Integrations / OpenAI Agents SDK
Keep the SDK in charge. See the whole run.
Sulcus instruments the SDK’s Runner calls and lifecycle hooks. The SDK still owns the agent loop, tools, handoffs, sessions and guardrails.
Available now · SDK 0.22.xSulcus adds
A runtime view across agents.
- Agent lifecycle, including handoffs between agents.
- Model calls and local tool activity.
- Guardrail events when a guardrail triggers.
- Reported token usage and an optional per-run token limit.
- Stop control and a retained history of every run.
The SDK keeps
How agents behave.
- The agent loop, model selection, tools and handoffs.
- Sessions, approval state, SDK limits and results.
- Hosted tools and provider behavior outside local hooks.
- Anything your application stores, and its side effects.
How it attaches
Around your normal Runner calls.
When a project selects the OpenAI Agents SDK, Sulcus activates its instrumentation in the run’s container and wraps Runner.runand Runner.run_sync. Your application calls the SDK as usual. Sulcus currently supports SDK version 0.22.x.
OBSERVATION BOUNDARY- Runs the agents
- OpenAI Agents SDK
- Instrumented calls
- Runner.run / run_sync
- Event source
- Lifecycle hooks and tracing spans
- Not captured
- Prompts, model outputs, tool payloads
Boundaries
What this integration doesn’t do.
If your agents request tool approval, the request can appear in Run Details. Sulcus Cloud can’t approve, deny or resume it.
Provider-hosted tools, internal retries and nested runs outside the instrumented Runner call don’t get full lifecycle coverage. The token limit applies per run, using reported usage.
Get started
Run your first agent in Sulcus.
Point Sulcus at a public Git repository, pick the entrypoint and framework, and watch the run live.
Python · public HTTPS Git repos · LangGraph, CrewAI, OpenAI Agents SDK
