Galileo Instrumentation for LangChain Apps
Wrap-up
You took the Monitoring Agentic AI Applications
multi-agent travel planner and instrumented it with Galileo (Splunk Agent Observability) by adding just two things: galileo_context.init(...) and a single GalileoCallback on the LangGraph run config.
With that, every agent node’s LLM call now appears as a nested span in a single Galileo trace per request — no per-node changes required and very low maintenance.
You now have the same workload traced in two observability tools (Splunk Observability Cloud from Monitoring Agentic AI Applications , and Galileo here), which is a useful basis for comparison.
Next, we’re going to expand on this by:
- Adding Galileo metrics (for example,
Context Adherence) to the captured traces. - Working through the features that help Galileo support better observability of agents.
- Leveraging powerful features like Signals.
- Using a dedicated
GalileoLogger(project=..., log_stream=...)to route specific runs to different log streams. - Adding more complexity to our agents.