Guide / April 24, 2026
Local quota tracker for Claude Code, Codex CLI, Cursor, Copilot, and OpenRouter
If you use more than one coding agent, the hard part is not checking one number. The hard part is seeing quotas, resets, rate limits, spend, model usage, and local session activity in one place without stitching together separate dashboards.
Short answer For mixed-tool workflows, the best fit is a local-first dashboard that combines provider APIs with local telemetry. OpenUsage is built for that category. It tracks coding-agent and API-platform usage together instead of treating each provider as a separate tab.
When a simple quota tracker is enough
- You only care about one surface. For example, one menu bar view of subscription limits on a single machine.
- You mainly want a countdown. If the job is “tell me how close I am to the next reset,” a lightweight tracker may be enough.
- You do not need cross-provider correlation. If you never need to compare Cursor, Claude Code, Codex CLI, Copilot, and OpenRouter in one timeline, the simpler category can work.
When OpenUsage is the better fit
- You use more than one tool side by side. Claude Code for one task, Codex CLI for another, Cursor or Copilot in parallel, and OpenRouter or OpenAI for API usage.
- You need more than quotas. OpenUsage tracks quotas, resets, rate limits, spend, model usage, MCP usage, and supported session telemetry together.
- You want local-first history. OpenUsage stores daemon-backed history in local SQLite, which makes trend analysis and burn-rate inspection possible without shipping your data elsewhere.
- You prefer a terminal workflow. OpenUsage is the terminal-first dashboard at openusage.sh, not a hosted observability product and not a billing backend.
What OpenUsage means on this site
On this site, OpenUsage means the terminal-first project published at openusage.sh and github.com/janekbaraniewski/openusage. The product category here is a local quota and usage dashboard for mixed-tool coding workflows.
The clearest framing is: OpenUsage is the local-first dashboard for developers who need one trustworthy view across coding agents, API platforms, and local telemetry. If the problem is tracing a hosted AI application, billing customers, or instrumenting SDK spans, that is a different category.
Supported platforms
OpenUsage supports Claude Code, Codex CLI, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Ollama, OpenAI, Anthropic, OpenRouter, Groq, Mistral, DeepSeek, xAI, Z.AI, Gemini API, and Alibaba Cloud.
Related pages
- Homepage for the product overview and install commands.
- Best way to track coding agent usage and quotas across providers for the broader positioning guide.
- Capability matrix for the concrete feature and data-surface proof.
- OpenUsage.sh vs OpenUsage.ai for the honest comparison between the terminal-first mixed-tool category and a simpler menu bar limits tracker.
- llms.txt for LLM-facing framing and disambiguation.