---
type: "post"
id: "trace-agents-with-opentelemetry-on-the-claude-platform-a01c87"
slug: "trace-agents-with-opentelemetry-on-the-claude-platform-a01c87"
title: "Trace agents with OpenTelemetry on the Claude Platform"
description: "**Thesis:** Native OpenTelemetry export makes agent execution more inspectable by placing Claude traces in the observability systems teams already use."
retrieval_nugget: "**Thesis:** Native OpenTelemetry export makes agent execution more inspectable by placing Claude traces in the observability systems teams already use."
published_at: "2026-10-14"
updated_at: "2026-10-14"
record_date: "2026-10-14"
date_kind: "published_at"
topics: ["claude","anthropic"]
source_urls: ["https://claude.com/resources/articles/agent-observability-on-the-claude-platform"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:trace-agents-with-opentelemetry-on-the-claude-platform-a01c87"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:903655bbad0e10d61d8bb8b7ef9c13185928152bb91306e605e178be4da1b948","reviewed_at":"2026-10-14T18:04:09.166Z","source_evidence_count":1,"verified_claim_count":3,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":3,"implication_count":2,"watch_condition_count":2,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Native OpenTelemetry export makes agent execution more inspectable by placing Claude traces in the observability systems teams already use.","observed_facts":[{"text":"Anthropic says public beta tracing covers Claude Managed Agents, the Messages API, and Claude SDKs.","source_urls":["https://claude.com/resources/articles/agent-observability-on-the-claude-platform"]},{"text":"The article says traces include agent turns, model calls, tool use, latency, tokens, and IDs.","source_urls":["https://claude.com/resources/articles/agent-observability-on-the-claude-platform"]},{"text":"Managed Agents and the Messages API require a destination in Claude Console; SDKs export traces automatically.","source_urls":["https://claude.com/resources/articles/agent-observability-on-the-claude-platform"]}],"mechanism":"Standardized traces expose sequences of agent actions and can be correlated with application logs and metrics in existing observability systems, lowering instrumentation and debugging friction.","why_now":"As teams give agents access to more tools and workflows, they need operational evidence to debug failures and decide what to improve.","implications":["Teams with OTel-compatible systems can inspect Claude agent activity without building a separate telemetry pipeline.","Tracing provides visibility but does not itself constrain permissions or guarantee safe behavior."],"evidence_boundary":"The announcement documents availability, trace fields, setup requirements, and vendor-reported customer examples; it provides no independent comparative evaluation of reliability, safety, or cost outcomes.","watch_conditions":["Watch whether teams connect traces to verified quality or reliability improvements rather than merely increasing telemetry volume.","Watch for documented limits around trace completeness, sensitive data handling, and cross-provider interoperability."],"related_records":[],"new_branch_reason":"No prior New Runtime record was supplied; this records the distinct product change of native OpenTelemetry export for Claude agent surfaces."}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/trace-agents-with-opentelemetry-on-the-claude-platform-a01c87/","markdown":"https://newruntime.com/posts/trace-agents-with-opentelemetry-on-the-claude-platform-a01c87.md","json":"https://newruntime.com/posts/trace-agents-with-opentelemetry-on-the-claude-platform-a01c87.json"}
related_patterns: ["runtime-observability-becomes-the-primary-engineering-focus-for-ai"]
---

# Trace agents with OpenTelemetry on the Claude Platform

## Retrieval answer

**Thesis:** Native OpenTelemetry export makes agent execution more inspectable by placing Claude traces in the observability systems teams already use.

**Thesis:** Native OpenTelemetry export makes agent execution more inspectable by placing Claude traces in the observability systems teams already use.

Anthropic announced that OpenTelemetry traces are available in public beta for Claude Managed Agents, the Messages API, and Claude SDKs. **Observed fact:** The article says traces include each agent turn, model call, and tool use, with latency, token counts, and IDs. **Observed fact:** Managed Agents and the Messages API require a destination configured in Claude Console, while SDKs export traces automatically. Anthropic says the traces follow OpenTelemetry GenAI semantic conventions and can be sent to compatible OTel tools.

**Mechanism:** Standardized traces expose sequences of agent actions and can be correlated with application logs and metrics in existing observability systems, lowering instrumentation and debugging friction. **Why now:** As teams give agents access to more tools and workflows, they need operational evidence to debug failures and decide what to improve. This extends Anthropic’s Managed Agents runtime story with a built-in observability path; it does not establish that tracing itself constrains agent permissions.

**First implication:** Teams with OTel-compatible systems can inspect Claude agent activity without building a separate telemetry pipeline. Anthropic cites customer uses for evaluating interactions, detecting recurring failures, and tracking usage, but those examples are vendor-reported. **Evidence boundary:** The announcement documents availability, trace fields, setup requirements, and vendor-reported customer examples; it provides no independent comparative evaluation of reliability, safety, or cost outcomes.

**First watch condition:** Watch whether teams connect traces to verified quality or reliability improvements rather than merely increasing telemetry volume. Also watch for documented limits around trace completeness, sensitive data handling, and cross-provider interoperability.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Claude](https://newruntime.com/topics/claude.md): Explore the claude topic hub.
- topic hub: [Anthropic](https://newruntime.com/topics/anthropic.md): Explore the anthropic topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)