Claude agents now hand over their work trail: model calls, tool use and MCP calls land in one OpenTelemetry trace

Agents built on the Claude Platform can now be traced step by step like ordinary services. Claude Code or Cowork traces could already go to an observability tool, while agents built on the Claude Platform did not show up in the same place. After adding a destination in the Claude Console, every agent turn, model call and tool use arrives in the observability tool as a standard OpenTelemetry trace with latency, tokens and IDs.

The traces follow the OpenTelemetry GenAI semantic conventions, so data from Claude and other providers works in the same tools and dashboards. In Grafana Cloud, Honeycomb, Sentry and Splunk, agent traces sit beside logs and metrics, so a slow or failed step can be traced to the service it came from. With Arize AI, Braintrust, Datadog, Langfuse, LangSmith by LangChain, New Relic and PostHog, real traces can become evals, catch quality issues and measure whether a change helped; traces sent to Databricks Unity Gateway let platform and security teams see token usage by team and flag unusual activity.

Claude traces also work with any other OpenTelemetry tool that accepts a standard OTLP endpoint. Managed Agents and the Messages API need a destination in the Claude Console, while Claude SDKs export traces automatically. OpenTelemetry traces on the Claude Platform are available in public beta.

The feature is aimed at teams building agents on the Claude Platform, sending model calls, tool use and MCP server calls into existing observability tools. Managed Agents and the Messages API need a destination in the Claude Console, while Claude SDKs export traces automatically. OpenTelemetry traces on the Claude Platform are available in public beta.

The feature is in public beta. Managed Agents and the Messages API require a destination in the Claude Console before traces are exported, while Claude SDKs export traces automatically.