Dash0 Release Notes

Need more room for that service name? Now you can drag column edges to resize tables in Trace Explorer, Log Explorer, Web Events Explorer, and Resources.

Running multiple sites on a single web integration? Now you can finally see which domain each session came from.

Connecting your AI coding tools to Dash0 just got a lot simpler. Dash0's MCP server now supports OAuth 2.0 authentication, so tools like Claude Code, Cursor, Windsurf, and other MCP-compatible clients can authenticate without requiring you to create and manage static API tokens.

Chart legends now behave the same way everywhere in Dash0.

When a check fails, the first thing you want to do is understand why. Until now, drilling down from a failed check into the underlying telemetry meant losing context — metric filters weren't carried over, and the charts you landed on showed largely unfiltered data, often not relevant for your investigation. Dash0 now provides a seamless drilldown experience from failed checks directly into the relevant spans, logs, web events, and metrics — with filters automatically extracted from the check's PromQL expression.

The Query Builder just got a major upgrade. We've unified it with the dashboard panel editor, giving you a cleaner, more consistent experience, and a few quality-of-life fixes along the way.

Binary operations in PromQL silently drop results when one side of the expression has no matching time series. If you divide requests by capacity, and a new instance hasn't reported capacity yet, that instance vanishes from your chart — no error, no gap, just missing data. This makes dashboards unreliable and alerts blind to exactly the moments that matter most. Dash0 now supports the fill modifier for binary operations, giving you explicit control over what happens when one side of an expression has no match.

Span events are being deprecated in OpenTelemetry. With one click, you can transform them into logs in Dash0.

Investigating a single log record rarely tells the full story. The new Related Logs tab surfaces the surrounding context — every log from the same resource and trace, organized chronologically around the record you're inspecting — so you can reconstruct what happened without leaving the sidebar.

Dash0 now supports exporting audit logs to external systems using the OpenTelemetry Protocol (OTLP). Stream organization-level audit events, including user actions, configuration changes, and access activity, to any OTLP-compatible backend, SIEM, or long-term storage system. Built on OpenTelemetry standards, this capability integrates natively with your existing observability and compliance tooling, giving security and platform teams full control over where audit data lives and how long it is retained.

AI coding agents are rewriting how teams build and operate software. But most CLIs were designed for humans — with colored tables, interactive prompts, and prose help text that agents struggle to parse reliably. Agent mode makes the Dash0 CLI a first-class tool for AI coding agents, with zero configuration required.

The Dash0 Terraform Provider is now also available on the OpenTofu registry.

Visualize the full topology of your distributed traces, dependencies, cross-boundary interactions, and service relationships, for faster root cause analysis.

Change the operator on an existing filter and keep the values you already picked. No more re-selecting the same things twice.

AI coding agents are transforming how teams write code, but they have a blind spot: observability. When an agent scaffolds a new service or adds a feature, it rarely sets up instrumentation — and when it does, the result is often noisy, non-standard telemetry that obscures more than it reveals. Dash0 Agent Skills close that gap.

Monitor the complete Kubernetes cluster with one command

Auth tokens already support restricting ingestion to specific datasets. Now, tokens can also be restricted to individual signal types — logs, spans, metrics, profiles, or web events — giving you fine-grained control over what data each token is allowed to send.

When investigating issues in production, the first question is almost always "what's different about the errors?" Until now, the Triage panel in the Trace Explorer, Logs Explorer, and Web Events Explorer defaulted to comparing against the current timeframe — a useful but generic baseline that required an extra click to switch to error comparison. Starting today, all three explorers default to error comparison mode. The triage panel immediately highlights which attributes correlate with errors, so you can jump straight into root-cause analysis without changing any settings.

Stack traces from minified JavaScript are now automatically translated into their original source locations.

Every metric, attribute, and event that Dash0 adds to your telemetry is now defined in a public, machine-readable registry. The Dash0 Semantic Conventions give you a single place to look up what each field means, where it comes from, and how to query it.