AppRising

Claude Code Usage Monitor

App : A product you use directly

Real-time monitor for Claude Code quotas and tokens in the terminal.

Useful for quota-based subscriptions; ccusage remains more complete for cost tracking.

GitHub · 8,764 ★

Overall score

8.1 /10

Security: nothing to flag

Utility

7.0

Reliability

8.0

Security

No caveat on record

Ease

5 min · Comfortable

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

When using Claude Code intensively, it is hard to know in real time how many tokens or messages remain before hitting a subscription limit. Claude Code Usage Monitor solves this by displaying a continuously updated dashboard directly in the terminal, without leaving the working environment.

The tool reads Claude Code's local data files, calculates consumption for the current session over a 5-hour window, and shows tokens used, message count, and estimated cost. It can also connect to official API rate limits via a statusline hook, allowing it to distinguish certified data from local estimates through provenance labels.

Claude Code Usage Monitor is aimed at developers using Claude Code with a quota-based subscription (Pro, Max5, Max20) who want to anticipate session interruptions. A custom plan automatically analyses the last 8 days of usage to calculate thresholds tailored to each usage profile.

The tool is not suited for those looking for a graphical interface, centralised multi-account tracking, or direct integration with other AI providers. It works exclusively from the command line and covers Claude Code only.

What it offers

·Real-time terminal dashboard showing tokens, messages, and cost for the current session.
·Default custom plan: analyses the last 8 days of usage to automatically calculate limits tailored to your profile.
·Provenance labels on every displayed figure: 'official', 'local_estimate', 'experimental', or 'unknown' depending on the source.
·Statusline hook to capture official Claude Code limits via the API rate_limits field.
·Machine-readable output in JSON, text, or CSV via --once, --compact, and --write-state options, with automation-friendly exit codes.
·Optional local warehouse to retain usage history beyond Claude's automatic 30-day cleanup, with export by project, model, and day.
·Consumption forecasting and pace calculation to estimate when the session limit will be reached.
·Automatic preference saving (plan, theme, timezone, refresh rates) to ~/.claude-monitor/last_used.json.

Our tips

Use the --statusline option to install the tool as a Claude Code hook: it then retrieves official API limits rather than local estimates, improving alert reliability.
Run claude-monitor --once --output json to get a snapshot readable by a script or another tool, without starting the interactive dashboard.
Enable --warehouse from the start of your usage to build a local history that will survive Claude's automatic 30-day cleanup.
If you frequently change plan or timezone, use --clear to reset saved preferences and start from a clean configuration.
For status bars or external dashboards, prefer --write-state with a fixed file path: writes are atomic and prevent partial reads.

Watch out

On recent Linux distributions (Ubuntu 23.04+, Debian 12+, Fedora 38+), installation via pip may fail with an externally-managed-environment error; using uv or pipx is recommended to avoid this issue.
The 'team' plan is labelled as unverified in the sources: limits displayed for this plan are estimates, not confirmed official values.

In practice

Price: Free
Commercial use: Commercial use: yes
Available on: cli
Licence: MIT

Security and data

Rating: Security: nothing to flag
Your data: Data stays local
Maintenance: Active
Last release: 27 June 2026

Links and repository checked on 25 October 2026 · added on 1 June 2025

How to install Claude Code Usage Monitor

1. Install uv if not already available: curl -LsSf https://astral.sh/uv/install.sh | sh, then restart the terminal. 2. Install claude-monitor from PyPI: uv tool install claude-monitor 3. Launch the tool with: claude-monitor (or cmonitor, ccmonitor, ccm) 4. To specify your subscription on first launch: claude-monitor --plan pro (or max5, max20, custom) 5. Security note: the tool only reads local Claude Code files on your machine and does not transmit any data externally; the repository is MIT-licensed and the code is auditable on GitHub.

uv tool install claude-monitor

In the same category

Whole category Route & save →