Last updated: July 16, 2026
About the Dash0 CLI
Introduction to the Dash0 CLI — what it is, who it's for, and the design principles that shape its ergonomics for humans, AI agents, and CI/CD.
The Dash0 CLI (dash0) is the command-line interface for Dash0.
It exposes the same primitives as the web UI — dashboards, views, check rules, synthetic checks, teams, notification channels — plus telemetry queries and OTLP send operations, as commands that humans, agentic AIs, and CI/CD workflows can drive.
Who it's for
- Humans managing Dash0 assets and querying telemetry from the terminal.
- AI coding agents (Claude Code, Cursor, Codex, Copilot, and others) that need discoverable, structured commands with predictable output.
Every command supports JSON output, structured
--help, and JSON error formatting. Agent mode makes those defaults automatic when the CLI detects an agent in the environment. - CI/CD pipelines that keep dashboards, rules, and other assets in sync with git via GitOps-style
apply, or emit deployment events at release time.
What it does
- Manage assets as code.
Create, list, get, update, and delete dashboards, views, check rules, recording rules, synthetic checks, notification channels, and spam filters — one command at a time or via
dash0 apply -f <dir>for GitOps flows. - Query telemetry. Search logs, spans, traces, metrics, and failed checks with a common filter syntax and time-range flags.
- Send telemetry. Emit logs, spans, and deployment events via OTLP directly from your terminal or from GitHub Actions.
- Manage profiles. Configure multiple Dash0 environments (development, staging, production, or several organizations) with named profiles and either OAuth or static-token authentication.
Design principles
- Ergonomic for agents by default. Structured JSON output, JSON help, JSON errors, no interactive prompts, no colored output — all triggered automatically when the CLI is invoked by a known AI coding agent.
- Consistent surface across asset types.
Every asset kind uses the same five subcommands (
list,get,create,update,delete), the same output formats (table,wide,json,yaml,csv), and the same idempotent-upsert semantics. - Idempotent by design.
applyand per-assetcreateperform create-or-replace (PUT) when the source document carries a user-defined identifier — safe to run repeatedly from CI. - No secrets on the command line. Authentication and connection settings live in profiles or environment variables; the CLI never asks for a token as a positional argument.
Next steps
- Installation — install via Homebrew, Docker, Nix, or from source.
- Quickstart — a five-minute walkthrough: log in, list assets, query telemetry, and send a deployment event.
- Command Reference — full syntax and examples for every command.
- GitHub Actions — use the CLI from your workflows.