription Value: Helps users maximize the value of their Pro plan

Additional Notes

  • Usage limits are shared between claude.ai and Claude Code, so showing aggregate usage would be most helpful
  • Consider showing both token-based and message-based estimates since different tasks consume different amounts
  • Weekly quota information (added in August 2025) would also be valuable to display
  • Similar functionality could benefit Claude Max users with their respective limits

Related Features

This complements existing features:

  • /context command (shows conversation context usage)
  • /cost command (shows API costs for team/enterprise users)
  • Custom statusline support
  • Analytics and monitoring tools

Priority

High - This affects the core user experience for all Pro and Max plan subscribers who rely on Claude Code for professional development work.

---

Submitted by: Pro Plan User
Date: October 15, 2025
Claude Code Version: 2.0.19

View original on GitHub ↗

12 Comments

github-actions[bot] · 10 months ago

Found 3 possible duplicate issues:

  1. https://github.com/anthropics/claude-code/issues/8464
  2. https://github.com/anthropics/claude-code/issues/8412
  3. https://github.com/anthropics/claude-code/issues/5621

This issue will be automatically closed as a duplicate in 3 days.

  • If your issue is a duplicate, please close it and 👍 the existing issue instead
  • To prevent auto-closure, add a comment or 👎 this comment

🤖 Generated with Claude Code

diwu-sf · 10 months ago

/usage doesn't work for you?
<img width="469" height="307" alt="Image" src="https://github.com/user-attachments/assets/4e7f9feb-c383-4f10-a2b8-aa5397e3ce9e" />

Redster1 · 10 months ago

The advantage of a status line is that it's always visible, so I don't have to interrupt the conversation I'm having with Claude-Code to check.

github-actions[bot] · 8 months ago

This issue has been inactive for 30 days. If the issue is still occurring, please comment to let us know. Otherwise, this issue will be automatically closed in 30 days for housekeeping purposes.

Redster1 · 8 months ago

This issue should remain open because the problem still exists and this issue has a clear proposal.

kion · 7 months ago

Having this data provided within the status line script input would be extremely helpful. Hope this gets implemented soon. 🙏🏻

rickardstureborg · 7 months ago

I just implemented a workaround by spawning a new claude session and using the /usage command. Not an ideal solution but it's here for anyone interested:
rickardstureborg/claude-rate-limit-status

ends up looking like this:

<img width="735" height="79" alt="Image" src="https://github.com/user-attachments/assets/6b40d49f-f37f-4578-bd73-b018afb43a89" />

br3nt · 5 months ago

This would also be useful to display in Mac status bar, and in the Windows taskbar.
Though it always makes me nervous and anxious when I see the usage reaching its limits

hyang0129 · 5 months ago

Built a small CLI tool that solves this for Linux users:

pip install claude-usage-plz
claude-usage
Session (5h):        59% used  (resets 11pm (UTC))
Week (all):          33% used  (resets Apr 7, 3pm (UTC))
Week (Sonnet):       12% used  (resets Apr 7, 5pm (UTC))

Also works as an importable Python function for scripting/automation:

from claude_usage import get_usage

usage = get_usage()                              # default ~/.claude
usage = get_usage("/path/to/.claude-work")       # specific profile

print(f"5h: {usage.five_hour_pct}%, 7d: {usage.seven_day_pct}%")

JSON output for piping: claude-usage --json

It works by spawning claude interactively in a virtual terminal (pexpect + pyte), running /usage, and parsing the screen — since there's no API for this data yet. ~15s per call.

Repo: https://github.com/hyang0129/claude-usage-plz

cc @deanandreakis @Redster1 @kion @rickardstureborg @br3nt

Solomonic · 4 months ago

Built something for this. BudgetClaw parses the local JSONL logs and tracks cost per project and per git branch. You can set daily/weekly caps and it will warn or kill the session on breach. Not a replacement for native window visibility, but it stops the "find out when you hit the wall" problem.

https://github.com/RoninForge/budgetclaw

claude[bot] contributor · 3 months ago

This is a duplicate of #18121, which was fixed as of version 2.1.80.

github-actions[bot] · 2 months ago

This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.