---
type: "post"
id: "claude-platform-can-now-emit-opentelemetry-traces-from-5c6a70"
slug: "claude-platform-can-now-emit-opentelemetry-traces-from-5c6a70"
title: "Claude Platform can now emit OpenTelemetry traces from Messages API, Managed Agents, and SDKs (public beta)."
description: "Thesis: Claude Platform is making operational visibility part of the managed API surface by exporting request and agent-session traces to customer-selected observability destinations."
retrieval_nugget: "Thesis: Claude Platform is making operational visibility part of the managed API surface by exporting request and agent-session traces to customer-selected observability destinations."
published_at: "2026-10-14"
updated_at: "2026-10-14"
record_date: "2026-10-14"
date_kind: "published_at"
topics: ["claude"]
source_urls: ["https://platform.claude.com/docs/en/manage-claude/opentelemetry/api"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:claude-platform-can-now-emit-opentelemetry-traces-from-5c6a70"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:b43df0baf7007767589bb3f66de2eb3483214e112e858cf4ec51f059a6a5b803","reviewed_at":"2026-10-14T19:06:51.784Z","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":1,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Claude Platform is making operational visibility part of the managed API surface by exporting request and agent-session traces to customer-selected observability destinations.","observed_facts":[{"text":"The documentation says the API emits a span for each Messages API request and each tool use, and also emits spans for Managed Agents sessions.","source_urls":["https://platform.claude.com/docs/en/manage-claude/opentelemetry/api"]},{"text":"It says teams configure a destination in Claude Console and can receive spans without changing application code.","source_urls":["https://platform.claude.com/docs/en/manage-claude/opentelemetry/api"]},{"text":"Destinations can cover all, selected, or all-except workspaces, and support metadata-only or full-content modes.","source_urls":["https://platform.claude.com/docs/en/manage-claude/opentelemetry/api"]}],"mechanism":"Administrators provide an OTLP/HTTP endpoint and authentication headers through the Console or Admin API; workspace selection determines which activity is sent, while content mode determines how much content each span carries. This moves telemetry configuration to the platform control plane rather than requiring application-side instrumentation for the documented flows.","why_now":"As teams move from individual model calls toward managed agent sessions, they need a way to inspect execution in the observability systems they already operate. Claude Platform is extending its runtime-facing surface with a standard trace export path.","implications":["Teams using Claude API or Managed Agents can evaluate trace delivery to their existing observability stack without first rewriting application code.","The ability to scope workspaces and content mode gives administrators controls over trace routing and payload detail."],"evidence_boundary":"The supplied documentation establishes a public-beta feature and configuration options, but does not establish default-on behavior, adoption, reliability at scale, or the precise sensitivity and retention properties of every downstream observability provider.","watch_conditions":["Watch whether Anthropic moves the tracing capability out of public beta and clarifies operational guarantees, content handling, and delivery behavior for production workloads."],"related_records":[],"new_branch_reason":"No prior New Runtime record was supplied. This creates a separate analytical record for the new trace-export capability without asserting a relationship to an earlier analysis."}
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/claude-platform-can-now-emit-opentelemetry-traces-from-5c6a70/","markdown":"https://newruntime.com/posts/claude-platform-can-now-emit-opentelemetry-traces-from-5c6a70.md","json":"https://newruntime.com/posts/claude-platform-can-now-emit-opentelemetry-traces-from-5c6a70.json"}
related_patterns: ["runtime-observability-becomes-the-primary-engineering-focus-for-ai"]
---

# Claude Platform can now emit OpenTelemetry traces from Messages API, Managed Agents, and SDKs (public beta).

## Retrieval answer

Thesis: Claude Platform is making operational visibility part of the managed API surface by exporting request and agent-session traces to customer-selected observability destinations.

Thesis: Claude Platform is making operational visibility part of the managed API surface by exporting request and agent-session traces to customer-selected observability destinations.

Observed facts: The documentation says the API emits a span for each Messages API request and each tool use, and also emits spans for Managed Agents sessions. It says teams configure a destination in Claude Console and can receive spans without changing application code. Destinations can cover all, selected, or all-except workspaces, and support metadata-only or full-content modes.

Mechanism: Administrators provide an OTLP/HTTP endpoint and authentication headers through the Console or Admin API; workspace selection determines which activity is sent, while content mode determines how much content each span carries. This moves telemetry configuration to the platform control plane rather than requiring application-side instrumentation for the documented flows.

Why now: As teams move from individual model calls toward managed agent sessions, they need a way to inspect execution in the observability systems they already operate. Claude Platform is extending its runtime-facing surface with a standard trace export path.

First implication: Teams using Claude API or Managed Agents can evaluate trace delivery to their existing observability stack without first rewriting application code.

Evidence boundary: The supplied documentation establishes a public-beta feature and configuration options, but does not establish default-on behavior, adoption, reliability at scale, or the precise sensitivity and retention properties of every downstream observability provider.

First watch condition: Watch whether Anthropic moves the tracing capability out of public beta and clarifies operational guarantees, content handling, and delivery behavior for production workloads.

This begins a distinct analytical record because no prior New Runtime record was supplied; it captures the new trace-export capability rather than assuming an earlier analysis or inventing a link to one.

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.
- 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)