The Changing Landscape of Coding and Observability
The way engineers write code is evolving rapidly, and so must the methods for monitoring systems and responding to incidents. Today, many developers spend a significant portion of their day working from the command line, thanks to powerful agentic tools like Cursor and Claude Code that handle routine engineering tasks with remarkable efficiency. While these tools accelerate code generation, they don't eliminate the friction of switching context when you need to leave your terminal and log into a separate observability platform.
Moreover, AI agents introduce a new kind of visibility gap. They can see your source code on your local machine, but they remain blind to what's happening in your production environment. They don't notice a sudden latency spike on the checkout page, and they have no awareness of whether your service is meeting its Service Level Objectives (SLOs). Agents generate code based on assumptions about what might happen, rather than what is actually occurring in real time.
Introducing gcx: Your CLI Gateway to Grafana Cloud
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 Grafana Assistant directly into your terminal—and into the agentic coding environment running within it. This allows you to spot and resolve incidents in minutes instead of hours, all without leaving your command line.
From Greenfield to Full Observability in Minutes
gcx is built to do the heavy lifting. Most services begin life with no instrumentation, no alerting, and no SLOs—that's the normal starting point. gcx treats this as a starting line, not a blocker. Simply point your agent at the service and ask it to bring the system up to standard. gcx exposes the primitives the agent needs across every stage of the observability lifecycle.
Instrumentation
Wire OpenTelemetry into your codebase directly from the terminal. Validate that metrics, logs, and traces are flowing, and confirm the data lands in the correct metric, logs, and traces backends—all without leaving your CLI.
Alerting, SLOs, and Synthetics Checks
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 immediately. Stand up synthetic probes so that your users aren't the first to report an outage. This ensures proactive monitoring from day one.
Frontend Observability, Application Observability, and Kubernetes Monitoring
Onboard a Faro-instrumented frontend: create the application, manage sourcemaps so that stack traces are readable in production. For backend services and Kubernetes infrastructure, use the Instrumentation Hub to get up and running quickly.
Everything as Code
Pull dashboards, alerts, SLOs, and checks as files. Edit them locally with your agent, then push the changes back. When a human needs to investigate further, open a deep link directly into Grafana Cloud with a single command. What previously required a multi-day ticket can now be handled in a single agent session.
Why This Matters for AI Agents
Having CLI access to Grafana and managing observability as code is already powerful, but gcx truly shines when you give your AI agents access to it.
Without production context, an agent is essentially pattern-matching on source files, hoping to find the right answer. With gcx, the same agent can read the live state of the running system and make more informed decisions based on actual data. This transforms the agent from a guesswork-driven code generator into a context-aware assistant that understands your production environment.
Conclusion
gcx is a game-changer for engineers and AI agents alike. By bringing observability into the terminal and making it accessible as code, it streamlines the entire lifecycle from instrumentation to incident response. Whether you're a solo developer or part of a large team, gcx helps you and your agents work faster, smarter, and with greater awareness of your system's real behavior. Try the public preview today and experience the future of observability at the command line.