Introducing Claude Platform Tracing in New Relic AI Monitoring

New Relic has introduced public-beta support for natively receiving OpenTelemetry traces from Claude Managed Agents, the Messages API, and Claude SDKs, allowing Claude model and agent activity to appear alongside application performance data in a single distributed trace. By configuring New Relic as an OpenTelemetry destination in the Claude Console and exporting application traces to New Relic, teams can connect model requests, agent turns, tool calls, evaluations, and custom service activity without duplicating span data or maintaining custom attribute mappings. New Relic AI Observability can display trace waterfalls, agent entities, conversation replays, alerts, and dashboards, helping users identify whether delays or failures originate in models, tools, APIs, or databases. Its AI Evaluation feature, currently in public preview, can screen captured prompts, completions, and tool activity for prompt injection, PII leakage, and toxicity when full-content capture is enabled, while a metadata-only option limits collection to timing and identifiers. Initial support covers first-party Claude Platform services, with Amazon Bedrock and Google Vertex AI support planned, and organizations can configure regional data ingestion for selected workspaces through the Claude Console.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.