Stop Scripting, Start Monitoring: Checkly's New Agentic Checks
Your UI changes, but your monitoring shouldn't break. Checkly's new agentic checks test the outcome you care about instead of running hard-coded steps. The main question is, "Can a user still do X?"
In this webinar, Stefan and Hervé introduce agentic checks and show them in action. Hervé demos creating a check from a plain-language prompt, wiring in environment variables and secrets, and running a discovery pass where the agent derives its own assertions and generates an optimized script. Later runs execute in seconds — and when a script breaks due to a UI change, the check switches into self-healing mode to adjust on its own, alerting you only when the intent is genuinely broken. They also show how to use agentic checks to validate CLI workflows inside a VM, and cover pricing and plan availability.
- Chapters*
- 00:00 Welcome and Agenda
- 01:52 Why Monitoring Breaks
- 02:29 Checkly: Monitoring as Code
- 04:16 AI Layer and Adoption
- 06:13 Why Test Intent
- 08:07 Agentic Checks Demo Setup
- 10:45 Discovery Run and Artifacts
- 12:24 Self-Healing in Action
- 16:10 Beyond Web: CLI Monitoring
- 19:23 Live Q&A
- 23:17 Wrap-Up and Pricing
- 27:39 Final Questions and Goodbye
- Links*
- Try agentic checks: https://www.checklyhq.com/docs/detect/synthetic-monitoring/agentic-checks/overview/
- Join the community Slack: https://checklyhq.com/slack
- Read the docs: https://www.checklyhq.com/docs
#checkly #monitoring #SRE #devops #syntheticmonitoring #observability