The Changing Face of Engineering Workflows
The way engineers write code is evolving rapidly. Agentic tools like Cursor and Claude Code now handle many daily tasks directly from the terminal, dramatically accelerating code generation. However, this new speed comes at a cost: engineers still face constant context switching when they must jump into a different tool—like a web-based dashboard—to understand what's happening in production. More critically, these agents themselves operate with a significant blind spot: they see the code on your machine but are completely unaware of the real-world behavior of your systems.
The Visibility Gap: Agents Need Production Context
When an agent writes code based solely on source files, it's essentially pattern-matching and hoping for the best. It doesn't know if there's a latency spike on the checkout page, or whether your service is meeting its SLOs. Without access to production data, agents are guessing at correct behavior rather than building on actual system performance. This gap can lead to irrelevant or even harmful code changes.
Introducing gcx: Grafana Cloud in Your Terminal
To bridge this gap, Grafana has launched the public preview of gcx, the new Grafana Cloud CLI. gcx brings the full power of Grafana Cloud and the Grafana Assistant directly into your terminal—and into the agentic coding environment running inside it. The result: engineers and agents can spot and resolve incidents in minutes, not hours, without ever leaving the command line.
From Greenfield to Full Observability in Minutes
Most services start with no instrumentation, no alerts, and no SLOs. That's normal, and gcx treats it as a starting line, not a blocker. Simply point your agent at the service and ask it to bring things up to standard. The CLI exposes the primitives needed across the entire observability lifecycle:
Instrumentation
Wire OpenTelemetry into your codebase, validate that metrics, logs, and traces are flowing, and confirm the data lands in the correct backends—all from the terminal. No need to open a separate dashboard or wait for a manual setup ticket.
Alerting, SLOs, and Synthetics
Generate alert rules based on the signals your service actually emits. Define an SLO against a real latency or availability indicator and push it live. Stand up synthetic probes so users aren't the first to report an outage. Everything is done via the CLI, making it trivial to iterate.
Frontend, Backend, and Kubernetes Observability
Onboard a Faro-instrumented frontend: create the app, manage sourcemaps so stack traces are readable. For backend services and Kubernetes infrastructure, use the Instrumentation Hub to get wired up quickly. No matter your stack, gcx streamlines the onboarding.
Everything as Code
gcx treats dashboards, alerts, SLOs, and checks as files you can pull, edit locally with your agent, and push back. Need a human to dive deeper? Open a deep link into Grafana Cloud directly from the terminal. This as-code approach turns a multi-day ticket into a one-agent session.
Why This Matters for Agentic Tools
The real power of gcx shines when you give your agents access to it. Without production context, an agent is just pattern-matching on source files. With gcx, the same agent can read the current state of the running system—latency, errors, traffic—and make far more informed decisions based on actual behavior rather than hypotheticals. It transforms the agent from a blind code generator into a true debugging and operations partner.
Conclusion
gcx bridges the gap between fast code generation and real-world observability. By bringing Grafana Cloud into the terminal and into your agent's workflow, it eliminates context switches and gives agents the production context they desperately need. The result is faster, more reliable development and operations—all from the command line you already use every day.