grafana/gcx

A CLI for managing Grafana and Grafana Cloud resources. Optimized for agentic usage.

gcx is a command-line tool (CLI) that brings Grafana — a platform for monitoring and observability — directly into your terminal or AI coding assistant. Instead of switching between browser tabs, you can query live metrics, check alerts, manage dashboards, and more, all from the command line or through an AI agent that runs shell commands.

What it does:

  • Queries data: Run PromQL (metrics), LogQL (logs), and similar queries for traces and profiles directly from the terminal.
  • Manages resources: Pull and push dashboards, folders, and alert rules as code (GitOps-style).
  • Checks alerts and SLOs: List firing alerts, review service-level objectives, and investigate incidents.
  • Works with Grafana Cloud products: Includes commands for SLOs, Synthetic Monitoring, IRM (incident response), k6 load testing, Fleet Management, and more — but only on Grafana Cloud.
  • Integrates with AI agents: Ships a bundle of “skills” that let AI coding tools (like Claude Code, OpenAI Codex) investigate alerts, create dashboards, and fix issues using gcx commands.

Compatibility:

  • Works with Grafana Cloud, Enterprise, and OSS (self-hosted) versions.
  • Full support for Grafana 13+; Grafana 12 works but with limited features; versions below 12 are not supported.
  • Cloud-only products (SLO, Synthetic Monitoring, etc.) require Grafana Cloud.

Installation:

  • Quick script: curl -fsSL https://raw.githubusercontent.com/grafana/gcx/main/scripts/install.sh | sh
  • Homebrew: brew install gcx

Example usage:

gcx metrics query 'rate(http_requests_total[5m])' --since 1h
gcx alert rules list --state firing
gcx slo definitions list

Costs: gcx itself is free, but some Grafana Cloud products it interacts with (like Assistant, Synthetic Monitoring, k6) are billed based on usage. Queries and resource push/pull are not billed.

In short, gcx is a bridge between Grafana’s data and your terminal or AI assistant, making observability workflows faster and more scriptable.

Related

  • Project
    bolna-ai/bolna

    Bolna is an end-to-end open-source platform for building production-ready voice conversational agents. It orchestrates speech recognition, LLM reasoning, and speech synthesis over websockets, supporting multiple telephony, ASR, LLM, and TTS providers, and can be configured via JSON or a simple Python API.

  • Project
    ryoiki-tokuiten/Deepthink

    A multi-agent framework that scales test-time compute through parallel strategy exploration and structured noise injection to solve complex problems without getting stuck in cognitive loops.

  • Project
    Gnosil/semantix

    A memory kernel for AI coding agents that provides cross-session learning and optimizes provider prefix caches to reduce API costs and latency.

  • Project
    getomnico/omni

    An open-source, self-hosted AI agent for companies that integrates with workplace tools to gather context, analyze data, and execute tasks across systems.