Operations | Monitoring | ITSM | DevOps | Cloud

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?".

We at Checkly shipped a major CLI v8 release. Here's everything new, and why it's built for your agent. The CLI v8 expands Monitoring as Code and streamlines your everyday workflows: Upgrade and try it: npm install checkly@latest Chapters.

The new checkly api command is an authenticated pass-through to all 100+ Checkly API endpoints. Give it a path, and it handles your token and base URL. Full API coverage for your agents, no wrappers required.

Alerts tell you something broke, but not why. So you're stuck digging through logs and trace IDs. Checkly monitors your app from the outside, like a real user. And Rocky, the Checkly agent, automatically pulls the right context to provide a root cause analysis for any failed check.

Your Playwright tests catch bugs. The hard part is figuring out what actually broke — and sharing that context with your team. This session shows exactly how the Checkly Playwright Reporter solves that: one shared home for all your test runs, AI-powered root cause analysis, and a direct path from failing test to production monitor. María de Antón, PM for Playwright features at Checkly, runs a live demo on a real app with real failures.

Not everything worth monitoring is on the public internet. In this 30-minute hands-on session, Daniel Paulus deploys four Checkly private location agents on AWS EKS with Terraform, then uses a coding agent to scaffold 200 internal checks in seconds — uptime, TCP, DNS, ICMP, and Playwright browser checks against legacy apps that never leave the firewall.

Coding agents are the fastest-growing audience for the Checkly CLI, and we're doubling down on them. In this session, Stefan hands Claude a real e-commerce app, lets it set up monitoring with `npx checkly init`, generate Playwright tests through MCP, and walk an actual alert end-to-end with Rocky AI in the loop.

Rocky, Checkly's AI agent, monitors production sites and provides an analysis for every failing check. Previously, a coding agent couldn't access this analysis, leaving incidents and agents disconnected. Now, you can access all the analyses via the Checkly CLI (or API) and tell your coding agent, "Hey, I got a Checkly alert. Please investigate!" With Rocky's structured analysis delivered inline, the coding agent can start with a strong hypothesis, fix issues, and propose a PR in one session.

Learn how the Checkly CLI uses a single function (`detectOperator()`) to detect whether the caller is a human, CI, or a coding agent by checking agent-set environment variables. This detection then changes how commands behave to provide the best agent experience.

Building Agent-Friendly CLIs: Why Your AI Agent Already Loves the Checkly CLI Stefan explains why products, docs, and CLIs must be AI-ready as coding agents rapidly become primary users of the Checkly CLI. He outlines key CLI features for agent workflows: Stefan demos how an agent initializes project-tailored Checkly setup from scratch without any human intervention and also shows how agents can entirely automate the incident life cylce from resolution to status page communication.