Skip to content

A programming framework for agentic AI

Migrates GAIA to try_get_underlying_agent_instance (#250)

afourneyb33c9602b219
diff --git a/python/benchmarks/GAIA/Templates/TeamOne/scenario.py b/python/benchmarks/GAIA/Templates/TeamOne/scenario.py
index dee4dfd53..20a0ec7f1 100644
--- a/python/benchmarks/GAIA/Templates/TeamOne/scenario.py
+++ b/python/benchmarks/GAIA/Templates/TeamOne/scenario.py
@@ -148,8 +148,7 @@ async def main() -> None:
run_context = runtime.start()
- actual_surfer = runtime._get_agent(web_surfer.id) # type: ignore
- assert isinstance(actual_surfer, MultimodalWebSurfer)
+ actual_surfer = await runtime.try_get_underlying_agent_instance(web_surfer.id, type=MultimodalWebSurfer)
await actual_surfer.init(model_client=client, downloads_folder=os.getcwd(), browser_channel="chromium")
#await runtime.send_message(RequestReplyMessage(), user_proxy.id)
@@ -186,8 +185,7 @@ async def main() -> None:
await run_context.stop_when_idle()
# Output the final answer
- actual_orchestrator = runtime._get_agent(orchestrator.id) # type: ignore
- assert isinstance(actual_orchestrator, LedgerOrchestrator)
+ actual_orchestrator = await runtime.try_get_underlying_agent_instance(orchestrator.id, type=LedgerOrchestrator)
transcript: List[LLMMessage] = actual_orchestrator._chat_history # type: ignore
print(await response_preparer(task=task, source=(await orchestrator.metadata)["name"], client=client, transcript=transcript))