Agent Cat
agentcat.app- Category
- Developer Tools
- Rank
- No. 2700Tools index
Previous survey · No. 2603 ·
- Pricing
- Free
- Type
- APP
- Use case
- Productivity & Collaboration
- Interfaces
- Desktop · Mobile
- Date
About
Agent Cat is a free menu bar app for macOS and Windows, with companion iOS, Android, and watch apps, that tracks token usage, quota limits, and reset times for AI coding agents including Codex, Claude Code, and Gemini CLI. It breaks down usage by device and project, sends notifications when a quota resets, and states that it never stores or transmits prompts, responses, or code.
At a glance
Agent Cat brings coding-agent activity, token counts and available quota resets into a menu-bar or tray view. It is worth evaluating if you switch between providers or accounts often enough that checking each one interrupts your work.
Good fit
Developers who want a view of activity across supported coding agents and are comfortable with the collector reading local usage records.
Consider something else when
Use provider dashboards for authoritative usage and billing, or when your organization restricts tools that can read agent log directories.
Start here
Start with one supported agent and leave sync off. Compare one account’s displayed usage and reset time with the provider’s own view before adding more accounts.
Cost and requirements
The app is free, with Mac and Windows versions and mobile views. The Windows connector needs Python 3. Agent subscriptions and usage are billed separately by their providers.
Limits to know
The publisher says the Mac collector reads local logs without uploading the files. Optional sync stores account-linked summaries including provider, model and token counts. Quota visibility depends on the provider and connection method.
Sources checked 2026-11-05. Source-based assessment.
Why it made the leaderboard
Coding-agent users get cross-device visibility into remaining quota, reset timing, and per-project token spend, with team dashboards for shared accounts.
Tags
Media
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.