OpenAI Agents SDK tracing
Trace OpenAI Agents SDK runs natively. A drop-in trace processor captures every agent span, tool call, and handoff — Python and JavaScript.
Trace every OpenAI Agents SDK run
Full span traces
See the complete execution tree of every OpenAI Agents SDK run — nested spans, tool calls, and LLM requests on one timeline.
Cost & latency analytics
Track token usage, cost, and latency for every OpenAI Agents SDK call, broken down by model, prompt, or metadata.
Rich metadata & search
Tag, score, and search every request. Filter production traffic by content, model, status, or custom key-value pairs.
OpenTelemetry-native
OpenAI Agents SDK streams traces straight to PromptLayer's OTLP endpoint — no proxy in your request path, no SDK rewrite.
Debug in the playground
Open any OpenAI Agents SDK trace in the Playground to reproduce, tweak, and fix the exact prompt that failed.
Turn traces into evals
Promote real OpenAI Agents SDK runs into versioned datasets and run evaluation pipelines to catch regressions.
Understand what your OpenAI Agents SDK app is doing
The OpenAI Agents SDK orchestrates multi-agent workflows. PromptLayer's native processor records every span, so you can see how agents reason and hand off work.
See the full picture
Every OpenAI Agents SDK run becomes a searchable, replayable trace — inputs, outputs, models, and timing.
Find the bottleneck
Pinpoint the slow span or expensive model call dragging down your OpenAI Agents SDK pipeline.
Catch failures fast
Surface errors, failed tool calls, and low-quality outputs before your users do.
Ship with confidence
Connect traces to evaluation pipelines so every change to your OpenAI Agents SDK app is tested.
Frequently asked questions
If you still have questions feel free to contact us at sales@promptlayer.com