Export one conversation's execution trace (OTLP / Langfuse)
Render one conversation’s execution trace into the wire format the caller’s own observability stack ingests: ?format=otlp (default) returns an OTLP/JSON ExportTraceServiceRequest ({ resourceSpans: [...] }) ready to POST straight at an OpenTelemetry collector’s /v1/traces (Datadog, Grafana, any OTLP intake); ?format=langfuse returns a Langfuse /api/public/ingestion batch of trace/generation/span events. One root conversation span, one span per LLM turn (GenAI semantic conventions), one span per tool. Span/trace ids are deterministic, so re-pulling stays idempotent. Elevated roles only.
Authorizations
Dashboard JWT token from Clerk
Headers
Sandbox opt-in for Clerk-session-authenticated requests. Set to true to route the call through the test-mode pipeline: no real provider delivery, no credits deducted, response meta.test_mode: true. Ignored for live API keys (dv_live_sk_*) — server-to-server clients must use a test-prefixed key (dv_test_sk_*) to exercise sandbox. Test-prefixed keys unconditionally enable sandbox regardless of this header.
true, false Path Parameters
The agent id.
1The agent conversation id.
1 - 200Response
The OTLP ExportTraceServiceRequest (resourceSpans) or Langfuse ingestion batch body, returned verbatim — not wrapped in the standard envelope — so it can be forwarded to the caller's collector as-is.
The OTLP ExportTraceServiceRequest (resourceSpans) or Langfuse ingestion batch body, returned verbatim — not wrapped in the standard envelope — so it can be forwarded to the caller's collector as-is.