← all repositories

onllm-dev/onWatch

A background daemon that monitors AI API usage, quotas, and spending across multiple LLM providers in real time.

661 stars Go LLMOps · Eval
onWatch
Velocity · 7d
+5.4
★ / day
Trend
steady
star history

onWatch tracks API quotas and token usage across providers including Claude Code, Codex, GitHub Copilot, and others via polling their public endpoints. It stores historical data in SQLite and presents a local web dashboard for real-time monitoring, alerting, and usage analytics. The tool runs as a lightweight background daemon and supports ingesting telemetry from custom API-driven workflows.

heatdrop uses Google Analytics to see which pages get read — nothing else. Your call. How we handle data.