onllm-dev/onWatch

Track AI API quotas across Synthetic, Z.ai, Anthropic (Claude Code), Codex, GitHub Copilot & Antigravity in real time. Lightweight background daemon (<50MB RAM), SQLite storage, Material Design 3 dashboard. Zero telemetry.

What it solves

Many AI providers only show a current snapshot of quota usage, leaving developers with "budget anxiety" and surprise throttling. onWatch provides a unified, historical view of AI API quota consumption across multiple providers, allowing users to track burn rates, project future usage, and avoid hitting limits unexpectedly.

How it works

onWatch runs as a lightweight background agent (using less than 50 MB of RAM) that polls configured AI providers (such as Anthropic, GitHub Copilot, Cursor, and others) using API keys. It stores historical usage data locally in a SQLite database and serves a Material Design 3 web dashboard. It also supports local JSONL ingestion for custom API-driven workflows.

Who it’s for

  • Solo developers and freelancers using AI coding agents (like Claude Code, Cline, or Roo Code) who need to avoid mid-task throttling.
  • Small development teams sharing API keys who need shared visibility for budget planning.
  • DevOps and platform engineers who want to integrate AI usage monitoring into their existing monitoring stacks via Prometheus metrics.
  • Privacy-conscious developers who require local, auditable monitoring with zero telemetry.

Highlights

  • Multi-provider support: Tracks quotas for a wide range of services including Anthropic, Codex, GitHub Copilot, MiniMax, Cursor, Grok, and DeepSeek.
  • Unified Dashboard: A local web interface with time-series charts, burn rate forecasting, and live countdown timers to quota resets.
  • Integrated Experience: Includes a VS Code extension for status bar visibility and a desktop tray companion (macOS, Linux, Windows) for quick quota checks.
  • Local-first architecture: Single binary deployment with all data stored locally in SQLite; no external telemetry.
  • Alerting: Supports email (SMTP) and browser push notifications when quotas cross warning or critical thresholds.
  • Custom Telemetry: Ability to ingest local telemetry from custom API-driven workflows via API Integrations.

Related

  • Project
    slkiser/opencode-quota

    A quota and token tracking tool for OpenCode that provides real-time visibility into AI model usage and usage limits across multiple providers.

  • Project
    VasiHemanth/tokentelemetry

    A local observability dashboard that tracks token usage, LLM costs, and session traces for AI coding and autonomous agents by reading their local log files.

  • Project
    opgginc/opencode-bar

    A macOS menu bar app that automatically monitors and aggregates real-time usage and billing for multiple AI providers in one unified dashboard.

  • Project
    akitaonrails/ai-usagebar

    A system utility that tracks and displays AI plan usage and reset timers for multiple LLM providers across status bars, a TUI, and CLI.

  • Project
    Javis603/token-monitor

    Token Monitor is a cross‑platform desktop widget that reads local logs and API‑key balances from 35+ AI coding assistants, shows live token usage, cost, and quota limits, and can sync those summaries across multiple machines via a self‑hosted or Cloudflare hub. It offers per‑session breakdowns, historical heatmaps, export to CSV/JSON, and a highly customizable UI, all while keeping raw prompt data private.