llms.txt
Compare with other sites' llms.txt files →# Alertee
> Alertee is SQL data quality monitoring for production Postgres and ClickHouse. It catches the failure mode uptime monitors miss — the app is up, but the database says the outcome stopped happening — or quietly went wrong: stale tables, missing rows, wrong or invalid values, stuck jobs, failed payments, quiet tenants. You write a SQL check (or let Alertee's AI agent author and tune one from your schema), it runs on a schedule, and when a condition breaks it opens an incident in your Inbox and alerts your team via email, Slack, or webhook.
## How it works
- Two ways to get a check: write plain SQL yourself, or let Alertee's AI agent inspect your schema and draft a baseline-aware query with a threshold. Agent-owned checks are "managed by Alertee" and re-tuned on a cadence; you can take any check back at any time.
- Checks run on an interval (1–60 min by plan). Single-value checks compare one number to a threshold; grid checks evaluate many dimensions (per store, region, tenant) at once.
- A failing check opens an incident in the Inbox — acknowledge, assign, resolve, and classify the outcome (real, transient, expected, noisy). Outcomes feed back to reduce future noise.
- Read-only database access; credentials encrypted per organization. Postgres and ClickHouse, self-hosted or managed (Supabase, Neon, RDS, Cloud SQL, ClickHouse Cloud).
## Plans
- Free: $0 — 5 checks, 2 connections, 15-min minimum interval.
- Starter: $29/mo — 25 checks, 5 connections, 5-min minimum interval.
- Team: $99/mo — 500 checks, 50 connections, 1-min minimum interval.
- Every plan includes AI-assisted setup and the data incident inbox.
## Docs
- [Introduction](https://docs.alertee.io/introduction): What Alertee is and how it works.
- [Quickstart](https://docs.alertee.io/quickstart): Create your first check.
- [Checks overview](https://docs.alertee.io/checks/overview): Anatomy of a check and ownership.
- [Managed checks](https://docs.alertee.io/checks/managed): AI-authored, self-tuning checks.
- [Conditions](https://docs.alertee.io/checks/conditions): Single-value and grid conditions.
- [The Inbox](https://docs.alertee.io/inbox/overview): Incidents, outcomes, and self-improving alerts.
- [Plans & billing](https://docs.alertee.io/billing/overview): Limits and upgrading.
- [CLI & MCP](https://docs.alertee.io/cli/installation): Drive Alertee from a terminal or an AI agent.
## Pages
- [Pricing](https://alertee.io/pricing): Plans and what each includes.
- [Data quality monitoring](https://alertee.io/data-quality-monitoring): The problem Alertee solves.
- [SQL data quality checks](https://alertee.io/data-quality-checks): SQL patterns that catch stale rows, empty imports, dropped tenants, and row-count collapses in Postgres and ClickHouse.
- [Dead man's switch](https://alertee.io/deadmans-switch): One curl at the end of any cron job, backup, or script — get alerted when it stops checking in or the value it pushes goes wrong. No agent, no credentials.
- [Compare](https://alertee.io/compare): Alertee vs Metaplane, Monte Carlo, and Soda.
- [Blog](https://alertee.io/blog): Guides and updates.