> For the complete documentation index, see [llms.txt](https://docs.collieai.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.collieai.io/monitoring.md).

# Monitoring

- [Monitoring overview](https://docs.collieai.io/monitoring/overview.md): How CollieAi monitoring works — built-in observability for every LLM API call, with logs, analytics, and threshold-based alerts, automatically scoped per project.
- [Logs](https://docs.collieai.io/monitoring/logs.md): How CollieAi logs work — a searchable, per-project history of every API call with event type, latency breakdown, tokens, triggered rules, and block status, stored in ClickHouse.
- [Exporting logs](https://docs.collieai.io/monitoring/exporting-logs.md): How to export CollieAi logs programmatically — authenticating against the Logs API, applying filters (event type, status, direction, date range, full-text search), and paginating a stable export to ND
- [Analytics](https://docs.collieai.io/monitoring/analytics.md): How CollieAi analytics work — aggregated KPIs and charts for request volume, blocked rate, error rate, P95 latency, and rule triggers, with a latency breakdown by stage, scoped per project.
- [Alerts](https://docs.collieai.io/monitoring/alerts.md): How CollieAi alerts work — define threshold rules on metrics like error rate, blocked rate, and P95 latency, evaluated every 60 seconds, with cooldowns and acknowledgement.